Search This Blog

Loading...

Follow by Email

Monday, December 1, 2014

In hospital versus out of hospital heart attacks: wow, things sure cost a lot of money!

An article from the JAMA (Journal of the American Medical Association) has been gnawing at my consciousness for the last couple of weeks. Dr. Prashant Kaul and colleagues out of the University of North Carolina reviewed records from hospitals in the state of California from 2008 through 2011, looking for patients who had been hospitalized with heart attacks. Specifically, they were looking for patients with ST elevation myocardial infarction (STEMI), which are generally the most damaging and deadly of the events generally known as heart attacks, due to the amount of damage they do to the heart muscle. The authors compared patients who were already in the hospital for another reason when they had their heart attack, versus ones who were admitted specifically for heart attacks. They found that the patients who were admitted specifically for the heart attacks were generally younger and healthier, more often male, and were much more likely to survive than the ones who were hospitalized with other illnesses at the time of their STEMI. This is not terribly surprising, since people who have some other problem bad enough to put them into the hospital and then develop a heart attack on top of it are clearly at a disadvantage, even though there are cardiologists with magical potions and procedures close at hand.

What was most interesting and disturbing to me was the sheer astounding magnitude of costs associated with these groups of patients. The patients admitted for STEMI stayed an average of 4.7 days and total costs were $129,000. About 9% of them died. The patients who were already in the hospital at the time stayed an average of 13.4 days, their costs were $245,000 and a third of them died in the hospital.

I don't think we should get jaded to numbers like this. This is real money, the kind of money that can buy a house in some places or at least a very hefty down payment, can support a person for years, and the co-pays on which can destroy a family financially. As a person is racking up such a bill, there are days of inadequate food and sleep, indignities of hospital gowns and waiting for someone to come with medication or to allow one to empty bladder or bowels, if it's not already too late. And death, in 9-30% of the people thus cared for. In a hospital. This money is not buying comfort and luxury. What costs so much? I'm not entirely sure. The interventions done on people with heart attacks include bypass operations, which are costly, but happen to very few of these people. There are the "percutaneous interventions" meaning high-tech catheters passed through arteries to place stents in clogged blood vessels in the heart, which are also terribly costly, sometimes as much as $40,000 for placing a tiny metal finger trap in a partially blocked artery. More stents are placed than need to be, per many studies, but an STEMI is definitely a good reason to place a stent and doing so is often life saving.

But why? Why so much money? There is nothing absolutely expensive about any of this. A little expensive, yes, but not hundreds of thousands of dollars. But the costs add up. The equipment is getting incrementally slightly better and is priced somewhere in the ozone. But it's not about raw materials or time or any of the resources that are truly set in stone. The costs just rise to the level that we agree to pay. The many places where money hemorrhages from the system feed our vibrant healthcare economy. We pay huge amounts of money to insurance companies who disburse it to the entities that charge this much. If there were limits on costs, or even goals for cost cutting, I'm confident we could slim down our spending. But there aren't and we don't.

Heart attacks and their treatment are just a tiny piece of the picture. There are still a few good values (a needle and syringe still costs less than $1), but generally everything that has to do with healthcare is overpriced. I learned a new computerized medical record keeping system last week and talked at length to the trainer who had been instrumental in adopting it. I complained because it was clearly clunky and lacking in the subtleties that would have made it really useful. I asked about another program I had heard about which was looked at as the best. According to her, the "best" cost about half a million dollars per hospital bed to implement. A medium sized hospital might be 200 beds. So 100 million dollars. Apparently hospitals, hoping for efficiencies, have gone bankrupt after adopting this Mercedes Benz of medical records. And the other systems aren't much cheaper. How is that even possible? There are almost no fixed costs in computer software. They charge this much entirely because they can.

There are no obvious solutions to this, while we remain attached to a non-centralized third party payment system. Payment structures are changing, but slowly, and the powerful interests who make money off of this system seem to escape ideas made to dampen profits. As individuals, though, it's important to continue to notice that things cost too much, they don't have to, and it's not OK.




Wednesday, November 19, 2014

Lions and Tigers and Doctors Unionizing, Oh My! Interesting developments in the field of hospital medicine. (Fresenius buys Cogent and Sound hospitalist companies. Is that good?)

