The “Parkinson’s glove”

In the last couple weeks lots of people have asked me about the “Parkinson’s glove.” Many had seen a report on the Today Show that made some extraordinary claims about reversal of Parkinson disease (PD) symptoms, and some read a much more muted report about the glove via the Michael J Fox Foundation. Many asked whether I had read the studies, and asked where they could get the glove. Although this was news to me also, because so many have asked, because it generated so much excitement, I have been looking into this and am responding with this short article, despite the fact that I am not a researcher. I just try to keep my eye on PD and will act as a science reporter here.

First, what are we talking about? A Stanford Medicine team led by Peter Tass, MD, PhD, is trialing a “vibrotactile” glove intended to lessen or alleviate motor symptoms caused by Parkinson’s disease (PD).  Per the 12/14/2022 Research News article by the Fox Foundation “results from initial pilot studies in roughly a half-dozen participants suggest that it may ease tremors, slowness and stiffness… Participants wore a glove on each hand for several hours per day. The gloves deliver light vibration through the fingertips. Study researchers suggest that this stimulation can ‘reset’ abnormal electrical activity in the brain, which happens in Parkinson’s disease.” I will discuss the pilot studies below.

Before that however, and as interesting a concept as this is, let’s linger for a moment on the point that pilot studies are not adequate to make an assessment of how people will generally respond to some intervention. Pilot studies are typically small and preliminary, meant to see if a device or medication is feasible, and for example, to see if the design should be improved before launching into larger studies. Pilot studies occur before or at the beginning of the typical phase I-IV trials needed for FDA approval.  Even if results are remarkable, we need to see how considerably more than a handful of patients will respond before we can make generalizations.

It is also a good idea to take into consideration that the lay press often unintentionally distorts what is being presented, in part because the reporters might not understand the science (or the scientific method in many cases).  Much of what has been presented in lay press so far around this topic has been testimonial, not science.  A lot of hyperbolic discussion has taken place in these reports: words such as “miraculous” or “transformative” that raise red flags. And, it is often problematic to present incomplete information via the media and generally not a good idea to have media coverage prior to completion of studies.

One more aside. Please don’t mistake my statements about the media reports for a lack of interest in, or a dismissal of this topic. That is not the case. We just don’t have have all of the information yet, even if the videos seem amazing. Looking at this from a different angle, it is also often true that revolutionary innovation is mocked or heckled prior to being understood. That is not my position either.  I would guess a lot of big breakthroughs first suffered this way.  And, I can only assume it is almost a rite of passage for any genius to first be called crazy. Considering those points, and given that we don’t want to be among those Einstein called “mediocre minds,” it is also a good idea to recall the words of Carl Sagan: “extraordinary claims require extraordinary evidence.”  Let’s keep all of that in our thoughts as we evaluate this with our mental toolbox.   

The Today Show December 13, 2022 presented clips of Kanwar Bhutani, a 58-year-old man who was diagnosed with PD at age 39.  Mr. Bhutani was shown prior to treatment with the glove on a day in 2018.  In that scene he was stuck in a doorway with freezing of gait.  Per correspondent Jacob Soboroff, Mr. Bhutani had been “bound to a wheelchair and taking twenty-five medications to treat his symptoms.  But, after his very first session wearing the gloves for just four hours, Kanwar saw remarkable results.”  The monologue is superimposed over a video of the Mr. Bhutani walking upright with a seemingly normal pace and near-normal stride.  The subject verified that this was “day one…I was in disbelief…the good news today is I only take two medicines.”   Mr. Soboroff then said “You’ve gone from twenty-five to two.”  

Point of clarification:  It is likely that Mr. Bhutani was taking twenty-five tablets daily, not twenty-five different medicines for PD.  And, it is more likely that he is taking “two” different medications with different dosing times, than two tablets daily, but that is only conjecture.  The information is not given in the story.  Perhaps he was able to reduce medications, but the representation is likely media distortion number one.

Next, there was a clip of Bhutani reportedly finishing the New York City Marathon “just three months after his first treatment.”  To this day he was stated to have completed multiple 5K races. Mr. Bhutani said, “It has changed my life, totally transformed me.”  It seems remarkable.

On the same Today Show segment a man was shown with “stiff steps”, and in the next clip it appeared the same man was working with gymnasts’ bars.  Likewise, another man with a shuffling gait was shown performing “happy dances.” The Today Show stated that 20 patients were involved in the “first round of clinical trials…So far, they say everyone that’s used the gloves has seen some improvement…for Dr. Tass, and the team at Stanford Medicine, they say they’ve had thousands of people already apply for their next trial. They hope to have the gloves approved by the FDA and available for purchase in about two years.” 

There is a lot to unpack there.  We don’t know for example, whether these patients were examined (and filmed) on or off medications, which would of course make a huge difference for many people with PD. In the “on” state medications are working and movement is better. In the “off” state, medications are not working and movement is worse: stiffness of muscles, slowness of gait, for example. In the case of advanced PD fluctuation of symptoms throughout the day is common, whether or not medications have been taken. If you catch someone in an off fluctuation, they will look much worse than in an “on” state.  I’m sure you know that walking is very different from gymnastics or dancing, and it would have been better to show efforts at any activities before and after use of the glove for a one-to-one comparison: here he is walking before the glove, here he is walking in the same location after the glove. There are also are numerous reports of preserved motor function in one task or another despite severe PD gait impairment. In other words, some abilities may be normal, even in a person with advanced disease.  In 2010 for example, there was a widely distributed report of a man with severe gait impairment who appeared normal while riding a bicycle.  As for the before and after of Mr. Bhutani walking through a doorway versus walking on a straightaway, that is also an uneven comparison.  Doorways can be notoriously difficult for PD patients and a prime location for freezing of gait, even when gait may look relatively normal otherwise.  Finally, PD is a clinical diagnosis and there can be mimicking conditions.  We don’t have any information about these patients on the basis of this news report to know how the diagnosis was made. Certainly, other points could be made, but I will stop here. Suffice it to say, the program was not science, and if anything, leaned a little toward sensationalized reporting.  Let’s not get swept up in that.  This is not to say there is no science reporting on this topic, or that the efforts of the Dr. Tass team are not legitimate. There will be a little more on that below.  I would also say that Stanford is known for cutting edge investigation. Still, a person’s or an institution’s standing alone is not enough. We should still apply all of our tools of critical analysis.

Where did this idea come from? It turns out vibration therapy is not new. French neurologist Jean-Martin Charcot (1825-1893) first reported that vibrations from a train or carriage ride might transiently make patients with PD more comfortable and sleep better.  Charcot lectured multiple times about vibration therapy, and even developed an automated vibratory chair (fauteuil trépidant) to simulate the rhythm of train travel. I can say anecdotally that I have many times heard from caregivers or patients that gait has been better in the exam room than they expected after a long drive to my office.  Maybe this is the same effect.  However, I would also note that data has not been consistent in that area of investigation, and attempts to evaluate vibratory chairs have not always been positive. For example Christopher Goetz’s team at Rush University found no benefit in 2012. There have been many other investigations into vibration therapy, whether focal to a specific part, or whole body therapy.

Some patients have pointed me towards a 2022 Stanford Alumni presentation by Dr. Tass  in which he explained the gloves. This was not a scientific, but more of an informational presentation.  Nonetheless, he stated that he was relying on medicine, math, physics and “self-organization systems theories” that allow one to understand how complex systems generate order. Per Dr. Tass “the standard self-organization issue problem is synchronization.” Here he refers to synchronization of brain neurons.

I want to interject here to try to simplify this and say that the basic cell of the brain is the neuron.  When one neuron “fires”, it sends a message to another neuron. That is the basis of thinking, moving, and feeling in the brain.  When two or more neurons fire together, that is synchrony.  Sometimes synchrony is good, and sometimes synchrony is bad, depending on several factors.  Many diseases are known to have some degree of abnormal synchrony.  For example, a seizure is a “hypersynchronous discharge” of multiple neurons. Instead of the back-and-forth communication of millions or billions of neurons, some begin to fire together, disrupting the normal background, and a seizure occurs. You can imagine, other dysfunction might also be possible.

I realize thinking about how brains work is not something most people do.  Here is an analogy that might help to understand the synchrony issue. There is a crowded auditorium where you’ve gone to hear someone speak.  Pretend the auditorium is the brain, and people inside (including you) are the neurons.  Before the start of the evening there is a din of talk, laughter, and other sounds (the normal background activity).  There is no obvious synchrony on the level of the floor, or if it is present, it is not overriding.  Things change when the speaker walks on stage. First a few, then many, then the whole place begin to clap.  That is synchrony.  It is fine in short bursts but imagine if it did not stop. That would be a noisy evening and you probably would not be able to hear the speaker (abnormal synchrony). You would want to break up, or de-synchronize (ultimately stopping) the clapping so you could hear the speaker.  That is idea. 

Dr. Tass notes that where synchronization in the brain is concerned, “too much of it massively impairs brain functioning” because different neurons have to process different types of information. “If everybody does the same thing that’s not good and can cause massive impairment.” Thus, he states “we were able to design stimulation techniques that allow us to move systems from this pathological synchronized state to a better state, to physiological, desynchronized states…desynchronizing stimulation enables us to make networks unlearn the abnormal connectivity and hence make them unlearn to produce abnormal synchrony. ..that’s the core of what we are doing.” 

So, why a glove?  The fingertips have a huge cortical representation.  This means the fingertips are very sensitive, with many, many nerve endings and a large area of the brain is devoted to deciphering information from them.  The glove system reportedly uses what is called “coordinated reset,” which “means that if you have a large neuron population you do not stimulate everybody at the same time. What you do is you stimulate at different sites, different times, weakly.  These are weak vibratory bursts. …in this way we disrupt synchrony and cause therapeutic effects.”  

How does that work? The Stanford group uses vibrotactile coordinated reset (vCR) fingertip stimulation in hopes that it will affect neurons in the sensory cortex and thalamus of the brain, both of which are connected to the basal ganglia. Vibratory stimulation may affect brainwave rhythms which can be detected by EEG as well.

There are multiple scientific papers published in peer-reviewed journals by Dr. Tass and his team going back over several years.  This brings us to the above-mentioned pilot studies. In 2021 the team reported on two clinical feasibility studies involving a total of eight patients in the journal Frontiers in Physiology.  The article is free and can be read by anyone.  If you want to delve into the science of this, to read a great deal more about the proposed mechanism, and observe the study design and results, take a look.  The results of these trials were also published in the journal Neural Regeneration Research in July 2022. Briefly, in group one during the acute phase of treatment “five out of six patients showed a clinically significant acute reduction of MDS-UPDRS III scores.” The MDS-UPDRS III is a measure of motor function in PD. After three months of treatment “all patients showed a clinically significant cumulative reduction of MDS-UPDRS III scores.” I would point out that five of the six had only a 4 – 6 point change in the MDS-UPDRS III scores. This is not a big difference on a scale ranging from 0 – 137. In group two “all three patients showed sustained cumulative therapeutic effect as demonstrated by a significant linear decrease of the off medication MDS-UPDRS III scores as well as off medication tremor subscores.” The changes were overall more impressive than group one. Additionally, two of the three patients were able to lower dopaminergic medication use. The data is hopeful, but again, larger studies are needed.  Even among this small group of patients the results varied a fair amount.

To summarize, the premise is that groups of neurons in the brain may begin to fire together in a way that is harmful and may cause symptoms of disease, and that is not a new concept.  That is part of what deep brain stimulation intervenes upon. However, DBS does not typically improve gait or sense of smell; whereas these reports indicate the glove does. The hope is that by vCR with a glove that causes tiny vibrations in the fingertips, information will be carried to the brain, abnormally synchronized neurons will become desynchronized, and symptoms will improve. These are extraordinary claims, and the videos are compelling, but raise so many questions. It is potentially groundbreaking, but we need more evidence. I will be watching and waiting for publication of the clinical trials.  I encourage you to do so also if this is something that interests you.  As time goes on perhaps a trail will be available near you. That is what moves science forward.

URLs checked on date of publication, 12/31/2022

The secret life of dopamine

This fall I gave a pair of talks, one in Brewer, and another in Brunswick.  What follows is a summary of that information, or at least the heart of those discussions, which essentially amounted to a review of frequently asked questions on the topic of dopamine (DA).  I think understanding DA, and its pharmaceutical friend levodopa a little better might help people with Parkinson disease (PD) understand how the disease becomes a problem, and how and why certain medication regimens are needed.  I also hope this information sheds a little more light on the issue of why you should not adjust meds on your own, a topic I covered in the spring 2016 issue of MPDN (1).  Please read that article if you are a medication “self-adjustor.” 

