Monday, 19 May 2008

Answer To Recent Case

Here are the answers to the recent case-- I am sorry if it is overly long to read but I hope it will cast some light on this difficult scenario.

Question 1: Please make a problem list soley from the history and physical examination findings.

Problem list

  • Unsteadiness
  • Slurred speech
  • Periperal numbness (on history and physical)
  • Diplopia
  • Meningoencephalitis
  • Seizures
  • Medications- Phenobarbital / Phenytoin / Sodium Valproate / Flunitrazepam / Metoclopramide
  • Fine tremor
  • Coarse Facies
  • Alopecia
  • Cerebellar signs
  • Absent reflexes
Question 2: What other specific question(s) would you like to learn from the history that might help you with this diagnosis?

The drug history is very important here and the drug doses and frequency should be investigated in addition to searching for hidden alcoholism.

The fact that the patient is happily married does not exclude her from being infected with a sexually transmitted infection by her partner and hence, a discussion with the partner might be in order especially if an unusual infectious aetiology were to be found e.g. neurosyphilis.

Moreover, the fact that the patient works in the 'fashion industry' is very non-specific. For example, she might make jewellery!! She might be using Lead!! Lead can cause cerebellar disease too ! Hence, a more in-depth questioning about her occupation would be in order here.

Question 3: What particular feature of the oral examination might give you a further clue regarding the cause of this problem.

Given the fact that the patient takes Phenytoin, one chronic side effect is gum (gingival) hypertrophy. This might give the clinician a clue regarding the cause of some of the symptoms and signs.

Question 4: What is the diagnosis?

The fact that the patient has severe cerebellar signs, peripheral neuropathy, a resting tremor, coarse facies, alopecia and gum hypertrophy, a diagnosis of acute on chronic phenytoin toxicity would be top of the list.

However, when one thinks of cerebellar dysfunction, one needs to consider various other causes such as:

Drugs- Phenytoin, flunitrazepam (can cause drowsiness, slurred speech, tremor), Phenobarbital (known to cause ataxia), Na valproate (known to cause tremor, ataxia, speech disorder.
Salicylates, aminoglycosides, sedatives ( e.g. benzodiazepines), chemotherapy agents e.g. fluorouracil, cytarabine
Infection- Viral encephalitis, HIV, CJD, abscess
Connective Tissue- SLE, Sjogren’s disease, Hashimoto’s thyroiditis, Auto-Ab associated Cerebellar Ataxia
Neoplastic- primary ( e.g. meningioma, haemangioblastoma) or secondary brain tumour (melanoma, lung, breast, thyroid etc), and Primary CNS Lymphoma. Cerebellar pontine angle tumour e.g. acoustic neuroma associated with neurofibromatosis type 2.
Paraneoplastic cerebellar dysfunction can occur with Breast Ca, Lung Ca (mainly SCC), Ovarian Ca, Uterine Ca and Cervical Ca.
Haematovascular- posterior circulation stroke (infarction or bleed), dissection.
Immune- Multiple sclerosis, Guillain-Barre Syndrome (Miller-Fisher variant)
Metabolic- Alcoholic cerebellar degeneration, Lead poisoning.

Granulomatous- sarcoidosis
Degenerative- olivopontocerebllar disease (may be part of a multi-system atrophy)

Congenital- Friederich’s ataxia- cerebellar degeneration, peripheral neuropathy, spinal degeneration, pes cavus, autosomal recessive, onset age 20; 1:50,000 prevalence (one of the several causes of absent reflexes but extensor plantar response). There are several other more rare congenital ataxias.

Question 5: Which one laboratory test would give you the diagnosis?

The primary lab test to be done is the serum phenytoin level. This correlates well with symptoms when there is toxicity.

Phenytoin is a well established drug used for various conditions including seizures and dysrhythmias (although the latter used less commonly these days). It can be given intravenously or orally.

Normal range for phenytoin in Japan is 10-20mg/L.

Above 20mg/L, nystagmus is common. Above 30mg/L, tremor, ataxia and slurred speech occur. When levels exceed 40mg/L, lethargy, confusion and unconsciousness occur. There have only been a handful of patient cases who have survived with levels >100mg/L.

Phenytoin is protein bound and hence, in patients with low albumin levels or renal failure, patients may experience toxicity at lower serum concentrations.

