Tuesday, June 30, 2009

Technology and Health Care

The problem with technology and health care is that technological advances seem to increase the cost of health care rather than decrease it. We got an amazing talk by Mark Smith, president of the California Health Care Foundation. In every other industry, technology decreases the price of goods and services; it gets cheaper and cheaper to buy a computer, make an international call, travel around the world, and buy GPS navigation. Yet in the U.S., producing a new MRI scanner or making a new kit or developing a new drug causes costs to skyrocket. Why?

There are a multitude of reasons; the cost of producing a drug that makes it to market is sky-high and patent protections prevent generics from competing. Although co-pays exist, consumers rarely bear the brunt of the costs so price has a blunted effect. But Dr. Smith brought up a very fascinating point, that the fundamental reason why technology doesn't reduce expenditures is a practice philosophy. We want the best for our patients and ourselves, not just "good enough." For example, a 64 slice CT may be good enough to diagnose appendicitis, but if a 128 slice CT is available, that becomes the standard of care. But this shouldn't be the case; the standard of care shouldn't be defined as the ceiling but as the floor. What is the minimum that guarantees safe, effective diagnosis and therapy? That should be the baseline. This way, new technologies entering the market must compete to make their added cost worthwhile. Right now, any technology, no matter the added cost, seems "better" than pre-existing equipment.

We need to stop hitting everything with a hammer. Compartmentalizing health care may also reduce costs. For example; patients shouldn't be going to the emergency department if there's no emergency. They shouldn't be going to acute care to get a pregnancy test. They shouldn't be getting prescription drugs when over-the-counters may suffice. Doctors don't need to be doing excuse notes from school or work, nurses don't need to be doing secretarial work. Doctors need to be diagnosing, nurses need to be triaging patients. We need to do the highest level of work that we are trained for; indeed, doctors are terrible at managing paperwork or fixing the computer system. Work efficiency is paramount to reducing costs.

Monday, June 29, 2009

Next Life

"Will someone please call a surgeon / who can crack my ribs and repair this broken heart..." ("Nothing Better" by The Postal Service)

If I had a little more courage, far fewer interests, and a masochistic penchant, I'd be a cardiac surgeon. They say the surgical soul is the pancreas, but for me, it's always been the heart.

Image is in the public domain, from Wikipedia.

Sunday, June 28, 2009

Poem: Glass Coffins

Glass Coffins

This kind of thing used to be civilized;
you'd kill someone, and they'd be dead
but like everything these days, there's all the paperwork,
forms and witnesses, documents to be faxed
before we can give death the go ahead.
Give me half a dozen meetings
as meetings go around here, and it'll be restricted
to Mondays through Fridays, business hours;
death will take twelve holidays off a year.

I remember when it was black and white,
when you were living or dead, and the in between
belonged to Michael Jackson music videos
and occulteers in dark alleys, when
there was no controversy; if you had a knife in your head
or the cough of consumption, we dragged in the box;
not this ridiculous business, shining lights at pupils,
an octopus sprouting from a dead man's mouth,
the hundred thousand dollar ambiguity.

They're like glass coffins, these rooms;
the white coats round and gawk and pat themselves
on the back since this is the closest they've gotten
to resurrection itself. I almost succeeded in rolling
a three day boulder in front of a room
before the nutritionist stopped me, said
"You can't do that, we need to give him tube feeds."

He never came back, this gentleman;
we didn't think he was Jesus anyway,
with more blood than brain in his head.
After a week of gumshoe-ing
we found a brother who wouldn't come in
but still we scheduled a time, 1600.
Even medical futility needed a schedule,
and though he aspirated at 1400,
a blooming pneumonia, an old man's friend,
we continued full steam for two hours
until morphine came waving down the caboose.

Saturday, June 27, 2009

Made for Medicine

One of my patients in the ICU is a young woman with hypertension, chronic kidney disease, crack cocaine use, and a seizure history who presented with a very bad intracerebral hemorrhage. She's still in a coma; she's not getting better. It is a sad story, and currently we're continuing aggressive care per wishes of the family because she is only in her 30s and has two children. However, one of her ongoing problems is her hypertension. She is currently on five anti-hypertensive medications: amlodipine 10mg daily, clonidine 0.4mg TID, labetalol 900mg TID, minoxidil 2.5mg BID, and a nicardipine drip. She's maxed out almost all her medications; because of her chronic kidney disease, we're avoiding renal-active drugs. Nevertheless, her systolic pressures still run into the 220s.

