Guide to the TV Show


Holmes/Homes/HouseShow InfoResourcesSitemapGlossaryBlogs/ AnswersAboutHome

House MD Guide :: Blogs ::  Differential Diagnosis:
A Medical Guide to House

Wednesday, May 09, 2007

The One Clue

As a big fan of mystery novels and short stories, I thought I would go to the library and check out one that I hadn't read. Mind you, this eliminates all of the Sherlock Holmes stories and some of Agatha Christie's works. Incidentally, I happened to discover that one of my favorite authors, Isaac Asimov, had written a series of mystery stories (which I never knew), and he had put together a series called The Black Widowers. I've slowly been reading the short stories and keeping up with House, when it suddenly hit me that there's something about the recent few episodes of House that just has struck me with annoyance. I couldn't initially put my finger on it, but reading The Black Widowers brought it from the tip of my tongue onto the keypad of my laptop. Knowing that most of you have never heard of The Black Widowers (as I hadn't), I'll give you a quick synopsis of a typical story. Bear with me. I'll get to the point soon enough.

1. Five guys, a waiter, and a guest meet up once a week to have a discussion about some obscure topic (intricacies of Shakespeare, limericks based on the Iliad). At some point, the discussion becomes heated as the characters try to prove why they're right.
2. At the climax of the argument, the guest usually brings up a problem that he is having (generally a work-related mystery), and then this leads the gang to solve the case.
3. All of the five upper-class professional men grill the guest with questions, and don't come up with a solution.
4. The waiter offers the solution at the end, explaining how the solution ties back to the initial argument that they were having before the case was brought up. The solution is usually absurd.

This isn't all that different from a typical House episode in the 3rd season (with some notable exceptions such as "Airborne," which pleasantly broke this mold). Just replace the five upper-class professionals with Chase, Cameron, and Foreman. Make House the waiter. And instead of obscure topic, substitute strange plot theme (House doping Wilson up with amphetamines, Chase asking out Cameron). Instead of a guest, make it a patient. And instead of grilling with questions, make it medical tests. And for the sake of blandness, at the end of each episode, make the final diagnostic question "autoimmune vs. infection," and leave it to the waiter to figure out which it is.

A good mystery writer is able to present clues without cheating. For example, in the classic mystery story where someone is killed and only 4 people could have done it, a good writer telling the story plants one clue that if caught by the reader could present him or her with the solution to the mystery. This is what is what is engrossing about mystery-based fictional works of all mediums. You try to outsmart the detective, fall a step short, and admire the detective's ability to present a logical explanation for how he or she arrived at the correct conclusion when you failed, given the same data.
During some episodes of House, the writers fire point blank and hit the mark. A patient has symptoms. I immediately begin thinking of some diseases that could explain the symptoms. All of these diseases are methodically ruled out. Just before the end of the episode, House comes up with an off-the-wall diagnosis, defends it with an explanation, and the patient is saved. These episodes, in my mind, are the absolute best. Some of the most memorable scenes of the show come from moments when House has his big epiphany and explains it to the patient. Remember the scene from "Mob Rules" where House calmly explains to a mobster that only one drug could raise a person's estrogen level... estrogen**! Remember the scene from "Fidelity" where a woman sleeps continuously for hours, prompting House to tell her husband that she must have African sleeping sickness, contracted from a sexual encounter with a man who recently visited Africa?! That was a moment of beauty.
Sherlock Holmes, trying to explain a seemingly impossible murder during The Sign of Four, says, "How often have I said to you that when you have eliminated the impossible, whatever remains, however improbable, must be the truth?"
Yet as the 3rd season progresses, House is breaking off from the Sherlock Holmes mold and entering a stage of poor mystery writing. The most recent episode titled "Resignation" is one of the biggest offenders when it comes to breaking the model of good mystery story-telling as I have presented it. Addie, a 19-year-old patient, coughs up blood. House says it's an infection. Chase says it's autoimmune. The tests can't locate an infection. The tests can't locate an autoimmune process. Chase puts the patient on steroids. Patient crashes. House has a moral victory. House is happy. House is too happy. House has an epiphany. He must be on antidepressents. Therefore (and I say "therefore" with some sarcasm given that the two events are not related), the patient must be depressed and poisoning herself. Roll credits.

