InnoCentive

The following updates represent a summary of our progress with the hypotheses generated by InnoCentive solvers. We encourage you to take this material and go deeper.

We want to see where the dialogue takes the community, as we digest feedback, move forward in our distillation of the material here and generate new hypotheses.

Update 09/06/2012


New feedback from our consulting neurologist on recently suggested hypotheses:

Regarding using blood (serum) tests to evaluate for limbic encephalitis (as an alternative to an invasive lumbar puncture):
The patient is showing no signs of encephalitis. With this condition, you see severe altered mental status, usually obtunded and unable to function. The seizures are much more severe and the EEG’s are grossly abnormal.


Regarding ordering CBC with morphology and plasma fragility and cold agglutination tests:
Yes, I agree that the ordering doctor must ask for a manual examination of both the red and white blood cells otherwise it is a machine differential. However, I have never seen a patient present with seizures that were caused by thalassemia.


Regarding the Protein S deficiency Hypothesis:
Everything this solver wrote about Protein S is correct. It manifests as a hypercoagulable state and causes abnormal clotting. It is important in neurology because strokes are a common manifestation. Strokes and central venous thrombosis causes paralysis, increased intracranial pressure and seizures. However, on the multiple MRI’s and CT’s there have never ben signs of stroke or venous thrombosis, so the patient is not likely to have Protein S deficiency.


Regarding the hypothesis that the the patient has an acquired condition in which the effects of GnRH releasing cells within the hypothalamus are able to excite not just gonadotropic cells in the anterior pituitary but also areas of the hypothalamus itself, triggering both hypertension and seizures in a periodic fashion:
This is an interesting hypothesis. However, seizures originating from the hypothalamus are extremely rare and have very characteristic manifestations. Seizures in the hypothalamus are almost exclusively seen with hamartomas of the hypothalamus. These seizures manifest as a sudden burst of laughing or crying. The blood pressure changes seen in seizures are believed to be due to the ultimate spread of seizure activity from one of the cerebral lobes through the hypothalamus. However, it would be reasonable to obtain a GnRH level when our patient undergoes a prolonged EEG in a hospital and a seizure occurs.

Update 09/04/2012

Our team has been contemplating this post by Daniel Labbé (Aug 31, 2012 11:19):
  • An elegant solution to a complex problem usually involves finding a parsimonious solution. As several suggestions have been posted, but considered on a case by case basis, perhaps it would be possible to synthesize some of the ideas into smaller reviews. Thus, several ad-hoc hypotheses could be joined into a coherent theory of the syndrome.
  • The challenge would be to put the specialists and expertise in the right place. For example, several MDs and other "experts" in their respective fields have suggested various solutions, while speculating on supporting hypotheses outside their field of expertise - including myself. Some subjective reports have been made as well, which could also be taken into account.
  • What do you say, as a community, could we summarize the "best of" for each commentary and together assess the plausibilities? Thereby, we would relieve Simon from some of the work. Naturally, the final decision is not for us to make. There are several good ideas posted already, and some have already been evaluated.
This is exactly the question we have been wrestling with. We have over 70+ pages of high quality hypotheses and suggestions. Now - how do we take our initial results from our ongoing community dialogue, and take action to advance that discussion to distill and refine these ideas, and see if that process sparks new hypotheses?

Here is one idea:
We have yet to converge on a single diagnosis. We believe that the body of responses, when analyzed in whole, could yield a differential diagnosis or at least the one or two areas where to focus medical treatment. However, we claim no expertise in bioinformatics or medical data mining.

Does anyone in this solver’s community have experience with these techniques? We are looking for some guidance about what methodology, visualization approach, and / or software tools to apply to the 70+ pages of postings.

Update 09/02/2012

More new updates from our physician team on specific hypotheses from Solvers:

Hypothesis:
Carbon monoxide poisoning.