One afternoon, I had a little free time so I decided to think about causes of refractory hypertension. It could simply be due to the kidney disease, but I wanted to rule everything else out. I reviewed her extensive records and found that she had been worked up in the past for Cushing's, Conn's, thyroid disorder, and even pheochromocytoma. She had a normal renal ultrasound in 2003, but I wondered if we should do more investigation into renal artery stenosis, a common cause of secondary hypertension. I off-handedly put it in my note, but since ICU doesn't handle that kind of work-up, I didn't push for it. This morning, the neurology fellow said she read my assessment and plan and liked it. "It was a very medicine note," she said. I realized, even in this non-medicine rotation, I keep a bit of that thought process.

Friday, June 26, 2009

Living in the Hospital

I'm nearing the end of the dreaded 13 day stint on this rotation. The truth is, this rotation is pretty benign, but even benign schedules make me wince. We're here for 12 hour shifts every weekday, and I have a Saturday call (so a 30 hour shift from Saturday to Sunday noon) followed by a Friday call. So I've been in the hospital for 13 days in a row. By old timers' standards, it's easy, only 84 hours a week; I'm splitting time evenly between home and the hospital (there are 168 hours in a week). I shouldn't be allowed to complain about that. Even on my surgery rotation, I reached a hundred hour work week. But it's still tiring, and I'm itching for a quiet call night and a free Sunday. On the other hand, it is quite nice to be on for such an extended period of time. I know all the patients, I know all their issues, there are no surprises.

Thursday, June 25, 2009


Last call night, I had a very terrifying incident. At around 3am (when things happen around here), the resident and I had to put in a central line for one of our patients. A central line is an IV that goes into one of the larger vessels: the internal jugular vein in the neck, the subclavian vein under the collar bone, or the femoral vein in the groin. It is used to monitor the volume status of a patient and deliver certain medications that are toxic in peripheral intravenous lines.

The procedure, while standard and common, has its risks which include arterial cannulation (putting a needle in the artery instead of the vein as the two run together) and pneumothorax (putting a needle into the lung, causing it to collapse). We minimize that risk using ultrasound to see the vessels in the neck. The resident and I gowned up, prepped the neck, and began the procedure. After a few attempts, I was unable to cannulate the internal jugular vein (I think I'm too timid; up close, the needle looks huge). Although we visualized the anatomy by ultrasound, it was not easy.

Then the resident attempted the central line. After a try, she managed to get blood flow back. I was relieved, but when she unscrewed the syringe from the needle, pulsatile blood spurted back. It was not venous; it was arterial. We pulled out the needle and held pressure. Holding pressure for 10 minutes at 3:45AM feels like forever.

Suddenly, the monitor began to beep. We looked up, and I panicked. The heart rate was reading 30 beats per minute, and then it disappeared. She was asystolic; she had flatlined; her heart had, at least according to our monitors, stopped beating. In retrospect, the telemetry strip showed only a five second pause, but it felt like hours. In a classroom or on a test, I would know what to do, but here, I was stunned, petrified. But the resident responded in stride, asking for atropine, a drug to speed up the heart. She also figured out the cause of the asystole; by holding pressure, she was giving the patient "carotid massage." The carotid sinus contains baroreceptors that measure pressure; by pushing on the artery to tamponade bleeding, she had tricked the sinus to thinking the pressure was really high and the body responded by dropping the pressure and slowing the heart (parasympathetic response). This surge of hormones had slowed and possibly stopped the heart.

By releasing the pressure, the heart beat returned. The nurse had also given a bit of atropine as well. The patient ended up doing fine; there were no lasting complications beyond the arterial cannulation. But it taught me a good lesson, the anesthesia mantra that vigilance is necessary at all times.

Wednesday, June 24, 2009

Brain Death

Last week, we declared brain death on one of our patients. I think the last time I broached this topic was when I learned about it as a first year medical student. Brain death is death. A person who is brain dead is dead, and though that is a tautology, it is easy to forget (after all, logicians make careers out of tautology). The patient had a massive stroke and his brainstem had herniated. Before we knew that, we had initiated measures to keep him alive, putting him on a ventilator. But the neurology exam was completely negative; he had no response to pain, fixed pupils, and no brainstem reflexes (oculocephalic, corneals, cough, gag). When we turned the ventilator off, he did not spontaneously breathe. Thus, although we could keep him breathing and his heart beating, he was brain dead, and thus, dead. His family wanted to donate organs, but unfortunately the organ donor network decided not to take his organs (he was over 80 and had some dysfunction of all the organs).