Where is the one clue? There should be some event, however fleeting, in the episode that a careful viewer could pick out that would suggest that Addie is poisoning herself from depression. It's almost as if the writers are so insanely paranoid that the viewer might arrive at the correct diagnosis that they are intentionally leaving out any trace of a clue that could make an attentive medical student stumble upon the diagnosis. This is not a problem limited to this episode, but it is one that is more prevalent during season 3 than during earlier seasons. Even when the one clue is present during the 3rd season, it's presented after the fact. Think back to the episode "House Training," where Foreman treats a woman with radiation for the ever-ubiquitous "auto-immune" process, only to reveal during an autopsy in the last seconds of the show that her brastrap had torn her skin and opened the way for an infection.

Even though Asimov does a terrible job of actually constructing the mysteries involving The Black Widowers, he does an excellent job in the Introduction section of describing the components that make a good mystery story. I know Asimov is capable of good mystery writing because I've read his I, Robot series of short stories (of fame thanks to Will Smith), which do a great job incorporating the elements of good mystery storytelling in the setting of robotics. Likewise, I think that the writers of House can do better because they've demonstrated the ability to do so in previous seasons. Two more episodes left this season.

I'll be watching for the one clue. Maybe you'll spot it.

** This is actually not true, as shown in the episode of House titled "Act Your Age," where a 6-year-old girl begins having periods as a result of converting testosterone received from her father's skin cream to estrogen, facilitated by the aromatase enzyme.

Saturday, April 07, 2007

House Wannabe and the Case of the Real McCoy

Foreman: "Looks like they got the pheo out successfully. So. what now?"
House: "Clarence goes back to death row."
Foreman: "Just like that?"
House: "He's cured."
Foreman: "That tumor caused random shots of adrenaline, which obviously led to the rage attacks that made him become a murderer in the first place."
House: "By God, you're right! Let's call the surgeons. We gotta save that tumor; put it on the witness stand."
Foreman: "We could testify at Clarence's appeal."
House: "You smell that? I think that is the stink of hypocrisy. You wouldn't even consider the notion that Clarence's social upbringing was responsible for what he became, but now you're sprinting to the witness stand to blame everything on a little tumor."
Foreman: "A person's upbringing and their biology are completely different."
House: "Yeah. Because you only overcame one of them."
If you think back to Season 2, Episode 1, you might remember the story of Clarence, played by LL Cool J. Clarence was an inmate who had been put in jail on the charges of multiple counts of murder. Just prior to being brought into the Princeton-Plainsboro hospital, he was experiencing an episode of uncontrolled rage and hallucinations. During the course of the show, House discovered that Clarence suffered from a pheochromocytoma, a tumor of the adrenal gland that oversecretes adrenaline. Too much adrenaline in the body typically causes episodes of headaches, high blood pressure, sweating, and heart palpitations. House figured that in Clarence's case, the extra adrenaline was making him have episodes of rage. Foreman took this argument even further and claimed that perhaps Clarence should be acquitted for his murders because they may have been conducted during one of his raging episodes. The episode ends with Foreman saying that he planned to testify in Clarence's trial.

This argument at best spurious. I agree that it makes sense on a certain primitive level that excess adrenaline should cause a person to be violent. But that's it. There is no data anywhere that suggests that a pheochromocytoma can actually cause a person to be violent. In fact, even in forensic medicine journals, pheochromocytomas are acknowledged primarily for their ability to cause sudden death and panic attacks. As people who've suffered from panic attacks can tell you, the last thing a person who is having a panic attack is interested in is killing someone else.

So consider my surprise when I saw in the Washington Post today that "the real McCoys," renouned for their violent feuds with the Hatfields, have a hereditary disease that results in pheochromocytomas. Von Hippel-Lindau disease is apparently the cause of the violent history shared by members of the McCoy and Hatfield families, according to Dr. Revi Mathew at Vanderbilt University. Dr. Mathew says: "This condition can certainly make anybody short-tempered, and if they are prone because of their personality, it can add fuel to the fire."