Phenytoin exhibits dose-dependent elimination kinetics, and toxicity can occur easily in patients on chronic therapy especially when there are drug interactions or slight alterations of daily dose. As the phenytoin level reaches therapeutic range, the liver elimination of the drug becomes saturated and therefore becomes Zero-Order in Kinetics. Hence, the half-life increases as serum levels rise e.g. T1/2 = 26 hours at 10mg/L, 40 hours at 20mg/L and 60 hours at 40mg/L respectively.

There is no specific antidote available for treatment of phenytoin toxicity whether iatrogenic or intentional and treatment is supportive e.g. charcoal for elimination (if orally administered) and maintaining airway, breathing and circulation.

In this case, the patient was also taking several other drugs and hence, drug-drug interactions must be considered here and include:

Phenytoin-Valproic Acid interaction- valproate displaces phenytoin from protein binding causing toxicity

Phenobarbital - Valproate Interaction- valproate reduced metabolism of Phenobarbital and can result in toxicity. One case report of decreased valproate effect through increased metabolism.

Phenobarbital – Flunitrazepam interaction- decreased effect of benzodiazepine through increased metabolism induced by phenobarbital. Increased risk of respiratory failure with combination of benzodiazepine plus phenobarbital (one case report 1986).

Benzodiazepine – Phenytoin interaction - decreased effect of benzodiazepine through increased degradation via liver enzyme induction. Possible toxicity when phenytoin and benzodiazepine combined although mechanism not established.

Benzodizepine – Valproate interaction – possible benzodiazepine toxicity by displaced protein binding by valproate and also possible decreased detoxification by the liver (glucuronidation).

Hence, there are several potential drug interactions at work in this patient. Other drugs being used to control the seizures may result in phenytoin toxicity, and toxicity or reduced activity of other drugs.

I have no doubt that drug-drug interactions contributed to the acute on chronic toxicity. However, another part of the history found out later was that the patient had poor daily compliance and she took sometimes more medication than prescribed in order to make up for lost doses.

The usual daily dose is up to 300mg of phenytoin and only several cases has there been doses up to 500mg/day. This patient was apparently taking doses far in excess of the recommended daily dose.

Other clues to the diagnosis of phenytoin chronic administration were the mild leucopenia and thrombocytopaenia which are also well described. The CT and MRI scan of the cerebellum were both entirely normal.

Case Response From Dr Masami Matumura from Kanazawa Medical University. As usual, Dr Matsumura provided an excellent breakdown of the case, and is as follows;

Question 1:Please make a problem list soley from the history and physical examination findings.

#1 Ataxia
She showed unsteadiness when walking, gait- wide based, staggering in nature, heel to toe test positive, heel to shin test positive R>L, finger to nose test positive bilaterally, dysdiadochokinesis positive bilaterally, nystagmus, slurred speech, and fine resting tremor.

#2 Polyneuropathy
She had numbness of her arms and legs, mildly reduced light touch, and no reflexes.
#3 Diplopia
#4 Pinky-red cheeks
#5 Coarse skin
#6 Alopecia.
#7 Viral meningoencephalitis
#8 Post-meningitis seizures treated with medication

Interestingly, she has two neuro-anatomical disorders, ataxia and polyneuropathy.
I have to pick up all possibilities in this case.

Differential diagnosis:
  • Vascular: Cerebellar infarction, Subdural hematoma
  • Infection: Cerebellar abscess, HIV, Acute viral cerebellitis, Lyme disease, Syphilis, Lepromatous leprosy,
  • Neoplastic: Paraneoplastic syndrome, Cerebellar glioma, Cerebellar metastasis
  • Collagen (autoimmune): Systemic lupus erythematosus, Sarcoidosis, Sjogren’s syndrome, Dermatomyositis, Primary biliary cirrhosis, Guillain-Barre syndrome, Post infection syndrome
  • Toxic/Metabolic: Phenobarbital, Sodium valproate, Metoclopramide, Lead, Aluminum, Hypothyroidism, Vitamin B12 neuropathy, Diabetes
  • Trauma/Degenerative: Spinocerebellar atrophy
  • Iatrogenic: Phenobarbital, Sodium valproate, Metoclopramide
  • Idiopathic: Amyloidosis, Multiple sclerosis
  • Congenital: Fabry’s disease, Friedreich’s ataxia, Porphyria
She could not describe onset of symptoms. I assume these conditions were acute course or subacute course. If it was true, infection, neoplastic, collagen, toxic, and metabolic causes must be considered.

Question 2: What other specific question(s) would you like to learn from the history that might help you with this diagnosis?