I am presently doing locum tenens shifts in a lovely community in Oregon as a hospitalist. (For people not steeped in the lingo, that means I am filling in as a hospital doctor.) I have been to this hospital before and was glad to return when they needed some help. I like this place and noticed on my first go around that patients got good care and that physicians and nurses all seemed to get along pretty well together.

When I first worked here, 2 years ago, they had just transitioned away from a national company that organized and provided hospitalist coverage. The company was expensive for the hospital and refused to work with the doctors to allow them to have reasonable work loads. Not having a reasonable work load as a doctor is not just an irritation, it is dangerous. An overworked physician is not available to respond to, or ideally avert, emergencies. Because doctors are primarily over achievers, being unable to provide good care because of being responsible for too many patients is incredibly demoralizing and often results in burnout, with its associated depression, anxiety and sometimes substance abuse. We hurry through our visits with patients because there are so many to see, missing important clues to diagnoses and ignoring social issues that are vital for providing appropriate treatment.

At this hospital, the doctors met with the administration and came up with a plan that involved getting rid of the hospitalist company, managing the staffing and billing themselves with the help of a hired coordinator. They still have days when there are too many patients to see, but they have hired enough staff that this is rare and they are free to adjust in order to maintain good patient care. This was not possible when they were employed by the national company.

Hospitals, though, often balk at the idea of managing their own hospitalist groups. This is for good reason. Hospitalists are expensive. Even though we take care of many patients, the money we bring in as revenue to the hospital does not cover our salaries. In fact, it doesn't even cover half of our salaries. Partly this is because many of the patients we see are unable to pay anything for their care. The sickest of patients often have poorly paying Medicaid or state funded insurance or have no money or insurance at all, which is partly why they are so sick, or at least strongly associated. Hospitals are willing to subsidize a hospitalist service, though, because having hospitalists attracts many of the physicians that do make a hospital money, such as surgical subspecialists. These folks need to be able to do operations and be available for emergencies requiring procedures without worrying that they will be called away by hospitalized patients who have complex medical problems outside of their area of expertise. Hospitalists are mostly internal medicine physicians who are good at managing all kinds of chronic medical problems and are in the hospital all the time, ready to take care of any patient who needs urgent help.

There is risk associated with managing a hospitalist program. Doctors sometimes get sick (god forbid) and replacements are needed, which can be difficult at the last minute. Occasionally a doctor who is hired for a position as a hospitalist, who sounded really good on paper, turns out to be pretty awful. They have a terrible temper and yell at nurses or they never do their documentation or they have bizarre practice styles. Whoever is in charge of a hospitalist group must find somebody to fill in, super quick, plus try again to find the right permanent member of the group. Running such an organization is not for the faint of heart.

Intrepid, expensive and mostly mercenary companies have arisen to fill the void. "Afraid to run a hospitalist program? We will do it for you and it will only cost millions or dollars a year!"

A California hospital where I have worked answered just such a clarion call. Their hospitalist group
had formed out of community physicians and new hires and had even involved a merger. The doctors kept track of their patients' billing information and billed them directly, and never worked particularly well together. Nobody was boss, so standards of care were a bit haphazard. They made considerably more money if they saw more patients, so even though they hated being overworked, they tended to end up with over-large patient loads. They knew exactly how much hiring doctors to fill in cost, since it came directly from their pockets, and they were loath to do it. They became burned out. The overwork and burned-outness made other physicians not want to work there. So they agreed to have a national hospitalist company take over. This is looking like a disaster, though only time will eventually tell. They will take a pay cut, and the company will expect them to continue to maintain higher than safe patient loads. Many have already quit.

Just yesterday I heard about a hospitalist group in Eugene, Oregon, which has decided to unionize because of overwork (local paper article here, with excellent commentary). Thirty six doctors at PeaceHealth hospital in Eugene and Springfield have decided to form a union, which will be part of the American Federation of Teachers. This large group also is a home to nursing unions, so the connection is not all that weird. Attempts to reduce costs associated with the hospitalists by the hospital administrators lead to unmanageable patient loads. The hospital dealt with the doctors' complaints by proposing having a national company take over the program. The doctors voted to unionize, a strategy that has helped nurses have a voice in the past, when reduction in staffing and salaries made them very unhappy.