An interesting statement, even if it is at least somewhat incorrect. Maybe he saw it on House?

Tuesday, April 03, 2007

House: Great Diagnostician?

One of the premises of House as a TV show is that Dr. House is a diagnostic genius. As a character, House's drug-seeking behavior and terrible bedside manner is forgiveable only because he is able to traverse through obscure symptoms and come up with a unifying diagnosis. As one of my attending physicians told me, there are two ways of approaching an arbitrary set of symptoms: Occam's razor and Hickam's dictum. For medical purposes, Occam's razor states that a patient with two complaints (ie. headache and fever) is much more likely to have one disease entity that is causing both complaints as opposed to two different disease entities. Hickam's dictum, on the other hand, says that a patient with two complaints is more likely to have different causes for each symptom as opposed to one unifying disease process.

Historically, physicians whose names are revered today are those who applied Occam's razor to identify new diseases. There's Charcot's triad for cholangitis, Cushing's triad for increased intracranial pressure, Wernicke's triad of thiamine deficiency, and so on. An exceptional physician is someone who is able to decide when to apply Occam's razor and when to apply Hickam's dictum. There is actually an example of a misapplication of Occam's razor which is still mentioned in textbooks, known as Saint's triad, named after C.F.M. Saint, a South African surgeon (not to be confused with Dr. Sanjay Saint, who published a New England Journal of Medicine article about Saint's triad).
The legend goes that Dr. C.F.M. Saint saw a number of patients with diverticulosis, gallstones, and a hiatal hernia, and decided that there must be an underlying disease that explains all 3 of those symptoms. He did something that doctors love to do and attached his name to this series. As it turns out, diverticulosis, gallstones, and hiatial hernias are all fairly common medical conditions, and they're much more likely to occur due to their own etiologies than to a shared etiology.
One of the reasons that House appears to be such an exceptional diagnostician is that he is able to correctly apply Occam's razor and Hickam's dictum to the nth degree to diagnose even the most obscure of diseases.

But consider how House comes up with his solutions. A patient comes in with a vague complaint. This complaint leads to another complaint, which then leads to a seizure, allergic reaction, or organ failure resulting in the patient requiring intubation. Is this really top-notch care? To ask this question in a different way, what would take place in an "ordinary" top tier hospital?

Order of steps in on an ordinary medicine service:
1. Patient is seen in ER and sent up to the Medicine floor
2. Patient is evaluated by resident (and possibly a medical student)
3. Resident presents the patient's history, physical exam, and labs to the attending physician
4. Attending physician repeats pertinent aspects of history and physical exam and suggests changes to the resident's plan

In this case, substitute the word "resident" for "fellow" because Foreman, Chase, and Cameron are done with their residencies and are now working on their fellowships. As evident on House's team, things work a bit differently.

Order of steps (on House's team):
1. Patient is transferred to House's team from who knows where
2. One of the fellows (Chase/Cameron/Foreman) takes a history
3. One of the others does a physical exam
4. House comes up with a differential diagnosis consisting of 10 things and orders every single test needed to sort out which of the possibilities is actually present

House's differential diagnosis is typically fairly extensive, and in this regard, I think that House is admirably intelligent. It takes experience to be able to name common causes of a spectrum of symptoms, but many of the things that House spouts off (and even a lot of what Cameron says on the show nowadays) reflect a lot of time spent with textbooks (or UpToDate). My primary critique of House's techniques is the same one that was pointed out by Vogler in season 1. Every patient with an isolated symptom doesn't need a complete work-up. Financially (and in some sense medically), it doesn't make sense to work a patient up for an autoimmune disorder until a more likely cause has been ruled out. I think physicians have an inner longing to be able to just order every possible test all at once, but this can be irresponsible, depending on the medical setting. A simple example of this is the difference between telling your family medicine doctor that you're having a headache and telling an ER physician the same. In the first case, you're likely to be treated for migraines, whereas in the second case, you're likely headed to the CT scanner for a head CT.

I think that the fact that House orders an MRI on just about every patient is acceptable only in the context of the fact that his patients are pre-selected to be complicated.