The issue of medication self-adjustment comes up almost daily in my office, and should not be ignored because these medications can have good and bad results, depending on how they are used. 

I should also note that this article might not be for everyone. It requires some small consideration of brain biochemistry, of study data, and numbers. I will walk you through it, but some people will not want to do that sort of mental heavy lifting, and don’t use math as a language to describe nature.   If you are among them, and don’t want to read this article, I will give you this pearl:

Levodopa is a tool, and should be used properly.  If it is not abused, it is less likely to be a problem.  However, it does affect the brain, and should only be adjusted by a doctor, and even then might not be well tolerated.  Also, like any drug, if taken improperly, it is likely to lead to problems.

To understand these issues, you should know that one of the key factors  in PD is a deficiency of DA in the brain.  DA is used for many purposes, and when it runs low people may experience increased muscle tone (rigidity) and/or unwanted movements such as tremors.  This might seem obvious today, but it was not until the 1960s that scientists confirmed DA was depleted in PD.  After learning this fact, the first logical step in attempting to help was to give people with PD DA.  Giving DA by mouth was often unsuccessful because it caused nausea.  Because of this side effect, doctors attempted to give anti-nausea drugs, and this caused another problem.  Anti-nausea drugs usually block DA receptors in the brain, including the ones that DA is meant to target (footnote #1).  Giving drugs that act against each other was, and is, generally not a good idea.   And, because of this blocking effect, to get enough DA to the brain researchers would sometimes give what seem today like very high doses of DA.  Within a couple of years, patients exposed to these high doses began to show significant advances in their parkinsonism, which led many to ask whether DA was toxic.  It also meant finding another way to replace DA in the brain.

Scientists had learned some key points that were helpful in taking next steps.   Under normal circumstances DA is produced in the brain.  The building blocks of DA come from certain foods rich in proteins.   Eating those foods allows us to break the proteins down into smaller components called amino acids.  Our bodies recycle amino acids to make other proteins and neurotransmitters like DA.  One specific amino acid we get from eating proteins is tyrosine.  In the brain, tyrosine is converted into L-dopa, and L-dopa into DA.  Giving tyrosine alone did not help, and this further supported the idea that it was a breakdown in the production of DA that caused the deficiency. 

Under normal circumstances DA formation takes place in the midbrain (a part of the brainstem).  Specifically, DA is made in the midbrain’s substantia nigra, or “dark substance.”  It had been known since the 1950s that this dark substance was missing or depleted in people with PD; thus, researchers would learn this meant DA could not be formed.

In 1967 it was found that L-dopa could also improve parkinsonian symptoms because it could be easily converted by the brain into DA. Some people use the terms L-dopa and levodopa interchangeably. Levodopa is the international non-proprietary drug (generic) name of L-dopa.  

The finding that levodopa helped PD symptoms was an important breakthrough. However, the drug still might cause nausea in some patients because when it is taken by mouth it has to leave the GI tract and enter the bloodstream, where it will pass through the liver.  The liver contains an enzyme which can break levodopa down into DA, bringing us back to the nausea issue.   To fix that problem, the specific enzyme was identified, and a drug designed to block it: carbidopa.  It followed that carbidopa/levodopa was trialed in people. 

Carbidopa/levodopa was a revolutionary drug, which enabled people to live without some of the severe parkinsonian motor symptoms.   And, patients began to describe a phenomenon known as “ON” versus “OFF.” 

When a person feels ON there is a decrease or absence of some or all motor and nonmotor signs and symptoms of PD, such as tremor, stiffness, slowness, low energy, fatigue, and soft speech.  When a person is OFF, the signs or symptoms return.   

Before levodopa, PD patients would slowly and progressively develop the signs and symptoms of disease.  Many would become bed-, or wheelchair-bound, and succumb to the effects of wasting, deconditioning, clots in immobile limbs, infections, pneumonia, and so on.  There were wards in some hospitals dedicated to the long term care of these unfortunate patients.

At first, doctors did not know how to titrate the drug.  And, in those early days of the levodopa era, very large doses were routinely given.  It seemed appropriate to treat until symptoms were completely controlled. There were also many advanced patients, treated with dopamine for the first time, who seemed to require high doses.  Just as had happened with DA treatment, it did not take long for disease to seemingly progress, for levodopa to lose efficacy, and for scientists, patients, and doctors to question whether or not the drug might be at fault.  This question has never been fully eradicated from the literature, though many attempts have been made to settle the issue.

Several retrospective analyses have compared death and disability before and after the levodopa era (2-5).  In short, when numerous sources of data were pulled together it was found that before the levodopa era, in the first five years of disease over 25% of PD patients developed severe disability or died; whereas these devastating results occurred in less than 10% after the introduction of levodopa.  The numbers over time are even more telling.  With PD present up to 10 years death or severe disability was seen in over 60% of patients in the pre-levodopa era, and about 20% after.  By 15 years, over 80% of the pre-levodopa era patients were either severely disabled or dead; and under 40% after.  The bottom line is that life spans generally normalized, or at least become significantly prolonged after levodopa was introduced as a treatment because people had fewer and less severe complications of PD.  

Still, it was reported in the levodopa era that after about 5 years of treatment with levodopa up to 50% of patients developed what are known as motor fluctuations, meaning that function might not be so predictable or that medications might wear off too soon (6).  This end of dose wearing off phenomenon began to occur in 70% or more patients after 15 years of treatment, as did unpredictable fluctuations and dyskinesias (involuntary twisting-turning movements which are not tremors).  It was not clear at this time however, whether this was all due to progression of disease or levodopa.  Confusing matters, doctors began to use the term “levodopa-induced dyskinesias.”  It is true that virtually any person with PD will have dyskinesia temporarily if given too high a dose of levodopa. 

It is not clear whether levodopa over the long term causes dyskinesia.  It seems more likely the answer is that levodopa doesn’t, though there might be some exceptions, and more than a few caveats.  The issue is complicated.

Over the course of disease with PD as less DA is produced in the brain, more levodopa is needed.  Levodopa is converted to DA and stored in a part of the brain called the basal ganglia.  Early in disease the basal ganglia is able to store the large amount of DA the brainstem is able to produce. In this case, a person will usually take relatively low and infrequent doses to make up for a small deficit.  A little goes a long way.  However, over time less DA is produced in the brain and the basal ganglia is less able to store as much because of the progression of the disease itself.  People with PD lose those cells that store DA in the basal ganglia.  It could be thought of like a car that starts out with a large gas tank.  In this car analogy, the gas tank is slowly shrinking.  A car with a smaller gas tank has to make more frequent stops at gas stations to avoid running out of gas and shutting off. By the same analogy, a PD patient with a smaller basal ganglia storage has to take more frequent doses of levodopa.  This means ON time is shorter.  

Unfortunately, as disease progression goes forward, the threshold at which dyskinesia occurs becomes lower also.  This means that a person can no longer tolerate large doses of DA without experiencing involuntary movements.  When disease advances, taking smaller and more frequent doses of levodopa is one strategy for avoiding or minimizing dyskinesias and OFF time. Even with the best of treatment though, by the time of advanced disease many patients will have dyskinesia whenever they are ON.  The key is in minimizing the severity of dyskinesia and other issues. 

Because of the advent of dyskinesias in treated patients, since at least the 1980s major thought leaders in PD have debated about whether levodopa could safely be started early or late in the course of the disease (7,8).  After all, most of the time dyskinesias only occur in treated patients in the ON state.  This might lead one to conclude that it is the medication causing the problem.  Several studies suggested that dyskinesias could be predicted by high daily levodopa dose and longer duration of levodopa treatment.  Delaying the initiation of levodopa was thought by many to be the best therapeutic strategy to prevent motor fluctuations and dyskinesias.  Thus, the term “levodopa phobia” was coined, to refer to neurologists and patients who withheld the introduction of adequate levodopa therapy as long as possible (9).    And, in the 1980s other medications such as DA agonists were developed, which were able to stave off some of the motor symptoms of disease.  However, this class of drugs is not always tolerated in doses that would be needed to reach the same efficacy as levodopa.

In the 1990s researchers set out to settle the issue as to whether levodopa alters the natural history of PD, or whether levodopa hastens the loss of brain cells (neurodegeneration) in the ELLDOPA (Earlier vs. Later LevoDOPA in PD) trial (10,11).   This was a randomized, double-blind, placebo-controlled, parallel group, multi-center trial at 33 sites in U.S. and 5 sites in Canada. The study was designed to include 360 early, mild PD patients who had not previously required levodopa treatment, but were symptomatically ready to try medication. Patients were placed in different groups of three times daily dosing of: placebo, levodopa 50mg, levodopa 100mg, levodopa 150mg, and levodopa 200mg.  The primary outcome variable measured was a change in total Unified Parkinson disease Rating Scale (UPDRS) scores from baseline to 40 weeks after randomization, and after 14 days washout from meds (to see if there was a persistent effect) (footnote #2).  Patients who took higher doses of levodopa showed the most improvement in motor scores. Some of that data Is summarized here. 

 3 x daily dose    dyskinesia  (%)    wearing off (%)        

Placebo                        (3.3)                (13.3)

50                                  (3.3)                (16.3)

100                               (2.3)                 (18.2)

200                              (16.5)                (29.7)

The table shows that the 200mg three times daily group had the highest rate of dyskinesia (at least five times as much as the other groups), and considerably more wearing off at the end of the study. It would seem that early in disease 200mg three times daily might be too high and associated with earlier motor complications. However, it is also interesting to note that even the placebo group had some degree of dyskinesia, evidence that dyskinesia is part of the disease-though it can be exacerbated by medications (footnote #3).  A key insight from the ELLDOPA trial is that it is difficult to determine when an early dose of a drug such as levodopa is too high.  It takes a great deal of investigation to detect these nuanced issues. 

Later, the CALM-PD trial followed long term use of DA agonists (12).  Post-hoc analysis of this trial showed that the onset of dyskinesias was an expected complication of disease that was independent of when levodopa was started.

Other authors have suggested that cumulative levodopa dose (how much a person has taken in their life) may be an independent predictor of dyskinesias in patients with PD (13).  This line of reasoning has led many to believe they should delay starting levodopa, or that there is a “shelf-life” for taking levodopa.  They seem to believe there is a set number of years they can take levodopa before it “stops working.”  However, evidence favors starting levodopa when it is needed, though avoiding a too much, too soon approach.  Observations and trials such as the above-mentioned ELLDOPA study have shown that high doses of levodopa early in disease can result in early appearance of dyskinesia and possible irreversible advance of disease.  One of the problems with proving it is not the drug itself, but high doses of the drug, is that it would be unethical to design a study to test this idea in people.  However, in PD model rats treated with higher levodopa doses for short periods dyskinesias developed earlier than in rats treated chronically with lower doses, although the lower daily dose rats ultimately were given higher cumulative doses (14).  At least in rats, the cumulative dose hypothesis does not seem to fit and when levodopa therapy was started did not change the risk of motor complications.

Finally, in 2014 Italian researchers set out again to determine if starting levodopa early in disease had an effect on progression of PD (15).  Investigators went to Ghana, Africa, where they found 59 PD patients who for a variety of reasons had never tried levodopa, and an additional 32 patients who took low doses of the drug.   These patients were matched by age, sex, and disease duration with groups of Italians (2282, only 50 of whom had never tried levodopa).  Patients were followed over several years.  In both groups the average time it took for patients to develop medication failure was 5-6 years.  The average time to develop dyskinesia was 6-7 years.  The only group with early medication failure or dyskinesia were those who took high daily doses of levodopa.  Duration of time one had taken levodopa was not a risk factor.  However, duration of disease itself was a risk factor, as one would expect if these complications are also simply features of the disease.

  As if to illustrate these points, one patient with advanced PD of several years had dyskinesias with the first ever dose of levodopa.   

In summary, levodopa is the strongest, most effective drug available for the motor symptoms of PD.  It is a tool, and must be used properly.  If you have made it to the end of this article and are still a medication self-adjuster, you should know that the understanding of that proper use is complex, and not something that should be taken for granted. And, you should know that there is much more to the story, not discussed here. The science of brain DA and levodopa in PD are not easy data sets to follow. I cannot stress enough that medication adjustment is something best left in the hands of the specialist. Thus, as complicated as this article might seem, it only touches the surface of the secret life of dopamine. 