I'm not quite sure how this will play out. The hospital can still hire an outside company to provide hospitalist services, though they will have to negotiate with the union if they want to use the local physicians who really know the patients and the system. Well paid scabs might fill the void.

I have worked for two national hospitalist organizations in the last 3 years, and although I appreciated some of their good ideas in areas like communication and education and patient safety, the workloads were mostly just awful. My very first locum tenens position was with Sound Inpatient Physicians, a large hospitalist company. After 1 day of orientation with 16 patients to see, I accelerated to the usual patient load of 19 patients who I didn't know, many of them complex, during an outbreak of Norovirus, so I had to gown up for about a third of them, and got 2 new admissions. I rarely finished my 12 hour shifts in less than 14 hours, and barely had time to go to the bathroom. A well deserved high level of anxiety plagued me most moments of each day, as I rushed from acutely ill patient to acutely ill patient while receiving multiple pages on my beeper. The other company, Cogent, employed me as a locum tenens physician in a PeaceHealth facility in Washington. The doctors were great, though not happy. At all. The administrators were ninjas of awesome skills, but the workload was ridiculous. Just a few days ago the large German healthcare company Fresenius, which had bought a majority share of Sound Inpatient Physicians earlier this year, purchased Cogent as well. I would like to believe that good qualities of German healthcare will now perfuse these two companies, but I seriously doubt it. Fresenius also owns a large amount of American kidney dialysis capabilities, and hopes to combine their dialysis arm with their hospitalist business to better coordinate care. I will not simply assume that this will raise prices and result in overuse of their technology, since some good might come of it.  Fresenius did lose my vote of confidence, though, when they threatened to sue a researcher who wrote a scientific article that showed that a product they sell, hydroxyethyl starch, causes more death and kidney failure when used to resuscitate critically ill patients.

All this said, I should really admit to the fact that practicing as a hospitalist is a pretty good job. The patients are always interesting and they pay us well. There is no pay, though, that is adequate to make it OK to take care of too many patients at one time when doing so puts them at risk. Allowing doctors to have weighty input on work load, be it as part of a union or by being part of a well functioning independent hospitalist group, is vital to having good patient care and sustainable job satisfaction.

Wednesday, October 29, 2014

How does a non-cardiologist learn echocardiography? What's the deal with all of these ads for "123Sonography"?

Last Spring I got a junk e-mail offering a free "Echo survival course" from the University of Vienna, in Austria. I just had to go to a website, enter my e-mail, and I would get 4 free modules on basic echocardiography. Cool, I thought. Free knowledge! I've wanted to know the deepest secrets of echocardiography since I was a wee medical student a quarter of a century ago.

But why, one might ask, would it be relevant for me to know echocardiography? I'm not a cardiologist after all. Cardiologists are the people who read most of the heart ultrasounds, or echocardiograms, that are performed in the US. The usual routine is that someone like me, a general internist, or a family practitioner, orders an echocardiogram for a patient with a suspected heart problem. An ultrasonographer, a non-physician with expertise in performing ultrasounds of the heart, obtains images of the heart from various views, saves representative images, performs calculations of movements and sizes of structures and sends the whole file to a cardiologist who interprets the data. The cardiologist then produces a document with various abstruse and arcane abbreviations and corresponding values which also, thankfully, contains a summary paragraph which says if the heart looked normal or not. This whole process can take up to a week from start to finish. The patient or the patient's insurance company will be charged one or two thousand dollars and at the return visit the patient's physician will likely say something like, "it looked pretty good" or "one of the valves is a little leaky so I'm going to have you see the cardiologist" or "it wasn't too bad for someone your age."

