- Squiggly red line : approx start of cavernous sinus/cavernous segment
- Dorsal (or inferior) ICA aneurysm -- not quite SHA, too distal, one would expect a SHA aneurysm to be more proximal. SHA aneurysms point medially and inferiorly
- There are some people who believe that more tortuous carotid siphon = increased likelihood of forming ICA aneurysms
- There is also a pericallosal aneurysm -- kind of weird anatomy. There are 2 pericallosal arteries and they come of ACA really late.

HYPOTHALAMIC LESIONS

All images from this paper: http://www.ncbi.nlm.nih.gov/pubmed/17620469

Embryology: anterior pit from rathke's pouch (from mouth), posterior + infundibulum from neuroectoderm

Lesions:

Craniopharyngioma: 

adamantinomatous - childhood variant, very cystic, very heterogenous 

papillary - adult variant, smaller, more homogenous 

Germinoma

young kids, can have pituitary/pineal synchronous lesions, homogenous, enhancing, non-cystic/heterogenous 

Hamartoma

older kids, tuber cinereum, can be pedunculated or sessile (more likely to be assoc with gelastic seizures & precocious puberty if latter). These do NOT enhance. 

Dermoid Cyst

 well-circumscribed, ectodermal cysts of fat/sebaceous tissue/hair/etc. Typically T1 bright and suppress on STIR, but not as bright and not as suppressy as lipomas. Typically occur in the midline although not commonly at suprasellar. rarely enhance or have calcifications

Vs epidermoid cysts, which have the MRI appearance of dense CSF, more commonly appear off-center - esp parasellar 

Vs arachnoid cysts which look just like CSF because they are 

Rathke's Cleft Cyst

 well circumscribed, midline, between A&P pituitary, typically do not enhance/have calcifications

Hypothalamic/Chiasmatic glioma: 

 very often NF1 associated. larger tumors = more heterogenous, enhancing. 

Ganglioglioma

 very rare, well circumscribed, can be cystic, slow growing. 

Encephalitis 

 usually viral, with clinical evidence of infection 

LCH 

LCH - has unexplained predilection for pituitary stalk and infundibulum. hypothalamic lesions enhance. 

Sarcoid

nodular thickening of chiasm, stalk, infundiblum from granulomatous involvement of dura. predilection for skull base -- neurosarcoid often manifests around hypothal/pit. meningeal enhancement. can invade virchow-robin spaces. 

Real life Pcomm aneurysm surgery

- Do not retract on the temporal lobe! these aneurysms can be buried in temporal lobe and retracting on temporal lobe significantly increases rupture rate
- If you place the clip on and it starts bleeding, do not remove the clip. Clip-on bleeding is usually small/slow and controllable with suction. Clip-off bleeding may be of the uncontrollable, hits-microscope-lens type
- After you place clip on, you can use the bovie on the aneurysm dome to shrink it down, possibly for purposes of better visualizing distal pcomm (which usually comes off ICA posteriorly) or anterior choroidal. If you bovie, make sure someone is irrigating on you so you do not stick to dome. Also, don't use those special spetzler bovies for this, they are way stronger than normal
- Do not bovie an unclipped aneurysm! 1. You are asking for rupture 2. the bovie only works because it makes the tissue shrink down, which will only actually shrink aneurysm dome if the pressure inside is decompressed from a clip.
- You can take the clip off and reposition if you need to
- You have to visualize anterior choroidal before you close to make sure its still patent. Have to.
- If the Pcomm is not fetal, you can sac it, but if you sac you have to take it right at the origin off the ICA, otherwise you risk incorporating perforators into the clip/compressing perforators with clip. Most famous are thalamoperforating arteries, but perforators off pcomm also supply optic tract/chiasm as well as hypothalamus, mammillary body
- If you can't visualize the origin of AChR well and don't want to put a clip on neck and risk taking it, one option is to place a long clip along midsection of aneurysm to shrink it down enough to see around it-- but you have to use a clip long enough to get all the way across the dome; a partial clipping will increase risk of rupture.
- Pcomm aneurysms do not typically require much dissection or retraction, as long as you open sylvian fissure you should not have to retract much on frontal lobe.