Footnotes

  1. This is a bit of biochemistry and cell biology, but DA is sort of like a key, and the DA receptor is sort of like an ignition: one fits specifically into the other and has an activating effect.  It is one way brain cells communicate with each other, or get each other to do something, such as move a muscle in the body.   There are many different receptors in the brain that respond to specific neurotransmitters such as DA.    Another tricky detail is that there are subtypes of the DA receptor in different parts of the brain.   Each has a different set of functions.  
  2. It was discovered in post-hoc analysis of this trial that a full washout would take 28 days.
  3. In fact, dyskinesia can be seen with DA agonists of enzyme blockers used to treat PD.  Levodopa does not have to be involved.  And, when patients have deep brain stimulation the device itself may trigger dyskinesia. 

REFERENCES

  1. Should you adjust your own Parkinson’s meds?  Spring 2016, MPDN
  2. Poewe WH et al. Neurology. 1996;47(suppl 3):S146-S152.
  3. Hoehn MM, Yahr MD. Neurology. 1967;17:427-442.
  4. Hoehn MMM. J Neural Transm Suppl. 1983;19:253-264.
  5. Diamond SG et al. Ann Neurol. 1987;22:8-12.
  6. Lang & Lozano. N Engl J Med. 1998;339:1130-1143.
  7. Fahn S. Parkinson disease, the effect of levodopa, and the ELLDOPA trial. Arch Neurol 1999; 56: 529–35.
  8. Fahn S. A new look at levodopa based on the ELLDOPA study. J Neural Transm 2006; 70 (Suppl): 419–26.
  9. Kurlan R. “Levodopa phobia”: a new iatrogenic cause of disability in Parkinson disease. Neurology 2005; 64: 923–4.
  10. Fahn S.  Parkinson disease, the effect of levodopa, and the ELLDOPA trial. Earlier vs Later L-DOPA  Arch Neurol. 1999;56:529-535.
  11. Parkinson Study Group. Levodopa and the progression of Parkinson’s disease. N Engl J Med. 2004;351:2498-2508.  
  12. Constantinescu, et al. CALM-PD Investigatorsof the Parkinson Study Group, Impact of pramipexole on the onset of levodopa-related dyskinesias. Mov Disord 2007; 22: 1317–9.
  13. Hauser, et al. Factors associated with the development of motor fluctuations and dyskinesias in Parkinson disease. Arch Neurol 2006; 63: 1756–60
  14. Tsironis, et al. The course of dyskinesia induction by different treatment schedules of levodopa in Parkinsonian rats: is continuous DArgic stimulation necessary? Mov Disord 2008; 23: 950–7.
  15. Cilia, et al. The modern pre-levodopa era of Parkinson’s disease: insights into motor complications from sub-Saharan Africa. Brain.  2014;137(Pt 10):2731-42

Lyme and parkinsonism: is there a connection?

 

It is a more common question than you might think.  And, people are sometimes confused about what Lyme disease is.  This is common with most medical conditions.  Medicine is complicated.  Without training and a lengthy, diligent effort, one can only have an incomplete understanding, and it will be easy to confuse the facts.  Even with training we are left with many questions in science and medicine.  That is probably true of any complex issue.  This is where logic is needed and the scientific method is crucial.  Science is still the best tool we have for trying to solve these difficult questions.  When it comes to treating illness, one should learn about the disease in order “to know what you don’t know,” because the opposite situation is dangerous.

I will attempt to explain Lyme as we know it, and answer the question as to whether there is a connection with PD so that there is hopefully less confusion, and perhaps some clarity.  This will be somewhat of long article for MPDN, in part because it is complicated, and in part because I am aware that even mentioning the topic of Lyme may upset some patients who have accepted a diagnosis or seem to believe that there is some effort to suppress information about Lyme disease.  It is for some, a touchy subject.  Still, because it keeps coming up, because it can be a very serious illness, and because the season for Lyme has come along, I want to try to address the issues that have been raised in my office.  The best that I can offer is to carefully take points apart, and through this, I hope I can dispel false ideas and bring us closer to understanding this subject.


Why is Lyme a problem?

Lyme disease is on the rise in New England, and can cause serious illness, sometimes affecting the nervous system.  The majority of infections are straightforward, and treatment is usually simple.  However, diagnosis can sometimes be tricky, especially early after exposure, when the screening test is less accurate.  There are, from time to time, complicated cases (such as when one has been untreated for a long duration), but antibiotics appear to be effective, even in that situation.  As we will see below, most confusion arises around the post-treatment if patients have persistent symptoms.  Some patients incorrectly interpret ongoing symptoms as an ongoing infection, and conclude that mainstream medicine is not the way to deal with diagnosis or treatment.  That feeling is not limited to Lyme disease, as we live in a time of open anti-science sentiment and distrust of medicine, even in our elected officials.  Out of this fringe, certain jargon has arisen around Lyme disease, reflecting confusion and misunderstanding.  Examples include the terms “chronic Lyme,” “Lyme flare,” and “Lyme-literate doctor.”  I will discuss the first two terms below. “Lyme-literate,” it is worth pointing out here, actually refers to a healthcare practitioner who accepts and uses alternative, unproven concepts and treatments.  The terms listed above are not accepted by most neurologists, infectious disease doctors, or scientists who study Lyme.  The language may still be spread by the occasional healthcare provider who does not accept or follow scientifically and evidence-based guidelines, or more commonly by patients who believe they are chronically infected by the organism that causes Lyme disease.  Patient advocate groups have arisen and Lyme has become a political issue in New England.  As we will see below, laws have been passed regarding Lyme which may be counter to medical guidelines.  All of this may lead to confusion and distrust among patients who feel caught in the middle.  So, what are the facts?  How can we understand Lyme?  Can we have some clarity about this disease?

In the beginning

Lyme disease in the U.S. is named for the small coastal Connecticut town where clusters of cases were first reported in 1975.  The condition has spread over time, and in 2017 Lyme is found only rarely outside of the Northeast, mid-Atlantic, and

Adult deer tick, Ixodes scapularis, public domain, photo by Scott Bauer. (USDA ARS)

upper Midwest states.  Lyme is caused by infection with Borrelia burgdorferi, a bacterium transmitted to humans though bite of an infected deer tick (Ixodes scapularis).  Deer ticks are black-legged, and smaller than the more common dog ticks (1).

In Maine, deer ticks tend to cluster in southern and mid coastal regions, probably because of the warmer temperatures, the abundance of mice (which thrive near people and also host deer ticks), and the abundance of deer.  It is still possible to catch Lyme disease in any part of the state, though cases are much rarer in western and northern regions where temperatures are lower and mice are probably less abundant (2).

Signs and symptoms of Lyme

Some patients with Lyme disease will have an expanding, and sometimes bullseye-shaped rash, called erythema migrans (EM),

erythema migrans (EM), public domain, photo by James Gathany

which may occur in about half of patients any time between 3 and 30 days after a tick bite.  In 2003, when Lyme was not as common, or as well known, a study published in the New England Journal of Medicine (NEJM) (3) screened 10,936 people from 10 states where Lyme was endemic over a 20 month period in an attempt to identify all infections of Lyme, and to describe presenting symptoms. Ultimately, 201 confirmed cases were found. Frequent early symptoms of Lyme disease in the NEJM paper included EM, fever, headache, migratory joint pains, muscle pains, fatigue, and palpitations.  The authors noted only 2 to 3 percent of the patients had later systemic involvement such as facial palsy, trigeminal neuropathy, or Lyme arthritis.  These are issues that tend to arise only if treatment is delayed (though I would insert that I have seen facial palsy as the presenting symptom).  Over time we have learned that another, albeit uncommon late sign of untreated Lyme is carditis (inflammation of the heart) causing AV block, which affects the rhythm of the heartbeat.  Further, Lyme as a cause of meningitis (inflammation of the protective tissue covering the brain and spinal cord) is also rare in the U.S.  Lyme is a reportable illness in Maine, meaning that doctors and labs are required to report cases to the Maine Centers for Disease Control (CDC).  *see Footnote

Maine CDC investigates all reports of positive laboratory tests or clinical diagnoses of EM. Maine CDC also is compelled to produce a formal report for the State Legislature, who as shown below, has been interested in Lyme disease for several years (4).

From the report, we know that Lyme is on the rise, and in 2015 (the latest data publicly available at the time of this article) there were 1,171 “confirmed and probable” cases by federal CDC criteria (5).  Among those cases, EM was present 51%, arthritis in 30%, Bell’s palsy or other peripheral cranial neuritis (infection of nerves of the face and head) in 10%.  Hospitalization occurred in 3% of cases (38 total).  There was no mention of meningitis or other neurologic illness in the report, except to say that these conditions are rare.  There were no reports of parkinsonism.

So why do some people think PD is related to Lyme?

To be clear, most people don’t.  Onset of PD during a Lyme infection is not common, and there is no evidence that Lyme will cause PD.  Still, I am frequently asked by patients with new symptoms of parkinsonism to test for Lyme, if it could be Lyme, or some variation on that theme, and I am not sure why.  Further, there are some patients who have been convinced that Lyme disease caused the parkinsonism they began to experience during or some time after a Lyme infection (whether or not they actually had an infection).   It is often difficult to convince someone otherwise when the two conditions seemed to come together.

However, one should remember that correlation does not equal causation: two events may occur at the same time and have nothing to do with one another.

Alternatively, there may be a relationship of a sort, but it is not intuitive.  Trauma or systemic illness may unmask early PD.  Think of it like this: the brain will often compensate for neurologic disease and it may be some time before you know there is a problem as it is slowly smoldering along.  Physicians refer to this as the “prodrome.”  The progressive loss of dopamine people with PD experience usually takes a minimum of 6 to 7 years to become severe enough to cause visible signs of disease such as tremor, stiffness, or slowness.  And, many conditions such as illnesses, surgeries, and traumas may stress the brain and body of a person on the verge of showing signs of PD.  Perhaps that stress also uses up dopamine.  Whatever the case, that person may then finally start to notice the early indications of parkinsonism that would have occurred in due time, but have become evident sooner.  PD is not unique in this way.  Several neurologic diseases become apparent a little sooner than they would have in the setting of some other illness.  When this happens patients may incorrectly assume one problem caused the other.  In fact, one author noted (5):

“Virtually every known neurologic disorder has been blamed on this infection.  For most (multiple sclerosis, amyotrophic lateral sclerosis, Alzheimer disease, Parkinson disease), evidence is scant, nonexistent, or coincidental.”

I can see why people might think there is a connection, and it is a very human trait to look for patterns and associations.  We are essentially hard-wired that way.  It is good idea for survival to remember dangerous correlations.  The flaw is that they don’t always connect.  We should reason through seemingly connected problems to see if there is a false lead.

The issue is further complicated as Lyme is a complex area in medicine.  There is a great deal known about Lyme in science, but much of it is not straightforward, and requires some medical knowledge.  There is ongoing research, and like any area of science, there is room for an elevated understanding.  People outside of medicine who are trying to learn about Lyme should be careful what sources are used.  There is unfortunately an excess amount of false and misleading information on the internet about Lyme.  There are also legitimate sources, such as the CDC.

How do you test for Lyme?

The CDC recommends a two-step laboratory testing process.  When Lyme is suspected, the first step is an enzyme-immunoassay (EIA).  Physicians may also choose the less commonly used indirect immunofluorescence assay (IFA).

If the first step is negative, no additional step is recommended.  However, the test checks for antibodies against Lyme which may not have developed in early disease, and may take as long as 8 weeks to develop.  Therefore, the EIA test may be falsely negative about 32% of the time in early disease.  Repeat testing is sometimes necessary, and often patients are treated empirically if clinical suspicion is high.  Doctors should also consider alternative diagnoses, such as other tick-borne illnesses.

If the first step is positive or equivocal, this may mean that similar antibodies are detected, but not necessarily those against Lyme.  Thus, the second step is an immunoblot test called a Western blot, a confirmatory test.  Results are interpreted as positive only when both steps of the testing are positive.  The CDC does not recommend skipping the first step.

There are other commercial lab tests available which are not approved by the FDA or the CDC, yet are still used by a minority of medical practitioners in Maine.

When unproven tests are used, the diagnosis is suspect at best, leaving the actual cause of suffering undiagnosed and untreated.

How is Lyme treated?

Treatment of Lyme disease in most cases is with the oral antibiotic doxycycline for 2 to 4 weeks.  Antibiotics kill bacteria.  The bacteria that causes Lyme disease is highly susceptible to the drug and a single round of antibiotics is curative in the majority of cases.  Infrequently, IV antibiotics such as ceftriaxone are used, such as when a patient has been left untreated for some time and there is evidence of spread in the body.

What is the story with “chronic Lyme?”