Though House's diagnostic acumen is flaunted in his inpatient cases, I think his real skill is more evident in general medicine clinic, where he diagnoses patients correctly without imprudently asking for unnecessary tests. This is especially impressive given the fact that House is boarded in 2 subspecialties and somewhat far-removed from primary care. There is a joke in medicine that goes something like this:
An internist, a pathologist, and a family physician go duck hunting. They see an animal that resembles a duck. The internist says, "Let me run some tests to prove that it's not a goose or a rabbit and only then will I proceed to shoot it." The pathologist says, "I'll kill it now and then figure out what it is." The family physician says, "I'm not quite sure what it is, and I don't really care. I have a gun and I'm killing it."
What makes a good diagnostician is the ability to maintain perspective despite maintaining a wide knowledge base. Sometimes, a cough is hereditary angioedema caused by C1 esterase inhibitor deficiency. Other times, a cough is just a cough.

Tuesday, November 28, 2006

A Simple Case of Vampirism

"Watson, if it should ever strike you that I am getting a little over-confident in my powers, or giving less pains to a case than it deserves, kindly whisper 'Norbury' in my ear, and I shall be infinitely obliged to you." - Sherlock Holmes, The Yellow Face
The patient: Alice is a 6-year-old girl who experiences problems with vision on a carnival ride and ends up in the hospital with abdominal pain, apparently due to pancreatitis. She has a low-grade fever and mild anemia. Her bile duct is dilated on a CT scan, indicative of stones in the bile duct (choledocholithiasis). The stones weren't visualized, so House states that the stones likely passed already and recommends an ultrasound to see if there are any stones left in the gallbladder. Gallstones are visualized in the gallbladder, and House recommends removal of the gallbladder to figure out what the cause is. After some court drama, the patient has her gallbladder removed (cholecystectomy). Following the surgery, Alice complains about pain from her stitches and Foreman notes a bright red vesicular rash with areas of denuded skin. She has no history of allergies. The stones are noted to contain calcium and bilirubin with a pigmented appearance. Because the patient presented with a low-grade fever and mild anemia, House suggests a bacterial infection. Foreman argues that bacterial infections do not cause a vesicular rash. House suggests a scratch test to rule out allergies. The scratch test is positive for all antigens. House is still convinced that the cause is bacterial and gives the patient a peanut butter sandwich to prove that she is not allergic to everything. House wants to start broad-spectrum antibiotics, but because of some more court drama, Cuddy is appointed guardian of the child and decides to give metronidazole instead. The patient becomes hypertensive and tachycardic. After she is stabilized, Alice's father takes her and leaves the hospital but returns promptly when the Alice becomes stiff. House suspects that the patient's babysitter gave her an aspirrin (leading to Reye's syndrome) and tells Cuddy to put the patient on charcoal hemoperfusion. In the middle of the procedure, Alice starts experiencing extreme pain and paralysis in her left hand. She starts to become hyperthermic and thrombocytopenic. The vesicular rash spreads to her left arm. Her fever increases to 103. House suggests that the patient may have varicella based on its appearance but this doesn't fit because the patient has paralysis and no itching. Cameron suggests Rocky Mountain Spotted Fever. House says to start the patient on chloramphenicol for Rocky Mountain Spotted Fever. The rash spreads throughout her arms and legs, so House diagnoses her with necrotizing fasciitis and recommends bilateral arm and leg amputation. As the patient is on the operating table, Chase has a realization that the patient may have erythropoetic protoporphyria and runs down to tell House. Someone calls the surgeon and the surgery is cancelled.

The diagnosis: Erythropoetic protoporphyria

Working backwards: When Chase made the diagnosis of erythropoetic protoporphyria, he did so working on the assumption that the patient's allergic tendencies could be attributed to light exposure. Based on the Harrison's textbook, the first two lab studies to check when a patient is sensitive to light is a plasma porphyrin (to rule out porphyrias) and an ANA/Ro/La (to rule out lupus and other similar autoimmune processes). Even though Chase didn't order either of these studies, it makes sense that he presumed that one of these two diseases was leading to the patient's symptoms once he made the connection to the light. Had he ordered the tests, he would have found increased levels of plasma porphyrin. Some of the common porphyrias are porphyria cutanea tarda (#1), erythropotoetic protoporphyria (#2), and acute intermittent porphyria. If you tested the urine for porphyrins, porphyria cutanea tarda would show a normal porphobilinogen level and an increased uroporphyrin and 7-carboxylate level; erythropoetic protoporphyria would show normal porphyrin and porphyrin precursor levels; and acute intermittent porphyria would show increased levels of porphobilinogen.