I would ask her whether she has episodes of skin rash after sun exposure or not. And I would ask her change of voice and feeling cold or not.

Question 3: What particular feature of the oral examination might give you a further clue regarding the cause of this problem?

I seek oral ulcers!

Question 4: What is the diagnosis?

Systemic lupus erythematosus (SLE) is highly suspected in this case. She had two neuro-anatomical disorders, pinky-red cheeks, coarse skin, and alopecia.
Second candidate is autoimmune thyroiditis. She is Japanese woman.
Interestingly, both SLE and Hashimoto’s thyroiditis can be associated in several cases.

I can’t explain the cause of diplopia. Opthalmopathy is not described.

Metoclopramide is often prescribed in many patients. This drug can cause extrapyramidal sign. Tremor might be caused by metoclopramide in this case.

Question 5: Which one laboratory test would give you the diagnosis?

Serum antinuclear antibodies.
If it is possible, I would order CBC, FT4, and TSH.

Comments from B:

Dr Matusmura raises some very interesting points with his case interpretation. SLE is still a possibility. However, the pink cheeks on this patient did not involve the bridge of her nose, which the classical ‘Butterfly Rash’ of SLE does. Moreover, skin rashes are known to be a manifestation of phenytoin administration as well.
Alopecia is indeed a feature of SLE although it is not a pathognomic feature and can be found in phenytoin toxicity too—as can hirsuitism!
However, interestingly, phenytoin has been associated with inducing SLE in several patients although this tends to be a discoid-type rather than systemic.

Lastly, when the phenytoin dose was decreased, the symptoms began to resolve, and if this were purely SLE, one might not expect such a rapid improvement.

In order to diagnose SLE, the must be several criteria met and they include the following:

1) Malar rash
2) Discoid rash
3) Photosensitivity
4) Oral ulcers
5) Arthritis
6) Serositis- pleurisy or pericarditis
7) Renal disorder- persistent proteinuria >0.5g/24h or cellular casts
8) Neurological Disorder- seizures or psychosis (having excluded other causes first)
9) Haematologic disorders
  • Haemolytic anaemia or
  • Leucopaenia <4.0x109/litre>
  • Lymphopaenia <1.5x109/l>
  • Thrombocytopaenia
10) Immunological disorders-
  • Raised anti-DNA antibodies
  • Anti-Sm antibody
  • Positive anti-phospholipid antibodies
11) Raised titre of Anti-nuclear antibody

SLE can be diagnosed if 4 or more of the above criteria are met either simultaneously or serially over time.
(1982 revised criteria of the American Rheumatologu Association [now American College of Rheumatology] for the diagnosis of SLE)

From the above criteria, at most, the patient may have only had two criteria for suspecting SLE, those being the haematological disorders and the possible Malar rash. Cerebellar dysfunction although well recongised in SLE is not one of the diagnostic criteria. Unfortunately, antibodies for SLE were not performed in this patient.

Hence, it is still possible that phenytoin may have induced some features of SLE, and checking autoantibodies would be certainly worthwhile.

Finally, the haematological disorders in phenytoin toxicity are related to depleted folic acid and hence, checking this blood level and giving replenishment would be one way to correct the problem.

I think this paper from 25 years ago sums up what needs to be said:

1: Acta Neurol Scand Suppl. 1983;97:49-67.

Side effects of phenobarbital and phenytoin during long-term treatment of epilepsy.

Phenobarbital and phenytoin have good antiepileptic effect, but clinically significant untoward effects occur during their long-term use. Phenobarbital may cause hyperactivity, behavioral problems, sedation, and even dementia; these effects are dose related to some extent. Side effects of phenytoin include sedation, a cerebellar syndrome, phenytoin encephalopathy, psychosis, locomotor dysfunction, hyperkinesia, megaloblastic anemia, decreased serum folate level, decreased bone mineral content, liver disease, IgA deficiency, gingival hyperplasia, and a lupus-like hypersensitivity syndrome. Especially susceptible to the neurotoxic effects of phenytoin are epileptic children with severe brain damage who are on multiple drugs. In those children, balance disturbance may develop and be followed by gradual loss of locomotion. Among 131 mentally retarded epileptic patients, phenytoin intoxication occurred in 73 (56%), of whom 18 experienced persistent loss of locomotion. There is experimental evidence that the toxic action of phenytoin lies at the cellular level, predominantly in the cerebellum. Many experts avoid the long-term use of phenytoin because of its insidious and potentially dangerous side effects.