I have been doing a rather quicker and more focused version of the echocardiogram as a bedside procedure for the last 3 years which frequently serves my purpose much better than the scenario I just described. It is possible to learn the basics of cardiac ultrasound through continuing medical education classes taught by emergency physicians, who have used bedside ultrasound for decades to more capably triage patients. Ultrasound machines have gotten smaller, more ubiquitous in hospitals (often you can find one nearby if you need it) and even affordable to own and carry around. Mine fits in a lab coat pocket and gets pretty good images from which I can make more informed decisions regarding my patients' diseases and appropriate treatment. It doesn't replace the full scale echocardiogram, except in cases where the details are unnecessary. I have done thousands of focused echocardiograms, reviewed them with experts when that was appropriate, compared the results to full scale echocardiograms when those were done, and know much more about the normal and abnormal heart than I did during the first two decades of my internal medicine practice when I relied primarily on my stethoscope. My bedside echocardiograms tell me if the heart is weak, and how weak, if there is fluid around the heart, whether there is evidence of problems with the blood vessels in the lungs and if there is damage from long term high blood pressure. I can identify the moment of death quite accurately when attending the dying and can see if an acute heart attack is causing a patient's chest pain or breathing problems. I can avoid giving medications that the patient's heart will not tolerate.

When it comes down to it, though, a good cardiologist is better at interpreting an echocardiogram than I am. I am not qualified to take the images and arcane numbers produced by an echo technician and produce a succinct but exhaustive summary. I am an internist, which is a fine and noble job, but not a cardiologist. So, since I look at hearts at the bedside all the time with a small ultrasound, I would like to know more about the fine points of echocardiography.

I have looked around for years for a good way to learn echocardiography and found that, for someone already in medical practice, it is pretty tricky to get echocardiography training. There used to be a guy, an echo technician by training, who ran courses for internists and anesthesiologists in performing and reading echocardiograms, but he got old and pretty much stopped doing it. He also really did not approve of bedside ultrasound. His courses were expensive and weeks long. After completing a course, to be credentialed to read echocardiograms required shadowing a cardiologist for a period of time and then reading echocardiograms which were over-read by a cardiologist. It would have been much easier if I had only known I would need this when I was still in training.

So it was very exciting when the offer of a free echocardiography course came to me in my e-mail. I followed the links and found that the 4 module free course was well taught and informative. I was still wary, but plunked down $757 for a full "master class" in echocardiography which was self paced and gave me 30 hours of training. I had 6 months to use the resource material. It was really pretty good. The faculty is all from the University of Vienna, in Austria, which is a real place, not like University of Phoenix or something. The faculty are real cardiologists, clearly interested in their material and their English is just fine, though spoken with Austrian accents. Both image acquisition and interpretation skills are taught and there is an emphasis on understanding not just the echocardiogram but the physiology of the heart and the underlying disease processes. There were 20 modules, with quizzes following each. The quizzes weren't particularly well written, but were detailed enough that it was necessary to really pay attention. There were exotic European spelling and grammar errors which did not distract from the material. I completed the course, got a certificate of completion and I feel pretty certain that I still am not qualified to read a full scale echocardiogram. I do, however, understand quite a few more subtleties than I did, and will continue to do bedside focused echocardiography with renewed appreciation.

I think this course is much more aimed at training physicians in resource poor settings where nearly all doctors are generalists, and having a physician who is able to read an echocardiogram, even without knowing the finer points, is life saving and really all that is available. Doctors in these situations will use a course like this to go from knowing close to nothing to being able to capably diagnose the majority of cardiac conditions, to their patients' great benefit. This was truly an educational niche that needed filling. The teachers also do in-person few-day courses in various locations, which I would love to attend at some point, not least of all to see what the student body is like. If they are fantastically successful (and I kind of think they are) perhaps there will be competition, and possibly even accessible education in the US. It has seemed like the lack of educational opportunities acted to protect a piece of lucrative turf claimed by cardiologists. The US is not primarily made up of urban areas, though, and large portions of our geography, like where I practice, for instance, are not served by cardiology clinics. I am glad a resource like this exists, even though online training in something that is hands-on cannot expect to fully cover the educational needs of practitioners.

Tuesday, October 21, 2014

Ebola!!! What about everything else? Influenza for instance.

Ebola virus has grabbed headlines since the epidemic started in West Africa nearly a year ago. The death toll is estimated at 4500 people, and the epidemic continues to spread. One person infected in Liberia returned to Texas with the disease and died, infecting maybe 2 healthcare workers.

Ebola is a nasty virus, surely, with a case fatality rate of 80%. Overall health and nutrition as well as living conditions have an effect on how sick a person gets with it. We have no good treatments, though antibodies and other biologically based treatments are being used and may be rapidly developed to combat the disease.