Loose ends:
Q: If the disease is genetic, why don't either of Alice's parents show symptoms of the disease?
A: The disease is autosomal dominant, which means that even one copy of the gene should give you symptoms. However, some people with only one copy of the gene have been noted to have no symptoms, so there is a documented variation in the severity of the disease.

Q: Why was a diagnosis of pancreatitis thrown around early in the episode after the patient presented with abdominal pain?

A: Generally, when a doctor says that a patient likely has pancreatitis, he or she does so based on a blood sample that shows increased levels of amylase and lipase. Because pancreases make amylase and lipase to help digest food, damage to the pancreas breaks down individual pancreatic cells and releases these enzymes into the bloodstream, thereby elevating their levels in the blood. A CT scan could then be performed 48 hours after the onset of symptoms, with the goal to try to visualize some kind of fluid or cyst in the pancreas. I'm assuming then that this diagnosis was made because Alice was experiencing excruciating abdominal pain and had increased levels of amylase and lipase. Erythropoetic protoporphyrias are actually associated with gallstones (composed of insoluble crystalline protoporphyrin), which could have lodged in the common bile duct (like the one in the picture), creating a blockage of the main pancreatic duct and increasing the risk for pancreatitis.

Q: When Alice first developed a rash on her arm, why was it attributed to thrombocytopenia?

A: Thrombocytopenia is a low level of platelets, which can result in a petechial rash. From way up in the surgery viewing booth, it's possible that the patient's rash looked petechial in nature (like in the picture). I can't really think of a better explanation because I don't think a blood test was every run to confirm or reject this idea. Petechiae generally appear as small little dots.

Q: What things besides pancreatitis (and other anatomical diseases like appendicitis) can lead to abdominal pain, like the kind that Alice had?
A: There's a mnemonic I use from the Saint-Frances Guide to Inpatient Medicine to remember all the non-typical causes of abdominal pain.
Puking My BAD LUNCH.

Mediterranean fever

Black widow spider bite
Addison's disease, Angioedema
Diabetic ketoacidosis

Lead toxicity
Calcium high
Herpes Zoster

Q: Did vampires suffer from porphyria?
A: There is some thought that legends of vampires were based off of people who suffered from porphyria. Porphyria causes severe rashes when a person is exposed to sunlight, and one way to treat porphyrias is to transfuse blood products. Centuries ago when blood products weren't readily available, it would be theoretically possible to receive blood by drinking it straight from a victim's neck veins. Here's an interesting article that goes more into depth on the topic.

Wednesday, November 22, 2006

A Case of the Itchy Feet

The Patient:
Jack is an 18-year-old boy who presents to the hospital after experiencing a heart attack and massive vomiting. During the work-up, he tests positive for a number of bacteria.

The Diagnosis: Chronic granulomatous disease (CGD)

Working Backwards: In this case, the diagnosis wasn't impossible to make. Fairly early in the episode, House and his underlings run into an interesting finding: positive tests for Hepatitis A, syphilis, Eikenella, and botulism.
"So, we knocked down one infection and three more pop up? I think this game is rigged." - House
Well actually, any disease that targets the immune system can make you prone to recurrent infections. A relatively common disease that is often tested for in patients with unusual or widespread infections is Human Immunodeficiency Virus (HIV). But since when do House's patients come in with relatively common conditions? Seeing as how 2 of the last 3 House diagnoses (Episodes 305 and 307) have been genetic diseases, it's not a huge leap of the imagination that the writers of House would pick another hereditary condition for this episode. In all fairness, medical students acquire a knee-jerk reaction to thinking about hereditary immune deficiencies when the words "recurrent infections" show up in questions, so the final diagnosis was in some ways not entirely unexpected. However, chronic granulomatous disease isn't the only disease that fits into the category of genetic immune deficiencies, and here's where the medicine becomes interesting.