About 10% of patients treated for Lyme disease with a recommended course of antibiotics will report ongoing symptoms of fatigue, pain, or joint and muscle aches (6,7). This is unlikely to be an ongoing infection.  Most of these residual symptoms resolve after six months, though an even smaller number of patients will report persistent symptoms.  Patients, and some health care providers have occasionally referred to this syndrome as “chronic Lyme disease,” which is inaccurate, and the term has been rejected by experts (8).

This is not to say there is no condition.  After bacteria are eradicated with antibiotics, if there is a residual condition as described here, it is known by the CDC as “Post-treatment Lyme Disease Syndrome” (PTLDS) (9).

Per the CDC many medical experts attribute lingering symptoms to tissue damage and probable activation of the immune system that occurred during the infection, producing what is called an autoimmune response, in which the immune system later attacks the body.  This would not be a unique phenomenon.  Examples of other infections that might cause persistent complications and autoimmune response include the bacteria Campylobacter jejuni, with resulting Guillain-Barre syndrome, or bacterial strep throat, resulting in Scarlet fever and later rheumatic heart disease.  These are not chronic infections, but the result of the immune system being essentially confused by the original bacteria into attacking some part of the body, even long after the infection is gone.  This is an area of medicine in which doctors of immunology and rheumatology are experts.  They, along with the boards of neurology and infectious disease, do not recognize “chronic Lyme disease.”

Still, there do not seem to be satisfying treatments for many of the symptoms of PTLDS, and patients may understandably feel very frustrated and suspicious. Adding to the confusion, there are some health care providers who do not follow the guidelines and explain chronic symptoms as being due to chronic infection with Borrelia burgdorferi.  I have heard from several Mainers over the last decade of how they have been given either very long, or repeated courses of antibiotics and other treatments (some of which are not FDA approved for any condition) to treat what they refer to as “Lyme flares.”  Unless active infection is detected, antibiotics are not indicated, not helpful, and may be harmful to the patient and to others (10-15).

What’s the harm in taking extra antibiotics?

One way unnecessary antibiotics are harmful is through drug resistance.  There are over 1,000 species of bacteria in the human gut, and many species in the upper respiratory system.  We have more bacteria than you may think.  There are more of them in us that there are cells in our own bodies.   If that is hard to fathom, consider that their cells tend to be much smaller than ours.

Though most bacteria we possess are benign or even helpful to us, we may harbor bacteria that can turn against us if the conditions are right.  And, if exposed to antibiotics, many species of bacteria are able develop drug resistance which may lead to infections that we cannot treat with medications that would have otherwise been curative (12).

Thus, a person may unwittingly develop drug resistant bacteria that may attack then, or lay in waiting for later.  In either case, the results can be deadly.  Drug-resistant bacteria, and there are some very scary ones, can directly affect the person taking the medication, and can also spread to others.

Antibiotics are not completely benign or risk-free.  People may have side effects which can be serious, and death due to treatment for Lyme has been reported (15, 23).   Between 2001 and 2014 the National Institute of Allergy and Infectious Diseases (NIAID), part of the National Institutes of Health (NIH), funded four placebo-controlled clinical studies evaluating whether or not giving prolonged courses of antibiotics to patients with persistent symptoms was helpful (16).   The studies were well-designed, and published in peer-reviewed journals after rigorous statistical and scientific review (17-20).  Specifically, studies evaluated:

  • whether Bburgdorferi was susceptible to the antibiotics used;
  • whether antibiotics remain at effective levels throughout the course of treatment;
  • whether antibiotics cross the blood-brain barrier to reach the brain and central nervous system;
  • whether the antibiotics kill bacteria living outside or inside cells of mammals;
  • safety and welfare of patients enrolled in the trials.

The bottom line was that prolonged antibiotic use was not recommended due to lack of efficacy and the risks of antibiotic therapy.  Side effects were common and included intravenous line infections, blood clots, allergic reactions, and gall bladder removal.

Is there a link with Lyme and PD?

There is very little in medical literature linking Lyme to parkinsonism.  In a review of the online NIH database Pubmed, which extends back decades and covers peer-reviewed medical journals in English and some foreign languages, there are very few reports of any possible association.  In 1989, 12 cases of Lyme disease with neurologic complications were reported in a review of available data (21).  Meningoradiculitis (infection of the protective covering of the brain and spinal cord) was seen in seven patients; facial palsy in two cases; infection of the end of the spinal cord, the cauda equina, was seen in one case; and inflammation of the spinal cord and its covering (meningomyelitis) was seen in two.  Severe pain was a prominent feature in most cases, and this “consistently and rapidly” improved on high-dose intravenous penicillin, while paresthesias or fatigue often lasted several months, suggestive of the later-described post-treatment syndrome.  In a patient who already had parkinsonism, the neurologic condition was not changed by the infection or the antibiotic.

A case which is sometimes cited was a 1997 report of a 78 year old man in Spain with sub-acute mental deterioration, and progressive supranuclear palsy (PSP) presentation (22).  PSP is a form of atypical parkinsonism, which is not PD, and is caused by a completely different mechanism in the brain.  While there may be some features that are similar to PD, PSP has distinct features: severe eye movement problems, very early falls, early and advanced stiffness of the spine, rather than the limbs.  In this patient, Lyme tested positive in the blood and cerebrospinal fluid.  The patient reportedly improved after treatment with ceftriaxone.  However, this single case report is in question as no organism was found in the brain, and no later follow up as to whether the patient developed parkinsonism was given.  Case reports in the medical literature are a description of an unusual situation.  Many are written to simply call attention to a possible phenomenon.  They are not proof of concept.  If no more cases are reported, the association is either extremely rare, or nonexistent.  As far as I can tell, no other compelling cases have been described in the literature.  There is simply is not adequate evidence to suggest a link between PD and Lyme.

Is Lyme a political issue?

Though Lyme disease is complicated, there is a great deal more known than I have covered here.  Obviously, keeping track of the intricate data is not simple. There are guidelines by the CDC, the American Academy of Neurology, and the Infectious Disease Society of America.  And there are regulatory agencies in our state: the Maine CDC, the Maine Board of Licensure in Medicine, and the Maine Medical Association.  Yet, the legislature of the State of Maine has weighed in contrary to guidelines on Lyme disease.

In 2013 “An Act to Inform Persons for the Options of Treating Lyme Disease” (23) recognized PTLDS, but confused the issue by stating “There are some doctors who believe that longer doses of antibiotics may sometimes be helpful.”  One familiar with science will immediately recognize the problems with this statement.  The first issue is that it promotes the idea of chronic infection, which is not supported by medical evidence.  The second problem is in the terminology.  Belief is not science.  Belief is akin to faith, acceptance of an idea in the absence of evidence, or in spite of evidence disproving the idea.  Doctors are not trained to follow belief.  Doctors and other scientists are trained to use logic and the scientific method to question, test, and interpret nature.  Through rigorous testing, studies, peer-reviewed publication, and the skills of critical analysis, doctors evaluate the available data to arrive at an understanding of disease.  Guidelines are the product of consensus among a designated panel of experts in a field after very careful consideration of the data.  Guidelines are meant to protect people and offer the best possible, and safest approach.  When the Maine Board of Licensure takes action against a doctor who does not follow guidelines, it is to protect patients.

The law further required the Maine CDC website to list “different alternatives for treatment” instead of simply stating that the evidence-based, expert consensus-driven, peer-reviewed CDC guidelines should be followed.  The federal CDC has described some alternatives to therapy (24) as a caution, and noted that none of these are proven helpful, and some are dangerous:

“Patients given a diagnosis of chronic Lyme disease have been prescribed various treatments for which there is often no evidence of effectiveness, including extended courses of antibiotics (lasting months to years), IV infusions of hydrogen peroxide, immunoglobulin therapy, hyperbaric oxygen therapy, electromagnetic frequency treatments, garlic supplements, colloidal silver, and stem cell transplants.”

The state was not finished with Lyme however.  In 2015 “An Act to Improve Treatments for Lyme Disease” (25) was passed in Maine, indicating that physicians may prescribe long-term antibiotic therapy against Lyme, thus blocking the Maine Board of Licensure in Medicine from taking action against doctors who treat patients outside of the guidelines for this condition.  In June 2015, the Portland Press Herald ran an article about the law, “Maine Legislature clears way for long-term Lyme disease treatment,” (26), in which patients who believed they had chronic Lyme disease gave positive statements in support of the new legislation.  However, Dr. Phillip Baker, executive director of the American Lyme Disease Foundation, was quoted to say the law was counterproductive, and it was “sad and most unfortunate that this bill was passed.”  He cited the above four NIH studies proving extended antibiotic therapy was not beneficial to patients and may instead be harmful.  He noted:

“To ignore such evidence to promote an unproven, unsafe therapeutic approach cannot be in the best interests of the public health and the citizens of Maine.” 

Less restraint was voiced by an online blogger for the Society for Science-Based Medicine (27), who reacted to the 2015 Maine law and noted regarding chronic Lyme:

“A whole health care cottage industry has grown up around this fictitious diagnosis, which includes, in addition to physicians, various alternative medicine practitioners, and labs which use unconventional tests to sell the patient on the diagnosis. Having dispensed with the need for objective testing and firm diagnostic criteria, a capacious range of symptoms is a ready rationale for all manner of CAM treatments, such as herbs, supplements, homeopathy, cranial sacral therapy, and such. The favored medical treatment for “chronic” Lyme is long-term antibiotic therapy, an odd choice in the normally pharmaceutically adverse pseudoscience universe.”

Finally, June 16, 2017 the CDC Morbidity and Mortality Weekly Report (MMWR)  “Serious Bacterial Infections Acquired During Treatment of Patients Given a Diagnosis of Chronic Lyme Disease — United States” described five patients, one of whom died from the treatment, and others who experienced serious infections after unnecessary treatment with antibiotics.  In one case a fatal neurologic diagnosis unrelated to Lyme was missed.


Summary

  • Lyme disease can be a serious illness.
  • Proper diagnosis and treatment should be sought (following guidelines).
  • There is no known connection between Lyme disease and PD.
  • Chronic Lyme disease is a term that is not accepted by the CDC, which has instead noted that after bacteria are eradicated with antibiotics, if there is a residual condition as described above, it is known as “Post-treatment Lyme Disease Syndrome” (PTLDS).

*Footnote:

Federal CDC defines cases for reporting as “confirmed,” “probable,” or “suspect,” (5) when a person has certain of the following:

1) EM or

2) at least one disseminated manifestation and laboratory confirmation of one of the following:

  • Positive culture for B. burgdorferi
  • IgG positive Western blot
  • Positive ELISA test and an IgM positive Western blot within 30 days of onset (and confirmed by IgG Western blot)
  • CSF antibody positive by EIA or IFA, when the titer is higher than in serum

Probable cases must meet one of the laboratory criteria mentioned above and be physician diagnosed.


REFERENCES

  1. https://extension.umaine.edu/ipm/tickid/maine-tick-species/deer-tick-or-black-legged-tick/
  2. http://www.maine.gov/dhhs/mecdc/infectious-disease/epi/vector-borne/lyme/documents/deer-tick-map-2013.pdf
  3. The Presenting Manifestations of Lyme Disease and the Outcomes of Treatment. N Engl J Med 2003; 348:2472-2474, June 12, 2003.
  4. http://www.maine.gov/dhhs/mecdc/infectious-disease/epi/vector-borne/lyme/documents/Lyme-Legislative-Report-2016.pdf
  5. https://wwwn.cdc.gov/nndss/conditions/lyme-disease/case-definition/2011/
  6. 20. Halperin J. Nervous systme Lyme disease: is there a controversy? Semin Neurol. 2011 Jul;31(3):317-24.
  7. Feder et al., A critical appraisal of “chronic Lyme disease”. N Engl J Med 2007; 357: 1422-1430
  8. Scieszka et al., Post-Lyme disease syndrome.   Reumatologia. 2015;53(1):46-8.
  9. https://www.cdc.gov/lyme/postlds/index.html
  10. De Wild, et al.,  Ceftriaxone-induced immune hemolytic anemia as a life-threatening complication of antibiotic treatment of ‘chronic Lyme disease’. Acta Clin Belg. 2016 May 12:1-5. [Epub ahead of print]
  1. Ettestad, et al, Biliary complications in the treatment of unsubstantiated Lyme disease. J Infect Dis. 1995;171:356–361.
  2. Holzbauer, et al.  Death due to community-associated Clostridium difficile in a woman receiving prolonged antibiotic therapy for suspected Lyme disease. ClinInfect Dis. 2010;51:369–370.
  1. Lantos, et al. Unorthodox alternative therapies marketed to treat Lyme disease. Clin Infect Dis. 2015 Jun 15;60(12):1776-82.
  2. Marks, et al. Antibiotic treatment for chronic Lyme disease-Say no to the DRESS. JAMA Intern Med. 2016 Dec 1;176(12):1745-1746.
  3. Patel, et al., Death from inappropriate therapy for Lyme disease. Clin Infect Dis. 2000;31(4):1107-9.
  4. https://www.cdc.gov/lyme/treatment/prolonged/index.html
  5. Berende et al., Persistent Lyme empiric antibiotic study Europe (PLEASE)–design of a randomized controlled trial of prolonged antibiotic treatment in patients with persistent symptoms attributed to Lyme borreliosis. BMC Infect Dis. 2014 Oct 16;14:543
  6. Klempner, et al.,  Two controlled trials of antibiotic treatment in patients with persistent symptoms and a history of Lyme disease. New Eng. J. Med.2001;(345):85-92
  7. Krupp, et al, Study and treatment of post Lyme disease (STOP-LD): a randomized double masked clinical trial. Neurology. 2003 Jun 24;60(12):1923-30.
  8. Fallon, et al. A randomized, placebo-controlled trial of repeated IV antibiotic therapy for Lyme encephalopathy. Neurology. 2008 Mar 25;70(13):992-1003.