Feedback:
Common morbidity involves myocardial and/or neurologic injury including delayed neurologic squeala. The signs and symptoms of carbon monoxide intoxication include; headache, difficulty with coordinating, dizziness, difficulty breathing, irritability, chest pain, confusion/memory loss and in more serious cases cerebral edema, disorientation and seizures.
Although you are correct, there is endogenous production of carbon monoxide, and adding to an environmental baseline would elevate blood levels. However, these levels have never been shown to cause seizures alone. Altered mental status and neurological impairment result from cerebral edema and basal ganglia damage. These factors are known to occur well before seizures manifest. In carbon monoxide poisoning there is classic MRI findings of basal ganglia injury or cerebral edema. Our patient has minimal signs of hippocampal sclerosis that are not associated with carbon monoxide poisoning.

Hypothesis:
Pilocarpine-induced seizures

Feedback:
Pilocarpine has been used as an ophthalmologic agent for decades; however, toxicity rarely has been reported in the medical literature. Oral pilocarpine tablets, as well as another muscarinic agent (cevimeline), are approved for the treatment of dry mouth (xerostomia). Reported cases of unintentional overdose of pilocarpine resulted in bradycardia, mild hypotension, and other muscarinic symptoms. Our patient did not have these symptoms. Seizures have not been reported in the medical literature as a primary manifestation of pilocarpine toxicity.

In animal studies high-dose treatment with pilocarpine hydrochloride, a cholinergic muscarinic agonist, induces seizures in rodents following systemic or intracerebral administration. Pilocarpine seizures are characterized by a sequential development of behavioral patterns and electrographic activity. Hypoactivity, tremor, scratching, head bobbing, and myoclonic movements of the limbs progress to recurrent myoclonic convulsions with rearing, salivation, and falling, and status epilepticus. The sustained convulsions induced by pilocarpine are followed by widespread damage to the forebrain. The amygdala, thalamus, olfactory cortex, hippocampus, neocortex, and substantia nigra are the most sensitive regions to epilepsy-related damage following convulsions produced by pilocarpine. Spontaneous seizures are observed in the long-term period following the administration of convulsant doses of pilocarpine.

Based on the fact that our patient does not manifest any of the common signs of pilocarpine toxicity, it is unlikely there is enough pilocarpine in her body to cause seizures. In addition, pilocarpine is an agent that causes low blood pressure and not high blood pressure as sen in our patient.

As an aside, the lip smacking in some partial seizures is not similar to the oral-lingual movements of tardive-dyskensia (TD). TD is repetitive tounge darting that is dramatic and occurs only when the patient is awake. The lip smacking seen in some partial seizures has limited tongue movemet, is relatively subtle and occurs only during seizure activity. TD only occurs in those that have taken traditional anti-psychotic medications such as Haldol. The mechanisms are completely different.

It is true that routine EEG’s record the surface brain electrical potentials and do not represent subcortical activity. Prolonged EEG’s are performed to record a seizure that originates sub-cortically that virtually always spreads to the cortex surface where the EEG would record the abnormal epileptiform activity.


Update 09/01/2012

More new updates from our physician team on specific hypotheses from Solvers:

Hypothesis:
Episodes of tachycardia episodes leads to dyssynchronous activity in the atria and ventricles due to the minor block observed in the ECGs, decreasing cardiac output, leading to hypoxemia, which in turn would lead to seizures.

Feedback:
The mechanism described above is correct in that prolonged, severe tachycardia can lead to eventual reduced cardiac output and under more extreme conditions, cerebral ischemia.

First degree AV block manifest on a surface ECG as a prolonged PR interval. The normal PR interval is from 120 ms to 200 ms in length. This is usually an incidental finding on a routine ECG. Typically, first-degree heart block does not require any particular investigations except for electrolyte and drug screens, especially if an overdose is suspected.

The degree of tachycardia and oxygen desaturation is not nearly severe enough to actually cause a seizure. The tracing showing tachycardia and reduction in oxygen saturation is a common result of seizure activity. It is therefore important to have a prolonged EEG to capture a seizure and more definitively correlate heart rate and blood oxygen levels as you suggested. However, isolated first-degree heart block has no direct clinical consequences. There are no symptoms or signs associated with it.

Hypothesis:
The patient’s symptoms are neurological manifestations of Behcet's disease.

Feedback:
Behçet's disease is a rare immune-mediated systemic vasculitis that presents with mucous membrane ulceration and ocular (eye) involvements. The classic triad of symptoms include recurrent oral ulcers, genital ulcers, and uveitis. As a systemic disease, it can also involve visceral organs such as the gastrointestinal tract, pulmonary, musculoskeletal, and neurological systems. The oral lesions are similar to those found in inflammatory bowel disease and frequently recur. Painful genital ulcerations usually develop around the anus, vulva, or scrotum and cause scarring in 75% of the patients.