Thank you very much Dr Matsumura.

Another response was from Professor Lefor of Jichi Medical University who made an excellent attempt to answer this very difficult case. Professor Lefor’s way of working up this case is also very thorough and indeed, the diagnosis for this case does fall into that ‘other’ group. His reponse is below:

1.My problem list includes:
a. Unsteady gait
b. Diplopia x2y
c. Sensory changes
d. Slurred speech
e. Hx seizures in the distant past, on meds
f. PE showed no bruits, nystagmus, slurred speech, tremor, decr sensation and wide gait

2. What other questions?

I would want to know more about the time course of onset of these various symptoms.

3. Oral Exam: I am not sure what to look for

4. What is the diagnosis?

This is a tough case. There is a wide variety of disparate neurologic findings here. It is difficult to put it all together. The major etiologies would include vascular, infectious, tumor, and other. I doubt it is vascular given her age, physical findings and onset. Infectious is possible, but the time course does not support this. Tumor is certainly possible, but the wide variety of symptoms makes it hard to assign to a mass lesion. My number one is in “Other” category, and I believe that this is most consistent with some global neurologic disorder such as Multiple sclerosis which affects a wide area of the CNS as a de-myelinating disease. She is a little bit old for the usual patient (in the USA the onset is rarely over age 40), but this disease, or some other de-myelinating condition seems most probable given the wide range of structures affected as evidenced by her symptoms.

5. What is the one test: For MS, there is no one absolute test, but MRI is quite good and is the one test I would get. MRI would also rule out a brain tumor, which is also a possible dx, and must be fully ruled out quickly to avoid “missing it”.

Thanks for letting me participate

Alan Lefor

Comment from Dr B: Professor Lefor’s consideration of MS is very thought provoking especially as diplopia can be a symptom of MS. Moreover, the onset of cerebellar dysfunction can also occur in MS and neurological deficits occuring in various places over time is one feature of this disorder. Indeed, exacerbations of MS with acute deterioration can occur although this is usually related to an underlying infection which there is no history of in this case. Moreover, the generalised loss of peripheral tendon reflexes is not a usual presentation of MS. However, with an acute deterioration, one might expect the loss of reflexes e.g. in a single limb, and for them to return in time and become hyper-reflexic signifying a previous upper motor neurone lesion.

However, the pinky cheeks, coarse face, alopecia etc do not fit with MS or with a tumour and hence, a metabolic or autoimmune condition would be considerably higher up on the list. However, things such as tumours and MS should always be ruled out in such conditions as has been described above.

Thank you very much Prof Lefor.

Learning Points From This Case

If a detailed history is taken, if it is a common disorder, the cause will usually become obvious quickly.

Taking a thorough drug history, drug doses and frequency is very important. One must not just re-prescribe the drugs just because that is what the previous physician did or that is what the patient says they are taking. You must check the correct doses if you do not already know.

In this case, the drug doses and frequencies were purposely omitted in order to draw your attention to the fact that they were missing. They should never be over-looked or omitted from the history and young physicians should always be asking about drugs the patient takes. Drug doses and frequencies must be checked to avoid toxicity. This case demonstrates this example very clearly.

Checking drug-drug interactions is also essential. I would suggest obtaining a handbook on drugs that deals with this or perhaps one for PDA e.g. Epocrates ( [free]) or the Handbook of Adverse Drug Interactions (

Generally, as physicians we all have to know an awful lot about drugs but it is something that can cure or curse a patient but we must know when that happens. The only way is learn throughly about drugs and especially about commonly used ones in daily practise. One textbook that I used as a student was Rang & Dale's Pharmacology with an updated edition just last year. You can never know too much about a drug.

Common Things Are Common and 'When You Hear Hoofbeats, Don’t Expect To See A Zebra'~~(Theodore E. Woodward M.D) are very important ideas. Obviously, things such as drugs, alcohol, stroke and infection are high up on any list for cerebellar diseases. These should be excluded initially and if ruled out, investigation for other causes such as MS, SLE and even Friederich’s ataxia which lie somewhere further down the investigative road.

Usually, if you prioritise your differential diagnosis according to the history and physical plus specific lab and radiological tests, the answer will soon become apparent.

But remember this, the answer usually lies somewhere within the history but you have to elicit the right answers to find it.

Please consider and have a good week!

Finally, thank you to Dr T and his team for showing me such a great case.