On the other hand, we had a patient who fell and hit his head a week ago. He had a massive head bleed but still preserved some of his brainstem reflexes. Thus, he was not brain dead (and also not dead; since I'm in a logic mood, "if brain dead, then dead" does not imply the inverse, "if not brain dead, then not dead"). The decision to stop the ventilator for this gentleman is much more complicated; since he is alive, withdrawing life support requires a declaration of medical futility or a decision to do so by the durable power of attorney. After several family discussions, we did end up withdrawing life support. Although it is a sad thing, I feel that it was the right choice.

Tuesday, June 23, 2009

The Basilar Artery

A 60 year old woman started feeling "off" a few weeks ago. Something was just wrong with her balance. She saw her doctor who noted gait instability but her work-up, including a non-contrast head CT, was normal. She was reassured and sent home, but the symptoms didn't abate. Then, she noticed something wrong with her talking; she couldn't speak clearly. Her speech was slurred. Still, she continued with her life as best as she could until she fell several days ago. Her son witnessed the fall and convinced her to go to the hospital even though she didn't lose consciousness or have seizure activity.

In the emergency department, she got the standard work-up for syncope (fainting) which didn't reveal any answers. But then over a half hour period, she went from talking coherently (GCS 15) to completely unresponsive (GCS 3). Even to painful stimuli such as rubbing her sternum or pinching her fingers, she would not wake up. Something had gone terribly wrong.

Neurology came down to assess the patient. Their exam was concerning; her pupils did not react (constrict) with light as they should. She had a few brainstem reflexes such as the corneal reflex (touching her eye causes her to blink) and a gag, but with such a sudden change in her mental status, they were concerned about stroke. She was rushed off to get a CT angiogram which showed a basilar artery thrombosis.

Thrombosis refers to a clot in a blood vessel. Clots in blood vessels to the brain account for the majority of strokes; a clot in the pipe prevents the blood from delivering oxygen and taking away toxins from the brain. Strokes are bad, but basilar artery thromboses are dreaded. The basilar artery supplies the posterior circulation of the brain and brainstem. While strokes in other locations present with a facial droop or one-sided weakness or difficulty with language, strokes involving the basilar artery kill.

Yet she was within the window of trying to dissolve the clot with tissue plasminogen activator, a clot-buster. These potent drugs are incredibly dangerous because they can convert a clot into an unstoppable bleed. They have to be used early, before the brain sustains too much damage. The neurologists wanted to administer this medication by artery (intra-arterial tPA) to break up the clot and restore blood flow to the brain; if they did not, the lack of oxygen would cause the brain cells to die, swell, and kill the patient.

They called interventional radiology, experts in using CTs, MRIs, and other imaging techniques to do precise procedures. They cannulated her vertebral artery, went up into her brain with a catheter, and injected the tPA clot buster. They then threaded a corkscrew-like coil, the Mercy Retriever, through the clot of fat and pulled it out of the brain, restoring blood flow.

The patient came to the ICU at around 3 in the morning. When I saw her, my exam was not reassuring; though she did have a cough, gag, and corneal reflex, she did not withdraw her arms or grimace to pain. I'd seen this disease, basilar artery thrombosis, one other time and that time, we withdrew care on the patient because he didn't get better. Despite heroic measures, I didn't think this woman would make it.

The next morning, I went to round on her. I was shocked. When I asked her to wiggle her toes, give me a thumbs up, and open her eyes, she did them briskly. She was able to nod and shake her head appropriately to questions (she had a breathing tube so she was unable to talk). Her mental status had changed remarkably. She went from comatose to appropriately responsive. The embolectomy (taking out the clot) and intra-arterial tPA had worked. I was floored (so was she, since we sent her out of the intensive care unit to the "floor" - bad pun). Modern medicine truly saved her life.