Neurological involvements range from aseptic meningitis to vascular thrombosis such as dural sinus thrombosis and organic brain syndrome manifesting with confusion, seizures, and memory loss. Neurological manifestations appear late in the progression of the disease.

According to the International Study Group guidelines, for a patient to be diagnosed with Behçet's Disease, the patient must have oral ulcers (any shape, size, or number at least 3 times in any 12 months period) along with 2 out of the following "hallmark" symptoms:

1) genital ulcers (including anal ulcers and spots in the genital region and swollen testicles or epididymitis in men)
2) skin lesions (papulo-pustules, folliculitis, erythema nodosum, acne in post-adolescents not on corticosteroids)
3) eye inflammation (iritis, uveitis, retinal vasculitis, cells in the vitreous)

On rare occasion patients have presented early in the disease with partial seizures, but is usually accompanied with visual and mental status symptoms. If the initial prominent symptom is seizures, then the other classic signs (eye and skin lesions) will present shortly after.

Our patient has not had any of the required triad of symptoms to consider Brehcet’s disease as the cause of her symptoms of hypertension and seizures.


Update 08/30/2012

Our neurologist has sent us this feedback on recent questions.

1. I note that 'endocrine tests were done'. What where they? Those listed merely relate to pituitary function. Have adrenal tests been performed? Has hypoglycemia been formally excluded?

Adrenal tests have been performed extensively at the NIH when they were evaluating for pheochromocytoma. Glucose levels were obtained in the emergency room after she had a generalized seizure. Hypoglycemia is a common cause of generalized seizures and all emergency rooms test for this upon arrival to the hospital.

2. Have the serum electrolytes - particularly potassium, calcium and magnesium - been checked during a crisis?

Electrolyte levels (potassium, calcium, magnesium) levels were obtained in the emergency room after she had a generalized seizure. These are relatively common causes of generalized seizures and all emergency rooms test for these upon arrival to the hospital.

3. Besides the two chem panels on 2010-11-23, are there any others in the medical history? Does patient have a chem panel with WBC/differential once per week for 8 weeks (to cover cycle)? Chronic low cellular electrolyte concentrations increases sensitivity to blood pH challenges. Challenges can include CO2 in environment, or from minor sleep apnea or breathing irregularities, or processed foods containing taste enhancers (MSG/excitotoxins) or carbonation etc.

There have been multiple chemistry panels obtained through our patient’s medical history. Only some were listed because the tests were all normal. We have not obtained a chem panel with WBD and diff once per week for 8 weeks. There have been no signs of low levels of electrolyte levels or CO2 levels on any tests to date.

Update 08/23/2012

We recently sent our consulting neurologist the list of labs that have been suggested by InnoCentive solvers. His feedback is copied below.

a. Ca-125: (cancer antigen 125 or carbohydrate antigen 125). CA-125 has found application as a tumor marker or biomarker that may be elevated in the blood of some patients with specific types of cancers. Ca-125 biomarker for ovarian cancer detection.
Our patient has no signs of ovarian cancer. Her CT and MRI of the abdomen revealed normal ovarian abnormalities and her hormone levels were normal as well.

b. B12- Levels have been normal twice and her symptoms are not consistent with B12 deficiency.

c. B6- Levels have been normal her symptoms are not consistent with B6 deficiency.

d. vit D3- Levels have been normal her symptoms are not consistent with Vitamin D deficiency.

e. recent prolactin- Although there is not a recent prolactin levels, we do have prolactin levels when she was post-partum and subsequent levels that were within normal limits for her stage at the time.

f. IMM- is a measurement of the different forms of human chorionic gonadotropic hormone and is considered by few to be a way to detect types on cancer. However, IMM has not been established as a cancer marker and not considered a marker for cancer.

g. TNF-alpha- Tumor necrosis factor primary role of TNF is in the regulation of immune cells. is able to induce fever, to induce apoptotic cell death, to induce sepsis (through IL1 & IL6 production), to induce cachexia, induce inflammation, and to inhibit tumorigenesis and viral replication. Dysregulation of TNF production has been implicated in a variety of human diseases, including Alzheimer's disease,[1] cancer,[2] major depression,[3] and inflammatory bowel disease (IBD).[4] While still controversial, studies of depression and IBD are currently being linked by TNF levels.

