Peri-Tonsillar Abscess

Thanks to Max W for presenting a case of a peritonsilar abscess in a patient with a hx of organ transplant who was found to have necrotizing HSV tonsillitis requiring incubation for subglottic/epiglottic edema

Peri-Tonsillar abscess in immunocompetent patients 

(bread and butter)


Screen Shot 2017-05-15 at 9.45.37 AM
This diagnosis can be made clinically based on suggestive sx – hot potato voice, pain with swallowing, 2/3 have truisms – and an exam with uvular deviation and a swollen tonsil.


  • Annual incidence is 30/100,000 people with a peak onset during adolescence
    • By contrast, appendicitis has an incidence of 233/100,000 people. Common enough that high volume emergency rooms see it on the regular but not as common as pneumonia.
  •  Risk factors include smoking, recurrent tonsillitis, and oropharyngeal cancer or radiation.


  • Often progresses from exudative tonsillitis->tonsillar cellulitis->abscess
  • Usually bacterial and polymicrobial with an over-representation of GAS, s milleri and anaerobes esp fusobacterium.
  • Always consider the ddx of other deep neck space infections from mouth to mediastinum
    • ludwig’s angina
    • epiglottitis
    • septic IJ thrombophlebitis (Lemierre’s disease)
    • retropharyngeal abscess
    • tracheal inflammation or infection


  • Evaluate for impending airway compromise
    • Difficulty managing secretions, voice changes, dyspnea. Stridor is a late sign
    • Consider direct laryngoscopy if any concern.
  • Aspirate that pus!
Screen Shot 2017-05-15 at 10.12.00 AM
  • We rarely admit paratonsilar abscesses. Usually ENT/the ED aspirates them, they get better and go home.
    • Ultrasound can increase the yield of aspiration.
    • If no pus is aspirated or symptoms do not improve, they should be admitted
  • Antibiotics
    • amox/clav and clinda are great PO options.
    • Consider MRSA like you would in any other cellulitis. If so, they may need admission/IV vanc
  • Steroids
    • Small studies  show evidence for a single dose of 10mg of dexamethasone to speed recovery.

PTA in severely immunocompromised patients

(ancient grains with caviar)

Main difference is in causative agents. Also consider…

  • pseudomonas, MRSA, HSV (case-reportably rare, that’s what this patient had), candida (usually more slowly progressive.)
    • initial abx management is often broad – don’t be afraid to call ID! This patient was started on vanc/piptazo/acyclovir.
  • If pus cannot be aspirated, consider biopsy to guide dx therapy
  • HSV tonsillitis improves with acyclovir.
References (and source of images)
Galioto NJ. Peritonsillar Abscess Am Fam Physician. 2017 Apr 15;95(8):501-506.
Hirzel C1, Nueesch S2, Wendland T3, Langer R4. Necrotizing herpes-simplex virus tonsillitis mimicking peritonsillar abscess. Infection. 2016 Apr;44(2):267-8.

Filed under: Morning Report

Diagnosing Hyponatremia!

Report this morning at ZSFG was inspiring. The case was super interesting – a woman who presented with hyponatremia which was found to be 2/2 LONG undiagnosed (like 30 year) hypo pit from postpartum pituitary hemorrhage.
Chief confession time
I think hyponatremia is really hard. In report today, a room full of super wise residents, medical students and Lisa Murphy helped develop a clear, comprehensible hyponatremia algorithm. Here is my deep, philosophical conclusion, that will also help you use this algorithm
  • Most diagnoses are simple with history, exam, and 3 lab tests. A small portion are wicked hard to diagnose.
    • Simple: this patient has been vomiting for a week is on a thiazide. The diagnosis is diuretics + hypovolemia. The patient has pneumonia and has been drinking plenty of water. The diagnosis is SIADH. In these situations, you can use the algorithm below to make confirm your diagnosis and ensure nothing else is going on.
    • Wicked hard: The numbers don’t fall into any of the above buckets, are in a grey zone, the patient is obtunded or you thought you knew the answer but the patient isn’t getting better.
    • GI bleeds are this way too! Most of them are clear from history and basic physical exam. Then there’s the occasional small bowel angioectasia that has you banging your head against the wall
  • Volume status
    • While basing your hyponatremia algorithm on volume status is laudable, in reality, assessing volume status is hard! So this trips people up a lot.
    • See below for a volume status-free algorithm.
Simplified hyponatremia algorithm
  • adapted from Miliones HJ, Liamis GL, Elisa MS. The hyponatremic patient: a systematic approach to laboratory diagnosis. CMAJ 2002; 166 (8): 1056-62. Chart made with Lucidchart.
  • A caveat: none of these branch points are true 100% of the time. The lower down you get on the algorithm, the less accurate it is, like the Brugada criteria. Uric acid levels are particularly iffy – they are most sensitive for SIADH when they are <4
  • The uric acid pearl, is one of VA LT’s favorites. ADH promotes Urea/uric acid excretion along with water retention.
hyponatremia! - Page 1-6
Tips for wicked hard hyponatremia
  • The algorithm in the Miliones article above which has more detail, explains what to do with intermediate values, and gives some specific suggestions for testing
  • Make a hypothesis, try a treatment, and recheck the big 4 – serum sodium, urine sodium, serum osms, urine osms.
  • Renal and endocrine can help you if you’re lost!
Managing hyponatremia is super important and pearls for another time. Search the archives for tons and tons of other hyponatremia pearls.

Filed under: Morning Report

Ramsay-Hunt Syndrome

Thanks to Adam Tabbaa and Megan Lockwood for presenting a 40 y/o healthy man who presented with multiple cranial neuropathies and ear vesicles concerning for Ramsey Hunt Syndrome
What is Ramsey Hunt syndrome?
  • classically, Facial nerve paralysis + Zoster oticus (zoster lesions in the ear) +/- hearing loss
  • Pathophysiologically – Zoster reactivation in the geniculate ganglion.
This case hinged upon identifying the patient’s cranial neuropathies and trying to identify an atomic location whether those cranial nerves ran together. So here’s a quick refresher on the facial nerve and the geniculate ganglion.
facial nerve 2
  • The facial nerve exits the brain stem, courses through the temporal bone, then forms a knee-shaped ganglion at its exit

facial nerve 1.png

  • As this less detailed image shows. The facial nerve runs *extremely close* to the ear when exits the temporal bone. So when VZV reactivates in the geniculate ganglion, it travels transaxonally through the nerves to skin around the ear, causing vesicles there.
  • When the facial nerve runs next to the ear, it also runs in close quarters with the vestibulocochlear nerve. So Ramsey-Hunt can also include deafness or vestibular symptoms and for this reason is considered an ENT emergency.
  • Ramsey Hunt is usually managed with acyclovir + steroids. A cochrane review showed no evidence of benefit for acyclovir, but the overall evidence quality was poor.
Multiple Cranial Neuropathies
  • Way more complicated – I’ll write pearls about this some time soon.
Whitley RJ1.A 70-year-old woman with shingles: review of herpes zoster. JAMA. 2009 Jul 1;302(1):73-80. doi: 10.1001/jama.2009.822. Epub 2009 Jun 2.
Rebecca Miller-Kuhlmann, MD. My very favorite neurologist who is tolerant of my many, many questions.

Filed under: Morning Report