21..Viader et al, Neurologic forms of Lyme disease. 12 cases. Rev Neurol.1989;145(5):362-8.

  1. Garcia-Moreno, et al.  Neuroborreliosis in a patient with progressive supranuclear paralysis. An association or the cause?. Rev Neurol. 1997 Dec;25(148):1919-
  2. CDC Morbidity and Mortality Weekly Report (MMWR)  “Serious Bacterial Infections Acquired During Treatment of Patients Given a Diagnosis of Chronic Lyme Disease — United States” online at  https://www.cdc.gov/mmwr/volumes/66/wr/mm6623a3.htm
  3. https://legislature.maine.gov/bills/getPDF.asp?paper=HP0416&item=1&snum=126
  4. http://www.mainelegislature.org/legis/bills/getPDF.asp?paper=HP0289&item=4&snum=127
  5. http://www.pressherald.com/2015/06/29/maine-legislature-clears-way-for-long-term-lyme-disease-treatment/

27. http://sfsbm.org/index.php?option=com_easyblog&view=entry&id=666

What is the Gut-Brain Axis?

We have long known that constipation is a common non-motor feature of Parkinson disease (PD) (1). One study of 7000 men over a period of 24 years showed that those with initial constipation (less than one bowel movement per day) had a three-fold risk of PD after an average of 10 years (2).  Thus, unexplained and persistent constipation in adults is associated with an increased risk of PD: 1/3 will have constipation a year prior to PD diagnosis, and 2/3 will have constipation after diagnosis (3,4).  Ultimately, greater than 80% of PD patients will have GI dysfunction of some type early in disease.  Investigators have tried to explain this for many years.  In the 1970s, one idea was the “slow transit hypothesis,” the idea that slowing down of the gut allowed the absorption of toxins which might trigger PD (5).  Though this fell out of mainstream discussion, and may have seemed an oversimplification, the idea is not so far from current thought.

To understand what we now call the gut-brain axis, it helps first to know that your gut, or your digestive tract, contains a lot of bacteria, tens of trillions of them. Largely because of the population in the gut, there are more bacteria than human cells in our bodies.   One way they get away with this is that they tend to be much smaller than our cells.  Bacteria tend to be about 0.1-5 micrometers in size (millionths of a meter).  Human cells range from about 10 to 100 micrometers, so our cells tend to be many times bigger than bacteria.  And, as long as they stay in specific locations, bacteria work with, not against us; but when they go to the wrong place, illness occurs: think diarrhea, nausea, vomiting, etc.  Under ideal circumstances, among several jobs, many gut bacteria help us to digest food and collect certain nutrients we otherwise would not be able to extract or produce.  In exchange for this, they are given food, shelter, and a chance to move along to some other environment – a great reason why we should all wash our hands when leaving a restroom.  Half of the dry weight of our stool is bacteria, and if it winds up travelling the fecal-oral route, e.g., hand to doorknob to someone else’s hand to mouth, the consequences can be bad.  A few minutes spent in a busy public restroom will alert you that hand washing is not a universal practice.

A healthy gut requires a healthy population of bacteria.  To understand this a little better, let’s define a few terms.  A single bacterial cell is a microbe.  The terms microbiome and microbiota are used a lot when discussing this topic, and for our purposes, they are the collection of bacteria in the gut.  There is a group of over a thousand different species of microbes in the gut.  Their library of unique genes is over 100 times larger than the number of genes in a person’s DNA library.  That is a considerable genetic power. The balance among types of organisms present in the gut is important as well.  When there is too much or too little of one type or another, this is imbalance, which we call dysbiosis.  Dysbiosis has been linked to inflammation, metabolic disease, certain cancers, and neurologic disease.

Dysbiosis can cause disease because one of the roles of a healthy microbiome is to regulate gut and bodily inflammation.  They do this by the release of tiny chemical messengers.  In PD, delayed gastric emptying and slow transit of stool may increase the over-production of some types of bacteria, which not only upsets the local balance, but creates the symptoms of gas, colic, and inflammation (6,7).  This makes it even easier to get their messengers into the bloodstream.  Further, as if pretending to be nerve cells, bacteria can produce neurotransmitters and neuromodulators such as GABA, serotonin, dopamine, or short-chain fatty acids (8).  This may allow bacteria to communicate with our nervous system (9).

It also goes the other way.  The drugs PD patients take may influence the microbiome (10).  One study evaluated 197 PD patients and 130 controls without PD.  They were able to identify the types and numbers of microbes taken from stool of each person.  Investigators took into account 39 potential confounders such as medications, diet, GI symptoms, and demographics.  The gut microbiome was different in PD, and multiple families of bacteria were much more robust.  In the PD patients there were also different balances corresponding to different drugs, such as COMTs and anticholinergics.  In the study, the effect of L-dopa could not be separated out as 90% of PD patients were taking the drug.

Another reason dysbiosis may be a risk factor for disease is that our gut bacteria protect us by breaking down xenobiotics (herbicides, flame retardants, insect repellents we encounter in the environment).  Living in agricultural settings is a known risk factor for PD and this may explain why.  In the lab we know that xenobiotics can cause the death of dopamine-producing cells and motor abnormalities in animal models of PD.  Therefore, dysbiosis might expose humans to toxins which would otherwise have been broken down.  The toxins may trigger disease.  The evidence that this may be happening is supported by the observation that the pathologic hallmark of PD, the Lewy body, may be seen in neurons of the gut years prior to the development of motor symptoms of PD, and may be seen in the gut of lab animals given xenobiotics.   These observations helped lead to the hypothesis that PD starts in the gut (11).

The most common protein in Lewy bodies is called alpha-synuclein (aSyn). This little protein is found at the point of communication between neurons called the synapse.  The job of aSyn is to stabilize little bags of dopamine (vesicles), and to help control synaptic plasticity (the formation of new neural pathways in the brain involved with learning).  This protein, like all others, has a very specific three-dimensional shape which it must maintain in order to work.  This is similar to a key, which must be shaped a certain way to fit into a lock and turn the tumblers.  If the key is bent, it will not work.  If aSyn is misfolded, it does not work, dopamine is not released, and new pathways are not made.  It may be that exposure to certain toxins triggers the misfolding of aSyn, or there may be other factors such as gene mutations.  Whatever the cause, misfolded aSyn may behave very much like a prion, where one bad protein warps others of the same class (think one bad apple spoils the whole bunch).

Testing this idea, researchers exposed lab mice to the pesticide rotenone, which has long been associated with risk of PD in humans (12).  The mice developed misfolded aSyn in the gut.  Abnormal aSyn in the gut activates immune cells in the brain known as microglia (13).  These microglia are then primed to destroy neurons containing abnormal aSyn.  This is one of the ways PD patients lose neurons.  If present long enough, misfolded aSyn will also travel up the long vagus nerve to the brainstem, where it will spread upwards in the same way the protein spreads in humans with PD.  Of note, mice and men with a severed vagus nerve have a lower risk of PD.

In order to test the effects of microbes and molecules on animals, researchers relied on the first gene abnormality identified in PD, the PARK1 gene (14).  The PARK1 gene causes the overproduction of aSyn.  When the gene is present in mice, they manifest a form of parkinsonism very early in life.  However, if the mice are treated with antibiotics to kill the microbiome, motor symptoms of disease become minimal, and there is reduced activation of microglia (15).   This led the investigators to raise PARK1 mice in a sterile environment with no microbiome.  These mice too, had minimal motor issues. When colonized with a microbiome (transplanted feces) from humans without PD, the mice were unchanged.  When given the microbiome from PD patients, the mice developed impaired motor function.  This supports the “two-hit” hypothesis, that PD is likely caused by genes and environment.  In other words, if someone is predisposed to develop PD because of genes, they may not do so until exposed to some trigger such as a toxin.   Researchers were also able to normalize the affected mice by adding normal chemical messages such as short chain fatty acids that a normal microbiome would produce.

Thus, the gut-brain axis raises several interesting questions:

  • Would a simple stool test reveal risk to PD and other diseases?
  • Would understanding the microbiome help physicians in selecting the appropriate drugs for people with PD, or lead to the formation of new treatments?
  • Would changing the microbiome improve disease for people with PD?

If you know, tell me.  Otherwise, stay tuned.  This is a big topic in PD.

 

Stylistic and copy editing of this article, as well as helpful insights, by Sarah Savard, RN, and Liz Stamey, RN.

 

REFERENCES

  1. https://mainepdnews.org/2016/06/12/constipation-in-pd/
  2. Abbott.  Neurology 2001; 57: 456–62
  3. Pfeiffer RF.  Gastrointestinal dysfunction in Parkinson’s disease. Parkinsonism Relat Disord 2011; 17: 10-15.
  4. Mulak A, Bonaz B.  Brain-gut-microbiota axis in Parkinson’s disease. World Journal of Gastroenterology.  2015;21(37): 10609-10620.
  5. Singharam.  Lancet 1995; 346: 861–64.
  6. Hasegawa, et al.  Intestinal dysbiosis and lowered serum lipopolysaccharide-binding protein in Parkinson’s disease.   PLoS ONE 2015;10, e0142164.
  7. Keshavarzian, et al.  Colonic bacterial composition in  Parkinson’s disease.  Mov. Disord. 2015; 30: 1351–1360.
  8. Lyte M.  Microbial endocrinology: Host-microbiota neuroendocrine interactions influencing brain and behavior.  Gut Microbes 2014; 5:381-389.
  9. Mayer , et al.  Gut/brain axis and the microbiota.  J Clin Invest 2015; 125: 926-938.
  10. Hill-Burns et al.  Parkinson’s Disease and Parkinson’s Disease Medications have Distinct Signatures of the Gut Microbiome.  Movement Disorders. 2017:00: (00)epub online Feb, 2017.
  11. Braak H.  Staging of brain pathology related to sporadic Parkinson’s disease. Neurobiology of Aging, 2003 24: 197.
  12. Pan-Montojo, et al.  Progression of Parkinson’s disease pathology is reproduced by intragastric administration of rotenone in mice.  PloS One  2010;5:e8762.
  13. Erny, et al.  Host microbiota constantly control maturation and function of microglia in theNS. Nat. Neurosci. 2015; 18: 965–977.
  14. Polymeropoulus, et al.  Mutation in the α-Synuclein Gene Identified in Families with Parkinson’s Disease.  Science 1997: 276 (5321);2045-2047.
  15. Sampson et al.  Gut Microbiota Regulate Motor  Deficits and Neuroinflammation in a Model of Parkinson’s Disease.  Cell 167, 1469–1480, December 1, 2016.

 

PD and diet

Recently, at a talk regarding PD in Brunswick, a common question came up as to whether diet influences the disease.  I gave sort of a stock answer, in that no dietary intervention has been proven to treat the condition, but that there are several points to consider about diet (see, for example, the article in MPDN about constipation).  Here, I will review some other issues.

Not surprisingly, the data seems to favor eating a healthy diet.  Many ask, “What is a healthy diet?”  That is a complex question.  Most studies seem to point to something akin to the Mediterranean diet, which contains significant olive oil, grains, vegetables, fruits, potatoes, seeds, nuts, legumes, and fish; and generally lower intake of red meats, poultry, dairy, and alcohol (though small amounts of red wine can be beneficial).  Numerous observations have been made regarding longevity and the reduction of cardiovascular or metabolic diseases among those who eat this way.