Random Neurorads

- Chordomas are almost always midline! A suspected tumor of the skull base that is exclusively lateral (i.e. condyle) is unlikely to be a chordoma. Chordomas also have "soap bubble" edges on scans (also physaliphorus cells on path) and enhance super bright with contrast, most commonly affect clivus/anterior occipital bone.
- A lateral skull base tumor that is erosive/lytic through bone, with popcorn calcifications, big time tissue edema - first on the differential is chondrosarcoma, which can originate from the cartilage of the formen lacerum or of the occipital condyle. These things also enhance like crazy on MRI. Other things: mets (esp breast cancer), chondroblastoma (tend to be younger patients, muted on T1/T2), weird stuff like langerhans cell histiocytosis, gout/CPPD, osteomyelitis.
- Petrous bone non-oncogenic pathology -- middle-ear ENT-type problems (cholesterol granuloma, cholesteatoma, middle ear effusions), aneurysms (petrous carotid), non-cancerous processes that affect the bone like Gout/CPPD or osteomyelitis (esp older, diabetic/immunocompromised people) -- petrous apicitis can present as Gradenigo syndrome, which is 6th nerve palsy from dorello's canal, facial or retro-orbital pain from meckle's cave.

Lymphoma on imaging:
- CT: looks like a solid, highly cellular/dense mass
- MRI: perventricular, more or less homogenously enhancing, lots of edema. However lymphoma typically does not push significantly into ventricles - a tumor that is mostly inside the ventricle, think more like ependymoma (can invade brain), CPC (CPP typically does not invade brain)
- Except for HIV-associated CNS lymphoma which is ring-enhancing and primarily T-cell based

Giant, gross looking tumor with minimal enhancement is not lymphoma. It is GBM or mets until proven otherwise. One thing that may distinguish is CBV -- edema around mets is purely vasogenic and so will have low CBV, while the edema around GBM is often partially or wholly infiltrative tumor and so will have high CBV

Another thing about GBM - typically tracks along the white matter, really likes the thalamus, heterogenous, non-restricting.

Things that restrict on DWI - A lot more than just strokes!
- Very cellular tumors - "small round blue cell tumors" like PNETs (ie medulloblastoma, pineoblastoma), and other cellular tumors like lymphoma, meningoma, chordoma, germinoma, or the solid/cellular parts of GBM (note: GBMs typically do not restrict but they can, so do not use lack of diffusion restriction to rule out a GBM. GBM can look like anything)
- Infection - empyema, abscess
- Epidermoid cysts (vs arachnoid cysts, which do not) - speaking of skull base lesions, bad epidermoids can become giant tumors along the petrous bone in the posterior fossa, extending all the way up into the midbrain.
- Cytotoxic edema/ischemia
- Weird poisoning syndromes - carbon monoxide, methanol, wernicke's, hyperammonemia
- Weird inhered stuff - X-linked adrenoleukodystrophy, Maple syrup urine, canavan
- Weird prion disease - kuru and CJD "cortical ribboning"
- Weird other weird stuff - CADASIL, wilson's

CMV encephalitis usually involves periventricular T2 hyperintensity. Usu CD4<50

Thought process for differential diagnosis in HIV CNS imaging
- Space occupying - toxo, cns lymphoma
- Non space occupying - PML, HIV encephalitis (usually symmetric), CMV encephalitis

MS usually causes incomplete rings of enhancement!

Multiple enhancing brain lesions with extremely high CSF pressures (put in EVD, CSF hits the ceiling) - think cryptococcus. The buzzword is"perivascular gelatinous pseudocyst"

Oligodendroglioma - on the boards, will classically be frontal, subcortical, heavily calcified (90% are calcified).

Acute management of Afib/flutter with RVR for dummies 


1. Are they hemodynamically stable? 
--> no --> synchronized electrical cardioversion (unsync'd may cause the rhythm to degenerate into vfib). You proooooobably should call MICU or cards if you're gonna do this. You know the whole, coding-someone-by-yourself-is-a-bad-idea thing. 
--> yes --> go to step 2. 

2. Is there any evidence that organs are being underperfused? 
--> yes --> be way more aggressive -- start with 5mg metop x3, and then go straight to esmolol gtt or dilt IV push x2 followed by dilt gtt if the BP Is good, or straight to amio load/gtt if BP Is not. Also, call for an ICU bed... few places allow this kind of thing to happen on the floor. 
--> no --> can let people hang out -- start with 5mg metop pushes, or maybe even just PO drugs depending on how fast the rate is and how healthy they are. 
which agent you pick depends on the BP and whether they have HF and what agents they are already on --> go to step 3/4/5 

3. Is the blood pressure good?
--> yes --> b-blocker preferred, then CCB 
--> no --> amio preferred, then dig (also give a bolus, unless HF)

4. Do they have horrible, decompensated HF? 
--> yes --> dig preferred, then amio -- amio is way faster, but dig has inotropic effect... depends on how sick/fast they are vs how bad their HF is. 
--> no --> beta blocker preferred, then CCB 

5. Are they already on something? 
--> yes --> increase the dose of that before you add additional agents. For example, if you're at 200mg/day of PO metop (50 q6) or 360 mg/day of PO dig (90 q6) and it's still not working, then add something else. 
--> no --> select first agent based on below system: 

Non heart failure patient with good blood pressure who isn't going that fast: 
IV metop first, then if that fails add PO metop, then if that fails add PO dilt 

Non heart failure patient with good blood pressure who is going fast: 
IV metop first, then if that fails IV dilt push, then if that fails IV dilt gtt, if that fails then amio. If at any point, the BP starts to fail to tolerate all the drugs, switch to amio earlier. 