Monday, June 22, 2009

Poem: Atop Parnassus

Atop Parnassus

If God heals and the doctor charges the fees
then take me atop Mount Parnassus please
where muses in their limestone caves
sing and dance and harp away the graves,
invoking great Apollo's grace
since potions could not best a God's embrace
or so the ancients thought years past;
now, modern medicine has surpassed
legend, casting aside antiquity and belief,
replacing it with science in sharp relief,
until the only thing left that is Greek
is the name of the street on which we speak:
Parnassus, oh Parnassus, named by some guy
sitting on high who knows we still rely
on a bit of faith, a bit of muse
in choosing the medicines that we choose.
Forgetting history, casting away humility
is to discard sense and sensibility.

Sunday, June 21, 2009


For ICU, call is infrequent but a killer (for me at least). We only take call once a week, on a Thursday, Friday, and Saturday, and the shifts are about 30 hours. It's tough because things are entirely unpredictable and patients are sick. Both call nights so far, I've admitted at least 3 patients, the maximum I'd admit on medicine (tonight I'm up to 5 and it's only 12:30; it's a Saturday after all). Although we only worry about sedation, pain, access, and respiratory status, I still find it busy and intimidating. Access can also take a while; it always takes longer than expected to put in an arterial or central line. Since patients are sick, there's a momentum when they hit the ICU doors and there's a lot to do for them all at once. Meanwhile, we have to worry about all the other patients in the unit whose catheters have fallen out, whose respiratory statuses are getting worse, whose families want an update. Indeed, tonight one of our patients has what appears to be delirium tremens, a dangerous form of alcohol withdrawal. His visual hallucinations were interesting. To me, it's incredibly intimidating being responsible for a unit of 20+ patients. Furthermore, things don't quiet down in the ICU; even at night, respiratory therapists are doing spontaneous breathing trials, patients are going to and coming back from the operating room, action is happening. The trick is to get sleep when you can (rather than blog, I suppose) because you never know when things will get crazy. Speaking of which, apparently there was a big fire and now we're getting a woman intubated for inhalational injury. Oh well, I wouldn't have been able to nap during the time of this blog anyway.

Friday, June 19, 2009


Every time we stick a needle in someone, take a scalpel to the skin, or put a tube down someone's throat to help them breathe, we must remember that we do harm. Even in choosing the potent medications we have, we do harm. We justify this by saying that such evils are necessary to effect a far better good, that the benefits outweigh the risks, that the process of healing is not free of pain. I believe that to be true. But nevertheless, I should never forget that in doing so, I may inconvenience, risk, or even harm those I intend to help.

Image is of the statue of Hippocrates at UCSF, from Wikipedia, shown under Creative Commons Attribution 2.0 License.

Thursday, June 18, 2009


One of the poignant things I've noticed on this rotation is human fragility. I get an incredibly skewed view of things because it is the surgical intensive care unit, but nevertheless, the view from here is grave (that eternal pun). The patients here fall under two general categories. Some patients are responsible for their dismal situation. I'm taking care of a young mother of two with bad hypertension who stroked while smoking crack cocaine. She had been admitted several times in the past for crack use, but this one was devastating; she's now in a coma, intubated, her skull removed to relieve the pressure on her brain. Another gentleman who will likely pass away today was drunk driving and crashed into a parked car. The stories go on and on: a man in his twenties involved in a gang fight, shot in the side of his head; a drunk woman who flipped a golf cart; a drunk man who crashed his car into a pole (drunk motor vehicle accidents are a recurring nightmare).

But I actually feel the most emotion regarding the other set of patients, those who are here completely by accident. We have a driver of a horse-drawn carriage whose horse stopped abruptly when an errant bicyclist sped in front of them. The driver, unrestrained, was thrown off the carriage and now has severe brain bleeds. Another previously healthy Cantonese grandmother was hit by a truck; she lost a leg, her pelvis was crushed, she's now septic (infection of the blood) and we're fiercely trying to resuscitate her. Yet another man was a long distance bicycle rider wearing a helmet, involved in a crash with a severe skull fracture and brain bleed; we're not sure he's going to make it.

Today we're transitioning two of our patients to comfort care; that is, we've felt that we cannot reverse what's happened and we're going to withdraw care and let them die in peace. The family is at the bedside. They're sobbing, praying, pleading, confused about this unexpected and terminal change of events. I feel an unease, distraught and miserable that a perfectly well person, on her way to a game of golf or driving home after a night of partying, can in a matter of moments, end up here in the intensive care unit, where despite "heroic" measures, we cannot salvage their lives.