The Medicine: In order to understand how House arrived at the diagnosis of chronic granulomatous disease (CGD), you need to know a little bit about how the normal immune system works. The immune system is made up white blood cells, which are often fancily referred to as leukocytes or WBCs. There are five major types of white blood cells that make up the immune system: lymphocytes, monocytes (aka macrophages), basophils, neutrophils, and eosinophils. Lymphocytes come in three flavors, B-cells, T-cells, and Natural Killer (NK) cells. B-cells can become plasma cells and produce antibodies. T-cells exist in 2 forms, helper T-cells and killer T-cells. Helper T-cells are a bit like antibodies and help tag bacteria for destruction. Killer T-cells and NK cells do exactly what it sounds like they do -- they kill bacteria. Monocytes are big and slow Pacman-like cells that eat other cells whole. It can take 3-5 days for these bad boys to arrive on the scene, and so the immune system has neutrophils, which are smaller and stealthier. Basophils are basically wimps. Eosinophils fight crime in the form of parasitic infections.

So where does CGD fit into all this? Well, CGD affects the ability of neutrophils to eat bacteria. Just before healthy neutrophils eat a cell, they like to spray it with it with detergent (Picture bleach being splattered onto a bacterium and the bacterium writhing in pain). Why the need for detergent? Well, as sad as it is, neutrophils lack teeth and so they need a way to chew up the bacteria so that they can swallow it (apparently, neutrophils have throats and stomachs). They take electrons from a compound known as NADPH (nicotinamide adenine dinucleotide phosphate-oxidase) and transfer it onto oxygen to make superoxide, which is toxic to cells. In CGD, the enzyme that makes superoxide is broken, so the neutrophils suck at their jobs. Infections run rampant.

Not all infections run rampant though. This is the basis for the kooky "test" that House runs by injecting Jack with all those different bacteria. Interestingly, bacteria are neutrophil-wannabes in one way. Bacteria also have an enzyme that makes superoxide. Instead of using superoxide to kill bacterial cells, bacteria release superoxide to kill human cells. Sometimes, with all the superoxide flying around, bacteria can get caught in the crossfire and die off. Certain bacteria have an enzyme (called catalase) that can disarm superoxide, which prevents them from being killed by the crossfire. Bacteria (and fungi) that have catalase include Staphylococcus, Aspergillus, and Gram-negative rods (which includes Serratia and sometimes Eikenella). This is why when House injected all those different organisms into Jack, Serratia won out. Syphilis (caused by the bacterium Treponema pallidum) and botulism (caused by Clostridium botulinum) generally do not have catalase, although this doesn't exlude their ability to infect Jack, given that he has other catalase-containing bacteria to help out.

One final point: CGD is usually passed from one generation of a family to the next on patients' X-chromosomes, which means that males are predisposed to manifesting the condition. Since neither of Jack's parents were seemingly affected, his mom was the likely carrier (since her other normal X-chromosome made up the losses of the mutated one).

What's the deal with the itchy feet? For some reason, House was really caught up on Jack's itchy feet. Apparently, according to Harrison's (a big thick textbook), CGD can be associated with seborrheic dermatitis, which could in theory cause itching. I actually got thrown off by this whole itching symptom because one of the immune deficiencies that is classically associated with itching (in the form of eczema) is Wiskott-Aldrich syndrome. Lastly, the itching could have been due to a fungal infection, seeing as how Jack's immune system isn't so great. Most likely, Cameron was right when she said that he got it from wearing sweaty socks while running around at work. That's how my feet feel when I'm post-call.


Quotations, etc. Copyright Heel & Toe Productions, Bad Hat Harry Productions, et al.
Pictures from the show copyright Fox Broadcasting Corporation
Questions, Comments (remember we are not related to the show):

And if you are in Europe:

Fitbit Banner - Affiliate Program
See The New York Times article about the FitBit.
Holiday Gift Guide: Save up to 50% on Christmas Gift Ideas at Spend $25 to get FREE Shipping!
MIS, Inc. for Website Hosting, Development, Design & Promotion