Non heart failure patient with suboptimal blood pressure:
Fluid bolus first, if that fixes the BP then go with the above pathway, if that doesn't fix BP then go the below pathway. 

Non heart failure patient with crappy blood pressure: 
Fluid bolus then amio (IV load and gtt, or IV load and PO maintenance, depending on how fast they are and how urgently you need control). If that fails call MICU. 

Decompensated heart failure patient who isn't going really fast and has OK kidneys: 
Dig first, then if that fails, add amio. 

Decomp heart failure, who is going fast and/or has bad kidneys: 
Amio first, then if that fails, add dig (if kidneys are OK)  

Also call cards... most of us probably shouldn't be managing afib/flutter with rvr in patients with decompensated heart failure... 

About the options: 

- IV Metop. Given as pushes of 5mg each, up to 3 times with a 5 minute wait between each one. It's fast, but there's no drip metop; so if it only transiently works then you have to switch to an agent that has a drip. If you think there's high chance of success that a few IV pushes of beta-blockade will make them better (i.e. you know the patient, metop has worked in the past, etc) then this is fine. Otherwise start with dilt or esmolol. Of note, the patient will need to go home on PO metoprolol and may need to remain on it indefinitely.
- PO metop. If IV metop works, then start a PO metop regimen - in b-blocker naive, start at 12.5 q6; can titrate up to 50q6. In a stable person who is not-that-fast and chillin' comfortably you can even just start with PO metop and titrate up as needed.
- Esmolol gtt: great drug, works really well, but not an option for non-ICU patients - half life 9-12 minutes means you need constant titration means there is no way the floor nurses will let you do this. In fact, even saying the word esmolol will cause the charge nurse to suddenly appear, bearing down on you with an angry expression, ready to bash your skull in with their ever-present clipboards. Try it. It's like a magic trick. 
- IV dilt: Push 0.25mg/kg. Wait 15 minutes. If it didn't work, push 0.35 mg/kg. Wait 15 minutes. If that still doesn't work, start drip at 5-15mg/hr. Preferred drug for patients you don't know (may have atrial thrombus) and for people with tough to break a-fib who you think may need a drip.
- PO dilt: If you're gonna pick an agent to start, you're better off starting with PO beta-blockade than PO CCB esp in older people or people who you think might have suboptimal heart function (i.e. older, fat, HTN, DM, PAD, smoker, etc etc etc) -- CCB have more negative inotropy. If someone is already on dilt and it has worked for them you can keep pushing it up to 120q6. 
- Amio: 150mg IV push, then 60mg/hr for 6 hours, then 30/hr for 12-18 hours depending on where you are. This is a great drug-- it works fast, it works well, it doesn't drop the BP (well, actually it does, but to a way smaller extent than CCB or BB), it doesn't have negative inotropy. The dangerous thing a is that it's a class III antiarrhythmic, and you can inadvertently cardiovert, and if they have an atrial thrombus you may cause an embolic stroke. Although you can inadvertently cardiovert with metop as well... this drug also has tons of long term side effects so its' not as a great of a choice for the long term, but if you're just trying to stabilize someone who is going fast and has terrible BP until someone who knows more about hearts than you comes along, its a fantastic choice. 
- Dig: slow, but has positive inotropy. 1 hour onset time. At 6 hours, a significant number of people will still be tachycardic. Second line in heart failure patients who are unable to be rate controlled with amio. Don't use in people with renal failure!! 

BP control and IPH recurrence 

N=1145
Study type: semi-prospective cohort.
Years: 1994-2011

Inclusion criteria:
- Age > 18
- Supratentorial IPH
- Alive after 90 days

Exclusion criteria:
- IPH from vascular malformation/aneurysm, tumor, trauma, or hemorrhagic transformation of ischemic stroke

Methodology:
- Study was separated into lobar and non-lobar IPH
- Follow up was conducted at 3, 6, 9, 12 mos and every 6 mos thereafter - inquiring about death, evidence of ICH recurrence, recorded BPs, medications.

Outcome:
- Recurrence of IPH

Results
- Overall recurrence rate of lobar IPH was ~20% and of non-lobar was ~7%

Comment:
- In multivariate regression, use of antiplatelet assoc with significant increased risk of ICH; warfarin - trending - likely underpowered - but big hazard ratio
- In nonlobar ICH - the relationships between drugs and recurrent ICH is less clear - both in magnitude and in statistical significance. This might have to do with statistical power, as non-lobar was much less common.