Wednesday, June 17, 2009

Arterial Lines

One of the fun things about ICU is that it is a procedures service. We maintain airways, monitors, and access for patients. One of the more common and less risky procedures is the arterial line. By cannulating the radial (or dorsalis pedis) artery, we can transduce the blood pressure and draw arterial blood gases. It's similar to but more difficult than placing an IV. Over the last week and a half, I've placed a few arterial lines (and attempted several more), and I've realized I like procedures. I like being hands-on, planning, troubleshooting, and the instant gratification once something works.

I've also seen a few emergency intubations when a patient's respiratory status acutely decompensates. It's really good watching the anesthesia residents approach the potentially scary airway and how they keep their methodical train of thought even in stressful situations. The other more difficult procedure is the central line, a catheter placed into the internal jugular or subclavian vein; I've seen the residents place a few for patients who needed better cardiac monitoring or medications that cannot be given through a peripheral IV.

Tuesday, June 16, 2009

Medical Malpractice

A while ago, we had a lecture on medical malpractice; much of it was on the nuts and bolts of the law, the process of going through a lawsuit, and reasons patients sue. But I was interested to hear about different views of the medical malpractice system. Proponents say that it is an indispensable tool to police a profession that won't police itself, that the threat of litigation prevents medical errors, and that it provides a venue for justice. Opponents of the malpractice system claim that it is random, only awarding a few patients, costly, encouraging defensive medicine, not correlated with negligence, and counterproductive.

Indeed, the data is shocking; in 1994, cases going to jury trial with a plaintiff verdict had a median award of $300,000 compared to $1.2 million in 2003 (Joint Economic Committee, US Congress, 2003 Data from Jury Verdict Research). Simply the cost of legal counsel is expensive; 64% of cases are dropped or dismissed but the mean defense cost is $18,330, 30% are settled with a mean defense cost of $50,452, 5% go to trial in favor of the defendant with mean defense cost $106,196, and 1% go to trial in favor of the plaintiff with mean defense cost $185,060 (PIAA Claim Trend Analysis, 2008 ed.).

The rising costs of litigation and rising payouts have driven up premiums for malpractice insurance; in Chicago 2008, ob/gyn had a premium of $138,434 (compared to general surgery $98,888 and internists $37,688). This varies widely geographically depending on malpractice reform. Luckily, California's 1975 MICRA act caps non-economic damages to $250,000 and limits contingency fees (that is, the percentage of award money that ends up in the hands of the lawyer rather than the victim).

Does the medical liability system compensate the negligently injured? This is a hard study to do. In 1991, an NEJM article argued that 2% of negligent injuries resulted in a claim and 17% of claims were related to a negligent injury, suggesting that there's poor correlation between negligent injuries and lawsuits. A 2006 NEJM article found that 3% of lawsuits didn't even involve an injury (yet 16% of those paid out). Of those with an injury, 63% had an error (73% paid out, average $522,000) and 37% had no error (yet 28% paid out, average $313,000). They found that 54 cents of every dollar did not go to the plaintiff, but instead went to administrative costs (mostly legal fees).

After these lectures, I am unconvinced that the legal system is addressing medical malpractice correctly. The biggest risk factor for being sued is not making a mistake, but rather having poor communication. Most patients injured through negligence are not compensated and even those who are compensated don't get the majority of the award. Most physicians who are negligent are not sued, and some physicians who are innocent are sued and lose. The costs of this legal system drive up not only premiums but the overall costs of medicine by forcing physicians to act defensively. I think the legal system is not the best way to encourage physicians to improve quality; many problems are systems problems not addressed by malpractice suits.

What can we do? I would not want to limit access to courts (by decreasing statute of limitations or necessitating pre-trial screening panels). Perhaps specialty courts such as the tax court would be better; after all, medicine is incredibly complex and I feel that it is absurd to think that a jury of laypeople can fully understand the issues at hand. Specialty courts with a trained judge may make more sense in such a specialized field. I do think that there should be caps on awards and regulation of attorney fees. The ultimate goals are to compensate the medically injured in an equitable and fair fashion as well as deter and prevent future errors and genuine negligence.

Sunday, June 14, 2009

Poem: Inventing the Calculus

No poem collection would be complete without a dedication to Leibniz, praise be his name. In parallel with Sir Isaac Newton, he invented the calculus; there was much controversy over whether he did this independently. This poem makes reference to his philosophy of monads, the ultimate elements of the universe with irreducible simplicity. Leibniz-Keks are a popular brand of biscuits in Germany. (Image is in the public domain, from Wikipedia).

