 |
| An insulin pump. Image from Wikimedia Commons |
Today's article, published in Medpage Today, is from a few days ago; it is about a paper just published in Diabetologia that shows a possible link between Enteroviruses and type 1 diabetes. Enterovirus has been in the news lately because the US has been suffering an outbreak of EV-D68 since August, putting many children in the hospital. Seven children have died. EV-D68 a nasty virus, a distant cousin of Polio, that causes a respiratory infection with coughing, wheezing, and fever in children. In rare cases the infection causes paralysis, much like Polio. EV-D68, Polio, and Coxsackie A virus, which causes Hand, Foot, and Mouth disease, are all specific types of Enteroviruses. Does this new research show that in addition to respiratory disease and possible paralysis that this viruses cause diabetes?
First let's all get on the same page about what type 1 diabetes is, and what it isn't--mainly type 2 diabetes. (You can skip the next two paragraphs if you're sure you already know) Type 1 diabetes (T1D) is an auto-immune disease (other auto-immune diseases include Celiac's, Rheumatoid arthritis, psoriasis, and many others). In general auto-immune diseases are caused by miss-regulation of a person's immune system such that, in addition to fighting infections (bacterial or viral), it also fights that person's healthy tissue; allergies are similar in that they are also immune miss-regulation, but the immune system attacks harmless things in the environment, like cat dander or peanuts. Specific to T1D, the immune system attacks the beta cells of the pancreas, eventually killing them all. These beta cells have a very important role, they monitor the amount of glucose (a type of sugar) in the blood and release insulin, C-peptide, and amylin as needed in response to changing blood glucose levels. Insulin is the blood sugar control hormone that most people are familiar with; it stimulates the uptake of glucose from the blood into the muscle and fat cells where it can be used for energy now or converted into fat and saved for later (the brain and nerves uses about 20% of all daily calories and takes up glucose in a non-insulin dependent manner). Without insulin the sugar is not available to the muscles. Stored fat must be used for energy, but without insulin, calories from sugar can not be stored as fat. This is why undiagnosed/unmanaged T1D causes rapid weight loss. The other two hormones have supportive roles to insulin, C-peptide protects nerve cells from damage due to high glucose levels and amylin slows digestion to prevent blood sugar spikes. People with T1D live relatively normal lives, but they must use blood sugar monitors and synthetic insulin to do their pancreas's job and manage the glucose in their blood. Too little insulin starves the muscles, too much starves the brain (as all the sugar is sent to the muscles and fat, leaving none for the brain and nerves). T1D is usually diagnosed in children, but the beta cell attack can begin at any age. The exact cause of T1D is not currently known. What is known is that diagnoses seem to occur in clusters or bursts; suddenly doctors will see several children from the same area of about the same age diagnosed within a short time frame, this has led to speculation that T1D could be caused by some kind of communicable disease. It is also known that children with type 1 diabetic relatives are more likely to develop T1D, but studies following identical twins show that genetics is at most about 50% of the explanation. This brings us back to a communicable disease as, potentially, the cause. This is what the Diabetologia paper investigates. But before we dig into that, I want to briefly explain type 2 diabetes to help distinguish it from T1D.
Type 2 diabetes (T2D) is a completely different kettle of fish! It really has a terrible name that was given to the disease when the pathology was not well understood. Doctors first saw T2D and noticed that these patients had high blood sugar--just like a T1D patient--but normal amounts of C-peptide, this led doctors to declare they had found a "second type of diabetes" and thus type 1 and type 2 were born. It was also noted to be more common in adults while T1D is more commonly diagnosed in children, this led to T2D sometimes being called adult-onset diabetes; this name is even worse as both types can affect people of any age. T2D is different from T1D in major ways: T2D patients are generally fat or gaining weight at diagnosis, not loosing weight; and they have healthy pancreases that produce normal amounts of insulin, C-peptide, and amylin. This second reason is why I hate the name T2D, it's a completely different problem! For T2D, I prefer the name "insulin resistance" because that tells you exactly what is going on. While T1D patients have no natural insulin to tell the muscles to take the sugar, T2D patients have insulin but the muscles ignore it. This leads the sugar to go to the fat cells (causing fatness) or stay in the blood (causing damage to the nerves). Because the muscles are ignoring the insulin they are starved for energy, this leads to hunger which fuels the cycle of weight gain and high blood sugar (as the muscles continue to ignore the food they just asked for). Insulin resistance (T2D) is treated very differently from T1D, patients are encouraged to eat less sugar (to stop fueling the cycle of high blood sugar and insulin that is just being ignored), exercise more (to encourage the muscles to stop ignoring the insulin), and take drugs like metaformin (which chemically encourages muscle and fat cells to listen to the insulin and take up more sugar). Insulin resistance is thought to be caused by genetic susceptibility in addition to a life time of too much dietary sugar/carbs and too little exercise. The current research into T2D and fatness is very interesting, and something I will go in depth into in another post. For now, it suffices to say, T2D/insulin resistance is NOT an auto immune disease and is completely different from T1D, which is NOT the result of any lifestyle choice the parent made for the child.
So what did the authors of this paper in Diabetologia find out about the cause of T1D?