These observations have led investigators to study diet for improving health and preventing disease.  The DASH diet (Dietary Approach to Stop Hypertension) is based on the Mediterranean diet, though using relatively more low-fat dairy and less fish.  The MIND diet (Mediterranean–DASH Intervention for Neurodegenerative Delay) takes elements from both diets and increases consumption of berries, nuts, and beans. A meta-analysis, which is an in-depth review of multiple similar studies, looked at 14 prospective trials totaling thousands of participants in the U.S., Greece, Europe, and Australia (1).  In these trials, people were followed from 3.7 to 18 years and had lower rates of Alzheimer disease.  One of the studies, conducted by the World Health Organization Study Group, followed more than 130,000 health care professionals for 16 years, and showed that those who ate a Mediterranean diet had lower rates of PD (2).  Studies of older people who followed the MIND diet showed less cognitive decline at a 4.7-year follow-up (3, 4).

The PREDIMED study included 522 people aged 55-80 who were at high risk for cardiovascular disease (5). These people were randomly assigned to one of three diets: a Mediterranean diet with supplemental extra-virgin olive oil (EVOO), a Mediterranean diet with supplemental mixed nuts, or a regular diet with reduced dietary fat.  Heart attack, stroke, and death from cardiovascular causes were all reduced in those eating the diet with EVOO, and those people scored higher on the Mini-Mental State Examination (MMSE) and the clock-drawing test at 6.5 years.

In a four-month study, 124 participants with high blood pressure started either a DASH diet, or aerobic exercise and a DASH diet (6).  Those with the combined approach had better psychomotor speed (basically, a measure of the time for the connection between thought and movement).  Though none of these patients were noted to have PD, the finding is interesting because one of the problems with PD is a decrease in psychomotor speed.

In a large population study of 1,260 people with cardiovascular risk factors for dementia, the Finnish Geriatric Intervention Study to Prevent Cognitive Impairment and Disability (FINGER) (7), participants were chosen who had average, or slightly lower than average, cognitive performance for age, and were randomly assigned to a combination of diet, exercise, cognitive training, and vascular risk monitoring, or to a health advice control group.  The diet included fruit, vegetables, whole-grain cereals, low-fat milk, low-fat meat, low sugar, margarine instead of butter, and two or more portions of fish per week.  Participants underwent a group of 14 neuropsychological tests. During the 24-month follow-up period, composite scores were 25% higher in the intervention group than those who received advice alone.  Executive functioning, which includes problem solving, critical analysis, and processing speed, were better in the intervention group.

There is some data that caffeine consumption may decrease risk of PD.  Over 8,000 men were followed for 30 years in the Honolulu Heart Program (8). Incidence of PD decreased with amount of coffee intake:  the more coffee consumed, the lower the risk of PD.  Similar data was found among those that consumed caffeine from sources other than coffee, such as tea. The National Institutes of Health-AARP Diet and Health Study prospectively examined whether caffeine intake was associated with lower risk of PD among over 300,000 men and women (9).  The effect was equal, with no gender difference.  Again, the more caffeine consumed, the less likely one was to develop PD.  Multiple other studies have shown similar results. Caffeine is hypothesized to protect dopaminergic neurons by antagonizing a neuronal receptor known as adenosine A2A (10).  Animal models have shown that chemicals which inhibit A2A can protect dopamine-containing neurons, and caffeine has been shown to improve some motor function in PD (11).  The effect on A2A receptors has led to a new class of investigational drugs, such as istradefylline.  Of note, caffeine is also a CNS stimulant which may help with daytime sleepiness, alertness, and cognitive function.

Curcumin is an active ingredient of turmeric. In volumes used in cooking it is non-toxic. It is able to cross the blood-brain barrier.  Curcumin binds to mutant α-synuclein (see the article in MPDN about alpha-synuclein), and thus may prevent aggregation and formation of Lewy bodies (12, 13).

Mucuna pruriens, the velvet bean, aka cowhage, has long been used in traditional Ayurvedic medicine for Parkinsonism.  In 1937, researchers isolated levodopa from the beans (14), though this was prior to a scientific understanding of the link between levodopa and Parkinson disease.  From 1978 to 2000 there were at least three open label studies (in which patients knew they were taking the study drug instead of taking a blinded pill, which might be treatment or placebo) (15, 16, 17).  These studies reported significant improvements in Parkinsonism for up to 20 weeks.  In 2004, London researchers demonstrated with eight PD patients that single doses of immediate release 50/200 mg carbidopa/levodopa were not as fast in onset of effect as a 30 g mucuna preparation (34.6 v 68.5 min), and this was consistent with time to peak blood concentration (18).  Average ON time was 37 minutes longer with 30 g mucuna than carbidopa/levodopa, and plasma concentrations 110% higher, implying the amount of levodopa was much higher in the mucuna preparation, essentially double the dose of the carbidopa/levodopa.  Each of the eight patients were trialed with mucuna, and two complained of mild nausea, whereas one dropped out of the study due to “short lasting vomiting.”  Acute side effects of levodopa are known to include nausea and vomiting.  This is the reason carbidopa is combined with the drug in tablets.  The authors suggested that domperidone might be combined with mucuna to prevent these side effects, and that larger randomized trials should be undertaken to evaluate mucunaMucuna is unfortunately still not endorsed by such trials, and no standard measurement of levodopa derived from the bean is as yet available.

Fava beans, Vicia fava, aka the broad bean, have been known to contain levodopa since 1913 (19).  One open-label study comparing 250 g of cooked fava with 100 mg synthetic carbidopa/levodopa showed lower peak plasma concentrations after eating the beans, though the effect was similar to synthetic levodopa (20). Unlike mucuna, however, the concentration of levodopa in fava beans is very low, and therefore would require a large number of beans to reach a benefit.  Likewise, there is no standard for the amount of levodopa in fava beans, making dosing unpredictable.  In addition, some people with a genetic deficiency of glucose 6-phosphate dehydrogenase may react to eating fava beans with favism, a form of hemolytic anemia.

Finally, because levodopa, the active ingredient in carbidopa/levodopa (Sinemet), Duopa, and entacapone/carbidopa/levodopa (Stalevo) competes with amino acids for absorption in the GI tract, it should be taken one hour before, or two hours after, meals containing protein.   This creates certain problems with timing of medications and foods.  It is better to stick with a consistent time to dose meds, and plan meals around this.  Often, patients report they have stopped eating proteins, or moved all protein to the nighttime.  There are problems with this approach as well, because we need proteins, but the all at once approach may not be the right path.

I discussed this issue with Alison Fernald, RD, LD, CDE, of Mid Coast Center for Diabetes & Endocrinology, in Brunswick, Maine.  She notes that “an average person can only digest 25-30 grams of protein at a time, and a 140 pound person needs at least 50 grams of protein a day.   So, they have to break it up, and not eat it all at once.”  For those who need detailed instructions, Alison suggested a way to tackle this might be the protein redistribution diet, which will be included in this edition of MPDN.

 

REFERENCES

  1. Sofi, et al., Adherence to Mediterranean diet and health status: meta-analysis. BMJ. 2008;337:a1344.
  2. Gao, et al. Prospective study of dietary pattern and risk of Parkinson Disease. Am J Clin Nutr. 2007;86(5):1486–94.
  3. Morris, et al. MIND diet slows cognitive decline with aging. Alzheimers Dement. 2015;11:1015-1022.
  4. Morris, et al. MIND diet associated with reduced incidence of Alzheimer’s disease. Alzheimers Dement. 2015;11:1007-1014.
  5. Martinez-Lapiscina, et al. Mediterranean diet improves cognition: the PREDIMED-NAVARRA randomised trial. J Neurol Neurosurg Psychiatry. 2013;84:1318-1325.
  6. Smith, et al. Effects of the dietary approaches to stop hypertension diet, exercise, and caloric restriction on neurocognition in overweight adults with high blood pressure. Hypertension. 2010;55:1331-1338.
  7. Ngandu, et al., A 2 year multidomain intervention of diet, exercise, cognitive training, and vascular risk monitoring versus control to prevent cognitive decline in at-risk elderly people (FINGER): a randomised controlled trial. Lancet. 2015;385(9984):2255-6
  8. Ross, et al. Association of caffeine and coffee intake with risk of Parkinson disease.  Jama.2000;283:2674-79.
  9. Lui, et al. Caffeine Intake, Smoking, and Risk of Parkinson Disease in Men and Women.  Am J Epidemiol. 2012;175(11):1200–07.
  10. Schwarzschild, et al. Caffeinated clues and the promise of adenosine A(2A) antagonists in PD. Neurology. 2002;58(8):1154–1160.
  11. Cieślak, et al. Adenosine A(2A) receptors in Parkinson’s disease treatment. Purinergic Signal. 2008 Dec; 4(4):305-12.
  12. Ahsan, et al. Curcumin Pyrazole and its derivative (N-(3-Nitrophenylpyrazole) Curcumin inhibit aggregation, disrupt fibrils and modulate toxicity of Wild type and Mutant α-Synuclein. Sci. Rep. 2015;5:9862.
  13. Ahmad B., Lapidus L.J. Curcumin prevents aggregation in α-synuclein by increasing reconfiguration rate. J. Biol. Chem. 2012;287(12):9193–9199.
  14. Damodaran M, Ramaswamy R. Isolation of L-dopa from the seeds of Mucuna pruriens. Biochem J 1937;31:2149–451.
  15. Vayda, et al. Treatment of Parkinson’s disease with the cowhage plant – Mucuna pruriens (Bak). Neurol India 1978;36:171–6.
  16. HP-200 in Parkinson’s Disease Study Group. An alternative medicine treatment for Parkinson’s disease: results of a multicenter clinical trial. J Altern Complement Med1995; 1:249–55.
  17. Nagashayana, et al. Association of L-dopa with recovery following ayurveda medication in Parkinson’s disease.  J Neurol Sci2000;176:124–7.
  18. Katznschlager, et al. Mucuna pruriens in Parkinson’s disease: a double blind clinical and pharmacological study. J Neurol Neurosurg Psychiatry.2004; 75(12):1672–1677
  19. Guggenheim M. Dioxyphenylalanin, eine neue Aminosaeure aus Vicia fava.Z Physiol Chem 1912;88:276–84.
  20. Rabey, et al. Improvement of parkinsonian features correlate with high plasma levodopa values after broad bean (Vicia fava) consumption. J Neurol Neurosurg Psychiatry1992;55:725–7.

 

What about medical marijuana and PD?

I am asked almost daily about medical marijuana for PD.  For the purposes of this essay, I will call marijuana “MJ.”  I am not alone in being asked this question; my movement disorder colleagues around the state have had similar experiences.  The National Parkinson Foundation (NPF) states that, among physicians surveyed at their 40 NPF Centers of Excellence, 95% of neurologists reported PD patients had asked for a medical MJ prescription, and 80% of patients with PD had used MJ, whereas only 10% of physicians had recommended it, and 75% of physicians felt that MJ would have a negative effect on short term memory (1).

MJ in PD is a complex subject, and to be clear, under Federal law, is not legal, though a couple of drugs derived from MJ are:  dronabinol (Marinol) and nabilone (Cesamet).  Since President Nixon signed the Controlled Substances Act in 1970, MJ has been classified by the FDA as a Schedule I drug, defined as a drug with a “high potential for abuse … no currently accepted medical use.”   Schedule I designation is an impedance to scientific study for potential medical use.   In other words, it is hard to conduct trials with this designation.  However, trials or not, in Maine medical MJ has been approved since 1999.  Under the Rules Governing the Maine Medical Use of Marijuana Program (MMMP), there are certain indications for the use of medical MJ, but PD is not one of them (2).  Regardless of medical law, the recent citizen initiative to legalize recreational MJ passed.  Access is legal (within certain limits) for adults under State law.  Many people with PD tell me they will be, or have already been, trying MJ.  I have heard from people who report improved tremors and dyskinesia, some who have better sleep, some who say it does nothing, and some who do not tolerate it due to side effects.  This is all anecdotal.  What does science show?

The pharmacology of MJ is complicated.  Most MJ strains come from two species of plant: Cannabis sativa, and Cannabis indica.  Over 60 neuroactive compounds have been identified in MJ.  Many of these work on the brain’s endocannabinoid system, which is known to affect neurotransmission in the motor system of the brain, and there are many receptors in the basal ganglia – the main region dopamine is used.  Endocannabinoids also are implicated in the control of mood, cognition, and pain.   THC is found in a higher concentration in sativa plants, and is thought to be the primary psychotropic compound in MJ, the cause of paranoia, and other psychotic features.  Cannabidiol (CBD) is a non-psychoactive substance, and is in higher concentration in indica strains.  CBD has sedating, anti-emetic (helps with nausea and vomiting), and analgesic (pain) properties.