Inventing the Calculus

Gottfried Wilhelm Leibniz once said
the world is composed of simplicity.
That was before the media of his day
accused him of duplicity.

Now, biscuits are his lasting legacy,
neither differentials nor integration.
We think of him when teaching scientists
to keep good documentation.

Saturday, June 13, 2009


This last week, I started my intensive care sub-internship at San Francisco General Hospital. It is an open surgical ICU meaning we act as a consult team but not a primary team. We handle all pain, sedation, ventilator, and access requirements for patients in the ICU. We can suggest to the primary team what else they might want to do, but they make the ultimate decision whether or not to take our recommendations. Being in this capacity is kind of fun; we don't have to worry about the day-to-day management of these patients and don't get paged for anything except sedation and airway problems. The surgical ICU takes mostly trauma and neurology patients, but we also take the occasional pediatric or ob/gyn patient.

The service is big; we've had about 20 patients all week. Many are victims of car accidents, falls, gun shot wounds, and strokes, so it's an interesting mix. Nearly all our patients are on ventilators and propofol, fentanyl, and/or lorazepam drips. They have invasive lines such as central venous catheters and arterial lines. We start early in the morning with radiology rounds; we look at all the new films for all the patients (nearly all of them get a daily chest X-ray to look at their pulmonary status and the positioning of lines and the endotracheal tube). Then we do bedside rounds which can take several hours, discussing and seeing all the patients. Each day, we have a lecture on an ICU-related topic. Then the afternoon is composed of admitting new patients and doing work on the olds, changing lines, adjusting ventilator settings, transporting patients. Since trauma can come in any time, admissions are unpredictable and since patients are sick, we need to see them immediately. It's very different from my medicine rotation last month.

Thursday, June 11, 2009

Medicine Sub-I

My medicine rotation went well. I learned what it would be like to be an intern (now I'm scared). I carried more responsibility, developed sharper judgment, and became a primary provider for a patient. I began to make decisions and commit to them which is awfully hard but educational. As a third year student, I had wishy-washy plans for management; "well, if she has a gastrointestinal bleed, we should give her fluids and transfuse. But since she has heart failure, we have to be careful in giving her any fluids." As a sub-i, I had to decide how much, how fast, and how to assess if things were getting better. As a whole, I cared for mostly bread and butter cases, sometimes with psychiatric comorbidities and complex social issues. I learned a lot, but more about time management, logistics of health care delivery, and how to interact with consultants than about disease processes. The learning curve was steep but I think I was prepared to tackle it.

How does this change my thinking about medicine? I feel that hospital medicine is mostly run by house staff (interns and residents). They assess the patients, decide on a plan, and execute it. Attendings merely supervise. And indeed, good attendings don't micromanage. They teach and act as consultants to the team. I'm not sure what to make of it. I won't like intern year regardless because of the grunt work. I would probably enjoy being a resident because there's an intellectual challenge in diagnosis and management decisions. But as an attending? I'm not sure.

I love medicine in theory. I love thinking through complex cases, learning about people, trying to put together that puzzle of diagnosis. I fit in with the people. Medicine interns and residents have been some of the best I have worked with; they are brilliant, inspiring, hard-working, friendly. But as a sub-i, I learned medicine in practice is very different than textbook medicine. Patients will yell at you, decline important procedures, leave the hospital to smoke. Laboratory results will be miscalibrated, equivocal, or missing. Presentations rarely fit classical paradigms and answers can be elusive. There is an emphasis on quantity over quality, on getting work done than doing things perfectly. It is a privilege and a frustration all at the same time. Typing this makes me seem jaded, and I think it's somewhat true; after a year on the wards, I have really learned the limitations of medicine.

Wednesday, June 10, 2009

Skin Exam

There are a lot of barriers to doing a good skin exam, especially in the hospital. We rarely undress patients to inspect their entire body, we assess people in the emergency department in gurneys in the hallways, we assume patients will tell us if they notice weird rashes. But they don't and the skin exam is crucial (even if all rashes are "maculopapular"), especially for certain chief complaints. This is simply a reminder to myself not to forgo the skin exam because it seems inconvenient, hard, or unimportant.

Tuesday, June 09, 2009


I really like this picture, taken by my friend Grace, PharmD, while on vacation in Japan.