Based on previous work that showed statistically significant clustering of T1D diagnoses in children, and a possible connection to infection with the Coxsackie A virus, researchers from China Medical University decided to look for a link between diagnosis with an Enterovirus and later diagnosis with T1D. To look for this link they pulled the medical records of over 500,000 children from Taiwan who had been diagnosed by a doctor or hospital with an Enterovirus (including EV, Echovirus, and Coxsackie virus, but not Rhinoviruses) between 2000 and 2007 (let's call this the EV+ group). They then searched the medical records for 500,000 more children who were similar to the EV+ children with respect to age and sex, but had never been diagnosed with an Enterovirus, and added them to the study as a control group. The researchers then monitored to medical records of these children from 2000-2008 to see how many from each group would later be diagnosed with T1D.
After statistical analysis of the data the researchers found that 141 of the 570,133 control children developed T1D, that's an incidence rate of 3.89 per 100,000 kids per year or 0.0039% chance that any one kid was diagnosed in any one year of the study (any one kid would have to live to be over 200 to have a cumulative 1% chance of getting T1D). Researchers also found that 208 of the 570,133 EV+ children developed T1D, that's an incidence rate of 5.73 per 100,000 kids per year or 0.0057% chance that any one EV+ kid was diagnosed in any one year of the study (any one EV+ kid would have to live to be about 175 to have a cumulative 1% chance of getting T1D). When comparing these two rates, researchers show that there is a statistically significant difference between 0.0039% and 0.0057% with a sample size this large; they found the risk ratio (how much more at risk members of the EV+ group was to T1D) to be 1.48, or roughly half again as likely to get T1D, with a 95% confidence interval of 1.19 to 1.83. A 95% confidence interval that excludes a risk ratio of 1.00 means that the researchers can be 95% sure that the risk ratio is NOT 1.00, and thus having been diagnosed with an Enterovirus does put a child at a higher risk for being diagnosed with T1D. There are also some analyses regarding whether having certain kinds of allergies or asthma put a child at higher risk of T1D, the statistical analyses suggest that these problems are actually slightly protective.
Based on these results the researchers make their conclusions. They feel that their data is robust enough to conclude that there is a probable connection that merits further study, and that development of an Enterovirus vaccine could be protective against T1D. Their study was large and not confounded by ethnicity (98% of all Taiwanese are Han Chinese) or access to medical care (socialized medicine), and they were able to control for other differences between the children by using the extensive data in their medical files (things like family history or sex). They also did not have to worry about participants (or their parents) lying or forgetting about medical care or symptoms. The researchers do note that their study is limited in a major way by using medical records: they have to rely on a doctor visit with an Enterovirus diagnosis to determine which children are EV+. Enteroviruses are know for being asymptomatic quite often (Polio, for example, was only ever symptomatic in about 5-10% of infected people), and often having symptoms mild enough that a parent might not take a child to the doctor (the rhinoviruses are enteroviruses, they cause the common cold). Also doctors rarely prove that the causative virus is an enterovirus and just base the diagnosis on the symptoms meaning some children in the EV+ group could have had a different, unrelated infection.
What are my conclusions?
Medpages Today was pretty even handed in how they reported this research, though they are not a normal mass media outlet (they specialize in medical news for doctors). However, they did lead with the scariest version of the findings: "children younger than age 18 who had been infected with enterovirus were 48% more likely to develop type 1 diabetes". Further down the page they go on to give the specific incidence rates for the two groups which show that it is still not very likely. Overall, I'd say this was an example of good science journalism.Speaking specifically to my practical conclusions about this work, I think it is interesting but not something a Mom or Dad who just had a kid diagnosed with an enterovirus infection should give a moment's thought. If the kid already has a family history of T1D, it's already on their parent's radar. If the kid doesn't, the odds are so low, with or without an EV infection, it's not worth wasting worry over. Also, as I mentioned above, many EV infections are asymptomatic, so just because your child has never been diagnosed, doesn't mean your child hasn't been infected (though this study had no way to address if an asymptomatic infection affected T1D). All that being said, if a broad spectrum EV vaccine (we already have one for polio) were to be developed my kids would be some of the first in line. EV infections are common, and usually harmless, but they cause hand foot and mouth disease and colds--which make kids, and thus parents, miserable--and in rare cases cause serious complications like sever respiratory distress, secondary pneumonia, meningitis, myocarditis (inflammation of the heart), pancratitis, and paralysis (like polio). And death. But to help keep perspective, since August, 7 children have died of EV-D68, but last flu season 105 children died of flu.
References
Lin, Hsiao-Chuan, et al. "Enterovirus infection is associated with an increased risk of childhood type 1 diabetes in Taiwan: a nationwide population-based cohort study." Diabetologia (2014).
Ludvigsson, J., and A. O.
Afoke. "Seasonality of type 1 (insulin-dependent) diabetes mellitus:
values of C-peptide, insulin antibodies and haemoglobin A1c show
evidence of a more rapid loss of insulin secretion in epidemic
patients." Diabetologia 32.2 (1989): 84-91.
CDC statistics on '12-'13 pediatric flu deaths. http://www.cdc.gov/flu/spotlights/children-flu-deaths.htm
No comments:
Post a Comment