There is a lot of data about MJ in the basic science, and some in the medical literature about people taking MJ as an intervention.  The type of study is important. Case reports may describe the experience of only one, or of very few patients with an intervention.   An open label study is one in which there is no placebo, and the concern is that there may be bias, whether conscious or unconscious, on the part of the participant or the evaluator.  In other words, if you know you are taking an intervention, you may believe you are changed by that intervention.  Belief is important in the mind-body connection, and is the basis of the “placebo effect.”  This is not to say that open label trials are useless, just that they must be interpreted with caution.

In one open label observational study at an academic movement disorder center, 22 PD patients were tested at baseline and at 30 min after smoking MJ (3).  The group showed improvements of tremor, rigidity, and bradykinesia in the UPDRSIII motor score used to evaluate PD patients, improving on average from 33.1 to 23.2 after consumption.  If the score means nothing to you, know that the UPDRSIII score ranges from 0, with no PD, to 108, the worst PD imaginable.  Therefore, the lower the UPDRSIII score, the better, sort of like golf.

The scientific approach to prove the effectiveness of any drug or intervention would be to progress through trials which demonstrate safety with a few patients (phase I), a larger group (phase II), and efficacy (phase III).  In all phases, side effects are noted.   In the phase III trial, the gold standard is the well-designed, randomized, double-blinded, controlled trial (RDBCT).  In this trial participants must meet certain criteria for entry, and take either the trial intervention (for example, MJ), or a placebo.  When people participate in studies, they are evaluated periodically to measure effect.  A study is double-blinded when neither the study participant nor the evaluator know who is taking the intervention, and who is taking placebo.  All data is eventually collected, interpreted, described and discussed in peer-reviewed medical journals.   Doctors then read these articles and use their own skills of critical analysis to interpret the study.   There are many variables at play, and a study may raise many questions.  A single study may need additional support.  A study that shows unique results should ideally be repeated by a separate group of researchers and study participants.

Larger study groups tend to provide more valid information about how people may respond generally because researchers are able to average a lot of data.  Unfortunately, there is very little study data with MJ in large groups of PD patients or the RDBCT.   Some authors have tried to go through the available data and produce a conclusion from several case reports and studies.  For example, researchers looked at a collection of papers and showed that oral cannabis extract (OCE) is probably ineffective for treating levodopa-induced dyskinesias (4).

Papers such as this led a subcommittee of the American Academy of Neurology to review multiple studies involving the use of MJ in the treatment of neurologic diseases (5).  These studies showed that, in PD, treatment with the compound  9-tetrahydocannabiol (THC) or with OCE are probably ineffective in treating tremor.  OCE is probably ineffective for treating levodopa-induced dyskinesias in PD.  They noted that among 34 studies the risk of serious adverse psychopathologic effects was about 1%. The authors reported that comparative effectiveness of medical MJ vs. other therapies is unknown for these indications.

In 2015, researchers published in the journal of the Movement Disorders Society another summary of study data, which concluded that MJ is probably not helpful for the treatment of tremors or dyskinesias (6).

There are concerns about MJ in PD beyond paranoia and psychotic features.  While MJ might help with pain, insomnia, nausea, and weight loss, it might cause side effects.   MJ use decreases reaction time, has negative effects on cognitive and executive function, may lead to risky behaviors, create apathy or a lack of motivation, cause dizziness and blurred vision, cause mood or behavioral changes, affect balance, or just make a person sleepy.  All of these are already potential problems in PD, which may be exacerbated.  Chronic use of MJ has been shown to unmask underlying psychiatric disorders.  Finally, smoking MJ is associated with increased risk of lung cancer and stroke.  Vaporizing MJ is also probably not a healthy option.

In conclusion, there is not much data to support the use of MJ in PD, but there is a huge political will to legalize it, and a growing feeling among patients that it is safe to try.  This may not be true.  As with any neuroactive substance, large RDBCTs are needed to demonstrate benefit and safety.  In addition, if effective, MJ, which is a very potent compound, would ideally go through approval process with the FDA.  MJ has not, and physicians do not have label or dose recommendations, timing instructions, or adequate description of all potential side effects.

References

  1. http://www.parkinson.org/understanding-parkinsons/treatment/complementary-treatment/medical-marijuana-and-parkinsons-disease
  2. http://www.maine.gov/dhhs/mecdc/public-health-systems/mmm/documents/MMMP-Rules-144c122.pdf
  3. Lotan et al. Cannabis (medical marijuana) treatment for motor and non-motor symptoms of Parkinson disease: an open-label observational study. Clin Neuropharmacol. 2014;37(2):41-4
  4. Koppel et al.  Systematic review of medical marijuana 1948-2013.  Neurology.  2014;82(17):1556-63
  5. Systematic review: efficacy and safety of medical marijuana in selected neurologic disorders: report of the Guideline Development Subcommittee of the American Academy of Neurology. Neurology. 2014 Apr 29;82(17):1556-63.
  6. Kluger et al, The therapeutic potential of cannabinoids for movement disorders.  Mov Disord. 2015;30(3):313-327

 

B6, friend or foe?

Pyridoxine, also known as vitamin B6, is a supplement that many people take, either by itself or in a multivitamin.  It is available over the counter, no prescription required.  Most of the patients taking B6 with whom I have spoken, however, have very little idea what B6 does, and other than a general recommendation they have been given at some point, are not very clear about whether they should take it at all.

It is true that B6 deficiency can lead to illness.  It is also true that taking too much B6 can be dangerous.   Here, I am going to briefly explain the basics of B6 and take apart the common issues PD patients may face with this vitamin.

B6 is classified as a coenzyme.  Coenzymes are small molecules that assist enzymes in chemical reactions.  There are thousands upon thousands of these reactions going on all the time in your body.   B6 is involved in over 100 enzyme reactions with proteins, amino acids, carbohydrates, and lipids.  B6 helps to maintain homocysteine levels in the blood; is a participant in immune function; and is involved the development of cognition through the formation of neurotransmitters – those chemicals like dopamine that allow one nerve to communicate with another.   Deficiency of B6 is therefore associated with many different conditions such as anemia, dermatitis, cheilosis (scaling of lips, and cracks at the corners of the mouth), immune dysfunction, glossitis (swollen tongue), confusion, irritability, seizures, neuropathy, and depression.  In short, it is important to have normal B6 levels.

We get B6 by eating a healthy diet.  It is found in fruits, grains, fish, poultry, beef liver and other organ meats, potatoes, and some other starchy vegetables.  Absorbing nutrients from foods is not always easy, however, and this leads some to think that they should supplement B6 and other vitamins to “make sure” they are getting enough.  This should not be done blindly.

It is true that some vitamins and minerals are absorbed by a special process, which may take several steps in the body.  A failure of any one of the multiple biochemical steps required may cause poor absorption of the vitamin, and thus a deficiency.  This is the case with B12, for example, but it is not the case for B6.   B6 is absorbed passively in the jejunum, a part of the small bowel.  When a vitamin is absorbed passively, absorption is not an energy-requiring process, and when one has good health, there is usually no obstacle to B6 moving freely into the blood stream.  It would therefore seem easy to maintain a normal level of B6 in the body.

However, a small percentage of people will have B6 deficiency for a variety of reasons, such as kidney disease; celiac disease, Crohn’s disease, ulcerative colitis, and other malabsorption syndromes; autoimmune disease such as rheumatoid arthritis; alcohol dependence; and exposure to certain medications such as antiepileptic drugs.   There is some data, also, that PD patients who take large amounts of levodopa may have low B6 levels (1, 2).  Most of that data comes from European case reports of patients receiving continuous carbidopa/levodopa infusion via the Duodopa pump (not to be confused with Duopa in the United States), but it is also known that PD patients taking high doses of oral carbidopa/levodopa have a higher prevalence of chronic, sensory, axonal polyneuropathy (3), in other words, nerve damage.  It should be noted that some studies point to a deficiency of B12 in these patients, and apparently B6 was not always measured.   There is mounting evidence for both. It is not entirely clear how a B6 deficiency is happening in these patients.  One possible mechanism involves carbidopa, which is meant to block an enzyme in the body called dopamine decarboxylase, so that levodopa makes it to the brain.  However, carbidopa also inhibits the action of B6.  It may also be that absorption of B6 is somehow blocked by carbidopa/levodopa, or that downstream biochemical reactions deplete B6 and B12.  According to the prescribing information of carbidopa/levodopa (Sinemet), B6 and carbidopa/levodopa may be given safely together (4).  Though there is no formal guideline recommending testing, it has been suggested by some authors that it might be worthwhile to check one’s blood level, especially if one is taking moderate to high doses of levodopa, or suffering from any of the above-mentioned disease states (1).  Anecdotally, in my clinic I have diagnosed multiple PD patients taking carbidopa/levodopa with B6 deficiencies, where no other clear cause is evident.

The good news is that measuring a B6 level is done with a simple blood test, and replacement may be given with over-the-counter tablets.  However, as with any supplement or medication, care should be taken in replacing B6, and to be clear, I re-emphasize that I am not recommending anyone take a B6 supplement without knowing one’s own blood level first.  Ideally, a qualified physician should interpret the test and tell you whether you need B6 supplementation, and if so, how much you should take.  The United States Recommended Daily Allowance (USRDA) daily dose for men over 50 is 1.7mg, and for women of the same age 1.5mg (5).   In most individuals, this amount is readily obtainable with consumption of a healthy diet. Higher doses are sometimes recommended for short periods, for specific conditions.  You should know that many over the counter vitamin supplements carry amounts of B6 that are far higher than this, sometimes into or over the 500% or “megadose” range.  The Food Intake Board of the Institute of Medicine has established that for men or women over 50, the tolerable daily upper intake level of B6 is 100 mg (6), a dose commonly found in grocery stores.  B6 at this dose is 59 times higher than USRDA for women and 67 times higher than USRDA for men.

Overdosing B6 can be dangerous for several reasons.  Merck Pharmaceuticals has stated that B6 in doses of 10 mg to 25 mg may actually reverse the effects of levodopa by increasing the rate of the enzyme activity that carbidopa is meant to inhibit (4).  In other words, levodopa is depleted in the body before it can get to the brain, where it is needed to work against PD.   There is no mention of the effect on the same enzyme when B6 is given at 100, 500, even 1000mg (all doses which are available in some health food, drug, and grocery stores), though one would suspect it is even more potent in driving down levodopa, and thus worsening the symptoms of PD.  In addition, much like the case when B6 levels in the body are too low, excess B6 is well documented to cause neuropathy, or nerve damage, as well as a disfiguring skin condition ( 7, 8,9). The Weill Cornell Neuropathy Center evaluated all new neuropathy consultations from July 2014 until June 2015 and found that 7% had elevated B6 levels; whereas only 1.5% combined had either B6, or B12 deficiencies (10). Among the total group studied abnormal levels of nutritional factors were implicated in 24%. Likewise, the Peripheral Neuropathy Center at Columbia University evaluated patients over a 10 year period ending in 2012 (11). These were new referrals with an existing diagnosis of idiopathic neuropathy, meaning another physician, typically a neurologist, had not yet determined the cause of the neuropathy. Among these patients, B6 toxicity accounted for 2.5%; whereas B6 deficiency was 0.3%, and B12 deficiency 1.4%.

In summary, some PD patients may have a high or low B6 level, and either may be harmful. There is some concern that carbidopa/levodopa may indirectly drive B6 levels down.  B6 supplements may contain doses that are far too strong for daily use, and toxicity has been linked to neuropathy, and possible negative impact on PD symptoms. If you have concerns about your B6 level, have a qualified doctor check your level and advise you about how and whether to take B6.

REFERENCES (online sources as of mid September, 2016)

1.  Müller et al. Peripheral neuropathy in Parkinson’s disease: levodopa exposure and implications for duodenal delivery.  Parkinsonism Relat Disord. 2013;19(5):501-7.

2.  Urban et al.  Subacute axonal neuropathy in Parkinson’s disease with cobalamin and vitamin B6 deficiency under duodopa therapy. Mov Disord. 2010;25(11):1748-52.

3. Uncini et al. Polyneuropathy associated with duodenal infusion of levodopa in Parkinson’s disease: features, pathogenesis and management.  J Neurol Neurosurg Psychiatry. 2015;86(5):490-5.

4. https://www.merck.com/product/usa/pi_circulars/s/sinemet/sinemet_pi.pdf

5.https://ods.od.nih.gov/factsheets/VitaminB6-HealthProfessional/

6.  http://www.nationalacademies.org/hmd/~/media/Files/Activity-Files/Nutrition/DRIs/New%20Material/4_%20UL%20Values_Vitamins%20and%20Elements.pdf

7. Schaumburg et al.  Sensory neuropathy from pyridoxine abuse: a new megavitamin syndrome.  N Engl J Med. 1983;309(8):445-448.