Although TNF is associated with many different symptoms, none are related to our patient. TNF is not associated episodic hypertension or seizures.

h. IL-6- is an interleukin that acts as both a pro-inflammatory and anti-inflammatory cytokine. IL-6 is relevant to many diseases such as diabetes,[15] atherosclerosis, depression, Alzheimers Disease, systemic lupus erythematosus, prostate cancer, and rheumatoid arthritis.

IL-6 is not associated episodic hypertension or seizures.

i. H. pylori titre. This is a bacteria that causes ulcers and is not associated episodic hypertension or seizures.

j. SPEP- Serum protein electrophoresis (SPEP) is a laboratory test that examines specific proteins in the blood called globulins in condition that involves immunoglobulins. Not related to our patients symptoms.

k. several repeat cultures for B. burgdorferi. Our patient has had 4 cultures and all have been negative. Several doctors are evaluating her for related spirochete conditions.

l. Whole genome sequencing. There is no known genetic syndrome with these specific symptoms. Complete genome testing is not available to the public at this time.

m. cytokine testing performed before and during and post episodes. There are an extensive amount of cytokines in the human body. One would have to order a test to look at specific cytokines, not all of them. In addition, here are not cytokine related conditions that would correlate with our patients symptoms.

n. polysomnographic observation

o. ANCA tests to rule out vasculitis associated pathology. An ANCA test and/or tests for MPO and PR3 are ordered when a person is suspected of having a systemic autoimmune vasculitis, especially when symptoms suggest Wegener's granulomatosis or microscopic polyangiitis. Early in the disease, symptoms may be vague or nonspecific, such as fever, fatigue, weight loss, muscle and/or joint aches, and night sweats.

Our patient has had 2 nor ESR levels which is a very good indicator of inflammation and vasculitis.

p. Expanded (not targeted) cytokine analyses. Condition related to abnormal cytokine levels are not related to our patient’s symptoms or hypertension and seizures.

q. Urinary metabolic profile- is a measure primarily of organic acids. Our patient does not show any signs of an organic acid disorder.

r. Full autoimmune panel. Our patient does not show signs of an autoimmune condition.


Update 08/21/2012


Feedback from our consulting neurologist on the proposed Endometriosis hypothesis:
This is an interesting presentation. Aberrant tissue of one from one organ can develop in other locations in the body as a result of congenital issues. Endometriosis has been documented in nearly every location of the body, including such remote and unusual sites as the brain. The endometrial tissue in the brain would react in a similar fashion as the uterine endometrial tissue during the monthly cycle of hormones, possibly resulting in irritating the adjacent brain tissue, thereby causing a seizure.

However, there have been several scientific publications of MRI’s in patients with brain endometriosis. (See attached photo.)

Our patient has 2 MRI’s that did not have any mass effect. One was read as normal, the second as possibly minimal signs of hippocampus sclerosis .

Feedback from our consulting neurologist on the proposed hypoglycemia hypothesis:
These are all very good points you raise.

Hypoglycemia can cause both hypertension and seizures. However, hypoglycemia virtually always causes generalized seizures and rarely partial events. Insulinomas’s cause random, intermittent episodes of hypoglycemia. This patient has not had symptoms of hypoglycemia (sweating, tremors) at any time during the day. She has a random, fasting and glucose levels in the hospital shortly after experiencing a seizure and the levels have always been normal. However, if this patient is experiences a seizure near or in a healthcare setting, a glucose levels is a standard lab test that should be drawn. Additionally, this patient has had normal insulin and c-peptide levels recently.

Adrenal insufficiency can result in hypoglycemia and is typically not associated with an adrenal mass. This patient has had 3 rather extensive endocrine work-ups including all adrenal related hormones, all of which have been normal. The pancreas has been imaged in 2 abdominal CT scans and 2 MRI’s. The anatomy of the pancreas was normal in all.