Monday, June 08, 2009

Poem: Apology

I had forgotten how difficult it is to write a villanelle. The villanelle is made up of five tercets and a quatrain. The first and third lines of the first stanza are rhyming refrains that alternate as the third line in the successive stanzas and form a couplet at the close. There are only two rhyming sounds in the entire poem. My favorite villanelle by far is Elizabeth Bishop's One Art, which takes some license with the form but does it masterfully.

This poem is a brief attempt at the form. It's very rough and I'm not sure it makes complete sense, but here it is.


Picking you out was more art than theft;
though many wear rings, I knew
you'd leave, and money would be all you had left.

No love, no children, no future, bereft,
so I convinced you I could see all your dreams through;
winning you over was more art than theft.

Talked myself up, talked him down, deft
in insinuations, until even you would rue
that marriage; regret was all you had left.

A goodbye to him, then you leapt
when I asked for your hand, said I do.
Making you love was more art than theft.

I never stole anything, it was you who left
with half his fortune on our wedding debut,
a broken broke man was all you left.

Now, I apologize for leaving you bereft
but I never loved you true.
Leaving you now is more art than theft,
and this poem is all you have left.

Sunday, June 07, 2009

Hypothermia II

This patient (described in the previous post) with a core temperature of 24.8 degrees was intubated for airway protection and rewarmed. During the process, he began to have cardiovascular collapse, possibly from the vasodilation during rewarming. He was admitted to the intensive care unit. This was truly an ICU patient; he had something wrong with every organ system. From a neurologic standpoint, he was altered and then sedated. When he was extubated, he was only oriented to person and his speech was incoherent. From a cardiovascular standpoint, he required vasopressors in the ICU to maintain his blood pressure. From a pulmonary standpoint, he was not only intubated, but subsequent scans found incidental pulmonary emboli and moderate bilateral pleural effusions. From a gastrointestinal standpoint, he had severe necrotizing pancreatitis. In my reading, I learned that cold is one of the lesser-known causes of pancreatitis (along with the dreaded scorpion sting). Repeat CT scans showed necrosis of >30% of the pancreatic body and tail; necrotizing pancreatitis is a dangerous variant, but it is not surgical unless the patient is infected. We covered him with ertapenem. He also later developed a transaminitis which we attributed to TPN cholestasis. From an FEN/renal (fluid, electrolyte, nutrition) standpoint, he developed renal failure in the ICU, probably from acute tubular necrosis and contrast nephropathy. He required CVVH renal replacement therapy. From a hematologic standpoint, he had a thrombocytopenia of unknown etiology in the ICU which resolved on its own as well as a severe anemia (Hgb 7.8) that was likely multifactorial (a borderline macrocytic nature and history of alcohol use suggested B12 and folate deficiency, the iron studies suggested a component of chronic disease, and a finding of body lice could also contribute). From an infectious disease standpoint, he had a rising leukocytosis of unknown origin. After his ICU issues resolved, he was transferred to us and given to me. It was a lot of hard work, but very interesting.

Saturday, June 06, 2009

Hypothermia I

One of the patients I admitted last call was a middle aged man found down on a sidewalk for an unknown period of time. His core temperature was 24.8 degrees Celsius (76 F). That's cold. The ambient temperature these days is 15-20 degrees Celsius. A normal temperature is about 37 degrees Celsius. Hypothermia is a temperature below 35; severe hypothermia below 28. At that temperature, enzymes in the body fail, molecules crystallize, the heart is at risk for life-threatening arrhythmias.

Hypothermia itself is a fascinating illness to me. Many of the things we normally rely on are untrustworthy. For example, hypothermia causes a "bleeding diathesis"; the normal coagulation proteins stop working and people bleed easily. However, the laboratory test for coagulation is done at 37 degrees and so results are misleading. Likewise, arterial blood gases are done at 37 degrees, electrolytes may be inaccurate, and hematocrit is concentrated as temperature drops. The EKG starts showing Osborne waves, present in this patient, and very cool to see since they are so rare. You can't use insulin in patients under 30 C because it will crystallize.