8. Barak N, Huminer D, Stahl B. Vitamin B6 (Pyridoxine) — excessive dosage in food supplements and OTC medications. Harefuah 2004;143(12):887-90, 910, 909.

9. de Kruijk JR, Notermans NC.  Sensory disturbances caused by multivitamin preparations. Ned Tijdschr Geneeskd. 2005;149(46):2541-2544.

10. Latov et al.  Abnormal nutritional factors in patients evaluated at a neuropathy center.  J Clin Neuromuscul Dis. 2016;17(4):212-214.

11. Farhad et al., Causes of Neuropathy In Patients Referred As “Idiopathic Neuropathy” Muscle Nerve 2016;53:856-861

What is the risk of melanoma in PD?

Melanoma is a dangerous form of skin cancer. The CDC reports that in 2012 almost 68,000 people in the U.S. were diagnosed with melanoma of the skin (1), with a rate of up to 23.6 per 100,000 people in the state of Maine (5). Melanoma has sometimes been associated with other diseases or medications. Increased risk of melanoma in Parkinson disease (PD) was first raised as a concern in 1972 with the report of a PD patient who experienced recurrent bouts of this potentially fatal disease while being treated with levodopa (2). This introduced the question as to whether the case represented a random correlation or an important link between the two diseases. There was also concern of a possible link between levodopa and melanoma because levodopa can convert into melanin, a pigment found both in the dopamine-producing cells of the substantia nigra, and in melanocytes, the cells which can become the tumor of melanoma.

The risk is not clear however. Not all cases of either condition are reported. The best we can do is study the available reports, epidemiology, and disease databases. One case-control study evaluated 862 malignant melanoma cases, compared with 862 age- and sex-matched controls to detect the incidence of PD among melanoma sufferers (3). Among the melanoma patients, 25 (2.9%) had PD. Among the controls, 11 had PD (1.3%). Thus, the odds of having PD was over twofold greater if one had melanoma versus those who did not. A meta-analysis (the combination of multiple studies) using the National Institute of Health (NIH) Medline search engine found that between 1972 and 2010 just over 50 cases of melanoma in PD had been reported in the worldwide medical literature (4). One chart review of 409 PD patients found two with melanoma where 0.3 would have been expected based upon incidence among the same age group (5).

Does levodopa raise the risk of melanoma?
Some authors included information about whether or not people were taking levodopa. A cross-sectional survey (6) to determine the frequency of skin cancers at 12 medical centers in Israel found 9 out of 1,395 patients had biopsy-proven melanoma, 14 had melanoma in their medical history (6 occurring prior to, 8 after PD diagnosis). Twenty patients total (1.4%) had a current or prior melanoma, with an overall rate of melanoma 4.4 times greater than expected for these patients. Melanoma did not correlate with PD duration, PD stage, or the duration of levodopa treatment. In North America, 2,106 PD patients underwent full skin examination with a dermatologist and 346 patients had biopsy of suspicious-looking pigmented skin lesions (7). Twenty biopsies confirmed in situ melanoma (0.95%), and four had invasive melanoma (0.19%). There was no observation of a relationship between levodopa use and melanoma. The DATATOP patient cohort (800 patents entered between 1987 and 1988), followed until 1994, yielded five melanoma cases (1.5 would have been expected, considering the age and gender of the group). Malignant melanoma was diagnosed in two of the patients prior to starting levodopa. The evaluators could not come to any conclusion regarding association between levodopa use and melanoma incidence (8).

Although there have been case reports of patients who started levodopa prior to onset of melanoma, there have also been cases in which patients who remained on levodopa had no recurrence or exacerbation of melanoma. And, no formal relationship between the two has been established in the lab. One epidemiological study attempted to determine if levodopa was causative (9). In this prospective study of 1,099 patients from the Melanoma Clinical Cooperative Group, a single patient was taking levodopa at time of diagnosis. The authors found no role for the induction of melanoma. The link is still not clear (10) and levodopa has not been found to be carcinogenic (cancer-causing) otherwise.

A prospective study of 157,036 people without PD at baseline as part of the Health Professional Follow-up Study and the Nurses’ Health Study took place over a 14-20 year follow-up (11). 616 cases of PD developed. A history of melanoma in a first-degree relative revealed relative risk of 1.85. The authors concluded that PD and melanoma “share common genetic components,” though genes were not identified.

In 2007, some experts considered the ongoing questions and cited the absence of convincing proof of an interaction between levodopa and melanoma, though still offering the following advice: “it would seem prudent not to treat with levodopa if other anti-Parkinson agents remain effective (12).”

In 2009, investigators reviewed five published studies exploring the associations between melanoma, PD, and levodopa (13). They noted the increased risk of melanoma is already present before PD is diagnosed, that as it is unlikely that levodopa plays any role in this phenomenon. The authors noted that while there is a need for further investigation, they also recommended removal of the warning from the drug insert leaflet, noting it might “lead to unnecessary fear on the part of the patients and physician resistance to prescribing this medication.”

Still, the current prescribing information for Sinemet from Merck Pharmaceuticals indicates in regards to reported cases of melanoma in PD, “whether the increased risk observed was due to Parkinson’s disease or other factors, such as drugs used to treat Parkinson’s disease, is unclear….patients and providers are advised to monitor for melanomas frequently and on a regular basis when using SINEMET for any indication. Ideally, periodic skin examinations should be performed by appropriately qualified individuals (e.g., dermatologists).”

Other anti-Parkinson medications
Azilect (rasagiline) is a monoamine oxidase B inhibitor, which boosts dopamine in the brain. The Azilect prescribing information notes, “The increased incidence of melanoma in the Azilect development program was comparable to the increased risk observed in the Parkinson’s disease populations examined in epidemiological studies.” In the lab, rasagiline has been found to actually decrease melanoma growth, and has been tested as a therapy in lab animals (13).
Requip (ropinirole), Mirapex (pramipexole), Neupro (rotigotine) are dopamine agonists. A PubMed search for each generic name and the term “melanoma” yielded zero citations. Prescribing information of ropinirole is silent regarding melanoma; whereas prescribing information of rotigotine and pramipexole note that there is a general two- to six-fold higher risk of melanoma in PD but the connection with drugs is unclear.

Summary
The incidence melanoma among PD patients is at least twice that of the general population, though still in the low single digits. And, if one has melanoma, the risk of developing PD is double that of the general population also. Thus, it would seem that there is a common risk or genetic cause for the two diseases, though the link has not been identified. In spite of the fact that levodopa can covert to melanin, the pigment of melanocytes and melanoma cells, there is no proof that levodopa causes or stimulates melanoma growth. The growth of melanoma seems dependent on genes, not the amount of melanin present. Still, some experts suggest it might be prudent to avoid levodopa in the melanoma patient if possible, and some recommend discarding the warning about levodopa and the risk of melanoma altogether.

1. http://www.cdc.gov/cancer/skin/statistics/
2. Skibba et al. Multiple primary melanoma following administration of levodopa. Arch Pahol 1972;93:556-61.
3. Rigel et al. Evaluation of Parkinson’s disease (PD) prevalence in patients with malignant melanoma. Mov Disord 2006;21:S58.
4. Ferrira, et al. Skin cancer and Parkinson’s disease. Movement Disorders 2010;25(2);139-48.
5. Jansson B, Jankovic J. Low cancer rates among patients with Parkinson’s disease. Ann Neurol. 1985;17(5);505-09.
6. Inzelberg, et al. High prevalence of malignant melanoma in Israeli patients with Parkinson’s disease. J Neural Transm 2011;118(8):1199-207.
7. Bertoni et al. Parkinson’s disease and melanoma: an epidemiologic evaluation. Ann Neurol 2006;60(Suppl 3):S71-72).
8. Constantinescu et al. Malignant melanoma in early Parkinson’s disease: the DATATOP trial. Mov Disord 2007;22:720-22.
9. Sober A, Wick M. Levodopa therapy and malignant melanoma. JAMA 1978;240:554-555.
10. Disse et al. A review of the association between Parkinson disease and malignant melanoma. Dermatol Surg 2016;42:(2)11-46.
11. Gao, et al. Family history of melanoma and Parkinson disease risk. Neurology. 2009;73(16):1286-91.
12. Fahn S, Jankovic J. Principles and Practice of Movement Disorders 2007, chapter 6,131.
13. Vermeij et al. Parkinson’s disease, levodopa-use and the risk of melanoma. Parkinsonism Relat Disord. 2009;15(8):551-3.
14. Meier-Davis, et al. Comparison of oral and transdermal administration of rasagiline mesylate on human melanoma tumor growth in vivo. Cutan Ocul Toxico 2012;31(4):312-17.

What’s so bad about alpha-synuclein?

Alpha-synuclein is a tiny protein found in the neurons of your brain. One of its important jobs is to stabilize tiny bags of dopamine so that they may be released at the synapse where one nerve communicates with another. This makes it a very important little protein.
The problem is that alpha-synuclein, like any protein, has a three-dimensional shape and must be folded correctly. If it is not, bad things happen.   One of the bad things that we have learned in recent years is that “misfolded” alpha-synuclein appears to be able to spread by causing other normal alpha-synuclein to also misfold. I think of it like one bad apple spoiling whole bunch. Once misfolded alpha-synuclein accumulates it cannot be used and cannot be broken down and forms Lewy bodies. Lewy bodies, clumps of proteins in the affected neurons, are the pathologic hallmark of PD.

Diagnosis
The good news is that in recent years studies have shown that alpha-synuclein may be used to diagnose Parkinson’s disease. These misfolded proteins have been detected on biopsies of the GI tract and in the salivary glands. It appears that the FDA may soon approve a test to diagnose PD by needle biopsy of the salivary gland. Studies of early and advanced Parkinson’s patients have shown positivity in about 75% of patients (1).

Therapy
Alpha-synuclein is also a target for therapy. After multiple animal trials, there have been three human studies involving monoclonal proteins targeting alpha-synuclein. These monoclonal proteins are antibodies which target alpha-synuclein. Antibodies are little Y-shaped proteins that our immune system produces to fight virus, bacteria, and any foreign protein. The idea with this drug was to target alpha-synuclein as a foreign protein. The first trial was with healthy subjects and was able to completely bind all alpha-synuclein found in the blood. The drug was very well tolerated. To learn more about this visit www.clinicaltrials.gov and type in the search bar NCT02095171.  An ongoing trial with six PD patients in multiple centers around the country will measure alpha-synuclein and the antibodies in cerebrospinal fluid (NCT02157714). A third study with a different monoclonal protein produced by Biogen is being tested and 40 healthy volunteers in Texas and Indiana. There are also vaccines under trial. It should be noted that animal studies have shown removal of Lewy bodies and resolution of abnormal behaviors in animal models of PD, making vaccine a very attractive possibility (2). AFFiRis has done the only human trial with vaccine in Vienna, Austria. The first phase I study with a vaccine called PD01A was given to 32 early PD patients and, as this was a safety study, was tolerated quite well (3). Patients were entered into long-term follow-up and the data is not back on that yet.

Other Drugs
There was a press release a few months back about the drug Nilotinib reporting marked improvements in a few PD and LBD patients.  Nilotinib is already FDA approved for the treatment of chronic myelogenous leukemia, and was shown in a mouse model to reduce the activity of an enzyme called c-Abl (4). c-Abl is activated in PD patients and is associated with overproduction of alpha-synuclein.  Animal studies have shown that injection of either alpha-synuclein or c-Abl will increase levels of the other; excess of either can lead to LB formation and cell death.  Nilotinib is currently in a phase I clinical trial for the treatment of PD and DLB (NCT02281474) at Georgetown U.  The study will measure changes in alpha-synuclein and the protein tau as the primary outcomes, and will include 36 iPD patients.

A phase III study of 86 patients with a form of parkinsonism called MSA at University of Munich is currently recruiting participants (NCT02008721) for the compound EGCG, a polyphenol found in green tea and widely used in dietary supplements.   It has been shown to inhibit the formation of toxic alpha-synuclein (5).  There is no data as yet.

There are other trials targeting alpha-synuclein, which should give patients hope. I think we are standing in the doorway to a new era of treatment.

-Bill Stamey, M.D.

1. Movement disorders.  2016:31 (2) 250-56
2. Neuron 2005;46:857-68
3. Park Relat Disord 2012;18(Suppl 1): S11-13)
4. Sci Rep 2014;4:4874
5. Proc Natl Acad Sci USA 2010;107:7710-7715