Notes on tests and appointments:
We are still evaluating the list of suggested tests. We will be scheduling a CA-125 test (to address the endometriosis hypothesis) and a thyroid peroxidase (TPO) test (to address the thyroiditis hypothesis) at her next doctor’s visit.

The patient has a consultation scheduled with a kidney specialist on 8/31, and a consultation with a cardiologist scheduled on 9/6.

Update 08/20/2012

We received this feedback today from our consulting neurologist, in response to the set of questions we posted yesterday.

Feedback regarding the noted fluctuations in MCV values:
You are correct that there is a relatively wide variation in this patients RBC MCV. It must be realized that nearly all hematology tests were run in different labs. These tests are run on fully automated machines that report these numbers. This variation of MCV is unusual if the same persons blood was tested on the same analyzer several times. However, the quality and validity of each machine is different.

But more important is considering “if” there was a wide variation in MCV, what might this mean. Vitamin B12 is the first to consider. This patient had normal B12 levels and her clinical picture does not resemble B12deficiency. Anemia is a prominent finding in B12 deficiency as well as enlarged spleen or liver and gastrointestinal symptoms and abdominal pain. Also, the neurological symptoms include delirium, confusion, spastic ataxia, and peripheral neuropathy. This patient does not have these signs or symptoms.

Feedback regarding the suggestion that the patient should start a challenge elimination diet:
This is always a reasonable idea when someone is experiencing episodes of physical events. This is most commonly used with people who experience migraine headaches. The idea is to record you daily intake on a calendar and also mark the physical events the patient is experiencing. After 3-4 events, go back and see if there is a correlation with any foods. Then eliminate that food and see if the physical event recurrences. This can be repeated every time you note a correlation between diet and the event.

Feedback regarding the suggestion that the patient should not take labetalol prn:
The onset of action of labetalol is about 5 minutes, so it works quickly after she takes it. I guess it depends on how often she has to take it and how long it lasts. It can cause postural hypotensive as it blocks high blood pressure. I would reserve the question of hypertension management to your internal medical doctor or cardiologist.

Feedback regarding the suggestion that we test to determine if the spikes in BP are causing the seizures:
I agree with this. In fact, the seizures may even be causing spike in BP. She could be having partial seizures with minimal observable manifestations, but the seizure could affect the part of the brain that regulates BP. The way to test for this is to have the 96 hour in-hospital EEG with continuous BP monitoring. Then, when she has a documented seizure with EEG correlated epileptic activity you can match it with the BP. Does the elevated BP begin immediately, during or after the seizure?

Answer to question regarding the nature of the NeuroScreen test:
This neuroscreen was done on a blood specimen. It is not a meaningful test in mainline medicine. I do not believe it is an accurate analysis of the neurotransmitters in the brain.

Feedback regarding testing for orthostatic intolerance:
Orthostatic intolerance is referring to autonomic nervous system dysfunction. We spoke about this in out last project. This is tested for by the Tilt-table test in cardiovascular labs. Anyone with persistent orthostatic hypotension should have this test.

Feedback regarding the proposed hypothesis of Hashimoto’s thyroiditis:
You are correct that TSH, T4, T3 are normal in the early stages of Hashimoto's thyroiditis. Elevates levels of antithyroid peroxidase antibodies and antithyroglobulin antibodies must be present for the diagnosis.

However, clinical symptoms will not manifest unless the condition has progressed and anti-thyroid antibodies are present with subsequent elevated TSH and low T4 levels. Patient with Hashimoto's thyroiditis suffer from hypothyroidism not hyperthyroidism, so thyroid storms would not occur and would not account for episodes of hypertension. The episode of hyperthyroidism that occurs before hypothyroidism sets in is in the condition of Graves’s disease, not Hashimotos.

Hashimoto's encephalitis is rare, but can manifest with seizures. More typically it presents with disorientation, aggression, ataxia, tremors, aphasia and personality changes that are not subtle.

You are correct that the first EEG was described as, “The EEG is mildly abnormal due to disorganization of the background and mild slowing, which is either due to nonspecific encephalopathy or drowsiness”. However, other EEG’s were entirely normal.