You can rewarm patients in many ways. Mild hypothermia can be managed passively with blankets (it makes me think of the mantra for newborns: warm, dry, stimulate). With more severe hypothermia, active rewarming can take the form of heating pads, warm blankets, warm humidified air by endotracheal tube, and warm fluids by IV. Even more aggressive measures include irrigating body cavities (pleural space, peritoneum, bladder) with warm fluids by chest tubes or peritoneal lavage. A patient can even be warmed by extracorporeal methods, removing blood, warming it externally, and returning it to the patient.

Of note, you have to warm the trunk before the extremities. Warming the extremities first can cause a paradoxical cooling because cold acidotic blood from the arms and legs returns to the core. Warming the extremities also causes vasodilation which can drop the blood pressure. Speaking of paradoxical phenomena, there is also this strange behavior called paradoxical undressing which happens with moderate hypothermia; people will take off their clothes, worsening the hypothermia; the reason is unclear.

Friday, June 05, 2009

Gender and Death

When I was on call, I admitted a patient who taught me two very important aspects of medicine. She was a male to female transgender patient with an end stage chronic disease admitted to our service for comfort care. She was a hospice patient but because they didn't have adequate staffing on the weekend and felt that death was imminent, sent her to the hospital. The first thing I learned was about gender. Though she was a transgender patient who identified as female, she was roomed with a man (it was terrible we couldn't even get a single room for a dying patient). I wonder how the hospital decided on this; certainly, it's not a rare thing for a transgender patient to be admitted. But it bothered me that nurses, staff, and even doctors referred to the patient as a "him." It can be hard to remember to call her by her preferred pronoun, but it's so important.

The second was about death and dying. When I saw her, she was extraordinarily cachectic with extreme wasting. Her thighs were the circumference of my wrist. I could see all the bones of her cheek, the zygomatic arch, the mandible. I could count her ribs, almost make out a liver edge. She was nonverbal at baseline. We quickly put in orders for morphine, titrate to comfort. It is a strange feeling, watching someone die. Her respirations were agonal; she breathed small gasps at 30 a minute (try doing that; it's extraordinarily uncomfortable). I checked in on her frequently and increased the morphine to decrease the air hunger. Over the course of the night, she became closer and closer to death; at around 5am, she had spontaneous conjunctival hemorrhages. A little after 8, we were called to bedside to pronounce time of death. The resident and I went over how to pronounce someone and we listened to her heart and lungs, felt her pulse, examined her pupils. After confirming death, I called the case manager, organ donor network, and medical examiner.

Unfortunately, the patient died alone. She was estranged from family and only had a case manager as her outside contact. I spent as much time as I could with her in her last hours because I remember learning that patients don't want to die alone. It was a very poignant evening for me. May Mother's embrace welcome her home.

Wednesday, June 03, 2009


One of the patients I admitted last night had a new diagnosis of HIV. He came in with nonspecific symptoms and one of the routine labs sent by the emergency department included the rapid HIV test with over 99% sensitivity (that is, <1% results are false positives). He was shocked. At San Francisco General Hospital where the incidence and prevalence of HIV is relatively high, we have a lot of resources for newly diagnosed patients. I was not there when the emergency department delivered the news, but I saw the patient afterward.

How many ways one could react to such news! He went through all the Kubler-Ross stages that night. Early, he could not believe it happened, reciting to me his avoidance of high risk activity, struggling to come to terms with this shocking news. He never really became angry, but there were moments of impulse and outburst, punctuated by severe anxiety and panic. He bargained; if only the laboratory samples had been mislabeled, how he would take care of his health. Finally, he moved into depression, a time of introspection and grieving before showing signs of starting to accept.

It was good for me to be involved in this patient's care. Although we see many patients with HIV, I'd never seen someone with a new diagnosis and facing the psychosocial burden that comes with it. It reminds us why we went into medicine, to care for people, not simply for their diseases.

Monday, June 01, 2009

The String Test

Confabulation is a fascinating symptom where someone forms false memories, perceptions, or beliefs of the self or the environment. A person with confabulation doesn't just have poor memory; they make up their memories. We have a patient on our service with possible Korsakoff syndrome, a late neuropsychiatric manifestation of chronic alcohol use leading to thiamine deficiency. The string test was amazing in identifying confabulation. The resident pretended to reach into his pocket and pull out a piece of string. He unfurled this imaginary string in front of the patient, saying "Now I have a piece of string here, can you tell me what color it is?" The patient looked at the expanse of air between the resident's hands and replied, "Well, I'm color blind, so it could be red or blue or white or black." It was one of the most amazing things I've seen this rotation.