Comment:
In case you've forgotten your JNC 7 guidelines
- Normotension: SBP <120 and DBP <80
- Prehypertension: SBP 120-140 or DBP 80-90
- Stage 1: SBP 140-160 or DBP 90-100
- Stage 2: SBP >160 or DBP > 100
note that normotension carries an "and" operator for SBP and DBP whereas the others carry an "or" operator

This is an incredible dose-response relationship between even mild HTN (ie >120) and recurrence; HR for just a SBP of 120-140 was 2.8 in lobar and 3.0 in non-lobar. Man.

JNC 8 recommended SBP goal of <150 (elevated from <140) in older adults (>60) without DM or CKD; the idea is that if you are too aggressive with your goal, you overshoot too low and old people have falls and leads to worse outcomes.
But perhaps in older adults with a history of IPH, we should be more aggressive...

PARKINSONS TREATMENTS

Supplements
- Vitamin E - not shown to have any benefit in a large trial
- Coenzyme Q10 - not shown to have any benefit in large trial
- Exercise - actually works

MAO-B inhibitors
- Selegiline - helps with some motor symptoms
- Rasagaline - may slightly slow progression of disease (ADAGIO trial)
- Rasagaline might be a little more effective than selegiline but its way more $$$
- This class of drugs is relatively ineffective but relatively benign; good to start with
- The whole wine-and-cheese tyramine thing is a bit oversold, you can probably eat tyramine containing food as long as you don't go overboard

Anticholinergics 
- Triheyxyphenyldil  (Artane): most effective for tremor. Causes urinary retention and all the other anticholinergic things. Causes significant confusion/cognitive fog -- which may be a dealbreaking side effect for people in school/taking classes, with jobs that take a lot of thinking, or in old people who are already pretty confused.

Dopamine Agonists 
- Ropinirole (Requip)
- Pramipexole (Mirapex)
- Come in TID or extended release formulations (reallllly $$$)
- Rotigatine - patch - q24h
- Generally first-line in people under 65 - as they can delay the onset of dyskinesias. In people over 70, these drugs are both more likely to cause side effects (like confusion), and you're less worried about the progression of the disease over time.
- Side effects - Fatigue/somnolence, sleep attacks (seriously, people fall asleep at the wheel and get into accidents), nausea/vomiting, confusion, leg edema, postural hypotension, hallucinations

Sinemet 
- Levodopa is the most effective drug we have in the management of symptoms - and most people will end up on it eventually
- Carbidopa is a DOPA decarboxylase inhibitor
- Levodopa crosses the BBB and is converted to dopamine in the blood - dopamine itself doesn't cross BBB. Carbidopa prevents conversion of levodopa to dopamine in periphery. Carbidopa also doesn't cross BBB.
- The old school "controlled release" sinemet (sinemet CR) was crappy - depended on acid degradation in the stomach, and on gastric transit and motility, so the blood levels fluctuated like crazy and were really unreliable. Sometimes now we use it at night, where it gives nighttime symptom relief or AM effects.
- There's a new extended release sinemet called Rytany that depends on differential sized beads and is much more reliable in terms of drug levels. Unfortunately its really, realllllly, realllllllllly expensive, and you have to take a lot of pills (ie like 3-4 pills TID or QID - the median is 3.6 times a day)
- However regular sinemet some people have to dose q2 hours, or q3 hours, which is just agonizing, and going from q2 to q5 or q6 dosing can be life-altering in a good way

COMT inhibitors 
- Entacapone
- Tolcapone - more potent because better CNS penetration, but more reported cases of severe liver failure-possibly a fluke in early trials as the reported incidence is dropping with time; have to get monthly LFTs for the first 6 months
- In the late game, they decrease off time when given with sinemet
- Only given in combination with levidopa, not given alone-- they don't have much effect on their own. Also they result in disease with worse dyskinesias if given early, alone or in combo with sinemet.

Apomorphine
- Injection, subQ, need to premedicate with antiemetic, lasts 30-60 min, can bridge - i.e. take your sinemet and the injection and by the time injection wears off, pills have taken effect.

Duodopa 
- gel delivered into the intestines, 89% rate of pump complications/malfunction/failure

DBS 
- Usually can reduce the dose of meds, but still need meds


Treating Side Effects

Dyskinesias
- treatment: amantadine is very good for this. Doesn't affect PD itself.
- can decrease dose of sinemet

Freezing
- Can increase sinemet

Halluciinations 
- stop Dopa agonist, MAOs, anticholinergics, amantadine
- decrease sinemet
- add seroquel
- add cholinesterase inhibitor

To be avoided in PD: anti dopamine drugs
- Neuroleptics - haldol, etc
- Antiemetics - promethazine, metoclopramide, etc