Feedback regarding the proposed hypothesis of limbic encephalitis:
Limbic encephalitis and neurological manifestations of a paraneoplastic syndrome are conditions that could manifest with this patients signs and symptoms. Both conditions rare and can be elusive. You are correct that both are characterized by personality changes, irritability, depression, seizures, memory loss and sometimes dementia, but all symptoms do not need to be present.

A major criteria that must be fulfilled for Paraneoplastic limbic encephalitis is: “at least one of the following must be present: CSF with inflammatory changes but negative cytology; MRI demonstrating temporal lobe abnormalities; EEG showing epileptic activity in the temporal lobes.” As you noticed in your case reports the condition may manifest in “recurrent episodes of altered behavior and 'odd' episodes”, but is also typically a significant baseline change in personality and/or behavior (“He had become apathetic and uninterested in his family)

MRI and CT of the brain did not show temporal lobe abnormalities. Two EEG’s where completed and no epileptiform activity was present. This does leave the question of spinal fluid with inflammatory changes. However, without more to support the diagnosis of paraneoplastic limbic encephalitis I would not recommend the patient undergoing a lumbar puncture for CSF analysis.

Others have raised the questions of a variant of Lyme disease, so if spinal fluid analysis is obtained for evaluation of central nervous system B. burgdorferi, white blood cell count, cytology for malignancy, as well analysis for anti-Hu, Ma2 Amphiphysin, anti-CV2, Ri and anti-NMDAR will be ordered to rule-out Paraneoplastic limbic encephalitis

Feedback regarding the proposed hypothesis that the seizures are caused by a pinched nerve or structural issue in the neck:
Although compression of specific peripheral nerves can affect function of the autonomic nervous system and therefore blood pressure, a compressed nerve cannot evoke a seizure.

There are fields of healthcare, such as chiropractic and massage therapy, that are based on principles of manipulation of peripheral nerves and muscles. Specifically, massage therapists treat clients by using touch to manipulate soft-tissue muscles of the body. Therapists aim to relieve pain, rehabilitate injuries, reduce stress, increase relaxation, and aid in the general wellness of clients. However, central nervous systems disorders, such as seizures, cannot be directed effected by these maneuvers.

As you mentioned, it is always important to monitor for any timing or activity that may be associated with a person’s seizures. If a pattern can be established, then a trial should be undertaken with a different schedule to see if there is a change in the frequency of the seizures.

Update 08/19/2012

The is the first set of questions we developed based on the InnoCentive postings to date:

  1. Questions pertaining to MCV values on labs
  2. Hashimoto’s as a hypothesis – plausible?
  3. Endometriosis as a hypothesis – plausible? On this same note, we have had repeated questions regarding undetected retained products of conception and also one on microchimerism.
  4. Hypoglycemia as a hypothesis – plausible?
  5. Limbic encephalitis – plausible?
  6. Pinched nerve, structural issue with neck - plausible?
  7. One solver (a physician) advises strongly against prn use of labetalol. Thoughts?
  8. Suggestion that we follow up more closely on the proteinuria. Thoughts?
  9. NeuroScreen test of 4 Feb ’11 is unclear. Was this plasma, serum, CSF, blood or what?
  10. One solver suggests that we test of high blood pressure directly causes the seizures in the following fashion: Give the patient medication in a hospital setting to trigger high blood pressure and see if seizures result. Thoughts?
  11. One solver asks if we have had tests performed to evaluate the patient for orthostatic intolerance. What would be the testing procedure for that?
  12. Several solvers have suggested a challenge elimination diet. Thoughts?
  13. Several solvers have suggested that this may be necrotic pheochromocytoma. How would that condition be different, and was it ruled out by the NIH testing? Also several have asked about non-excreting pheo.
  14. Feedback on suggested labs:
    1. Ca-125
    2. B12
    3. B6
    4. vit D3
    5. recent prolactin
    6. IMM
    7. TNF-alpha
    8. IL-6
    9. H. pylori titre
    10. SPEP
    11. several repeat cultures for B. burgdorferi
    12. Whole genome sequencing
    13. cytokine testing performed before and during and post episodes
    14. polysomnographic observation
    15. ANCA tests to rule out vasculitis associated pathology
    16. Expanded (not targeted) cytokine analyses
    17. Urinary metabolic profile
    18. Complete autoimmune panel