But have we forgotten influenza? Influenza is a nasty virus, with a few marginally helpful treatments. Its symptoms are fever, headache, cough, sore throat, runny nose, sometimes also heart failure, respiratory failure and brain dysfunction. It will likely kill tens of thousands of Americans this season, mostly the very old and very young, but also perfectly healthy people.

How about other globally relevant diseases? One and a half million people in the world die of AIDS yearly, the same number of tuberculosis, about 700,000 of malaria and more than the total of all three of these of malnutrition.

So we can't do very much about Ebola at this point, other than perhaps support the overseas programs that are trying to help the affected West African countries--Liberia, Sierra Leone, and Guinea. These
will probably have a more significant second wave of misery due to the economic devastation brought about by the epidemic. In the US Ebola is not likely to be an issue of significance. Perhaps learning to don and doff HAZMAT suits will benefit us in some as yet unknown way.  I am open to scorn and censure if I am wrong.

What should we do about influenza, since that routinely kills many thousands of us? Should we all get flu shots? I am not entirely convinced. I get them every year and induce my loved ones to do the same. The flu is utterly nasty, even if it doesn't kill you. It comes on like a freight train, with upper respiratory symptoms and intense misery. It hurts to move your eyes. You can barely think and you certainly can't work, at least after the first day when you go in and infect everyone else. The cough lasts weeks. You are infectious for at least a week. Old folks go from flu to pneumonia and sometimes die. Anything that will lessen my chance of getting it is welcome. Still, wholesale vaccination programs of healthy people do not statistically decrease hospitalizations or work loss, and 70 people must be vaccinated to avoid one case of influenza. This is per a large Cochrane Collaboration evaluation of many well designed studies.

As we enter influenza season probably the most important thing to do to avoid the disease is not to spread it. People with classic flu symptoms should not go to work or school, should not fly or travel in any public conveyance, and if they try, they should be sent home. Will we actually heed this advice? No, at least we haven't so far.

What about malaria and AIDS and tuberculosis and hunger? There are lots of great ideas out there, from newer medications to public health campaigns, promoting peace, helping women farmers, digging wells, curbing environmental abuses, that sort of thing. Committed and creative people are working all the time. We just aren't hearing much about them since apparently the most pressing global health issue right now is Ebola.

Thursday, October 16, 2014

Why do drugs cost so much? Confused and fuming about the unfairness of it all...

Drug prices are a difficult issue to write about because real data about the workings of pharmaceutical companies is very difficult to uncover. Still, last week I came face to face with something that seemed extremely not right and so I feel I should at least make some comment. It started when I prescribed a patient sumatriptan for her recently more frequent migraines. Her cost exceeded my wildest expectations.

Sumatriptan is a nearly magical medicine which was FDA approved in 1991 for treatment of acute migraines.* It is similar to the neurotransmitter serotonin and reduces inflammation of arteries in the brain which is associated with migraine headaches. It does other things as well, and may have a much more complex mechanism of action. Although it has some side effects, it works well for most people, can be given as an injection, pill or nasal spray and doesn't cause drowsiness, constipation or nausea like many other pain medications can. When sumatriptan was first released, under the brand name Imitrex, it was astoundingly expensive. I can't remember what it cost, but it was a bundle. People were willing to pay because it often saved them a trip to the emergency room or many hours of misery. Global sales of this drug top 1 billion dollars yearly. It has been approved as a generic since 2008.

Back to the patient. I prescribed thirty 50mg pills of sumatriptan, generic, via a national pharmacy chain, hoping that she would be able to treat her migraine at home, have leftover medications for future migraines and avoid a trip to the emergency department. The pharmacy told her it didn't take her insurance and that the prescription would cost her $550. She didn't have $550. I called the pharmacy, told them that I meant generic, not brand name, and they told me that they understood that and that the cost was $550. I called an independent pharmacist who I know well and asked if this drug was still ridiculously expensive or if there might have been some mistake. She said that her cost for 9 pills (they come in 9 packs for no particular reason) was $6.50. At her pharmacy she would charge an uninsured patient cost plus a dispensing fee. That would be about $30 for 30 pills. I got online to see how Canada charges for this drug, and"Canada Drugs" which sells medications at Canadian-ish prices to people in the US charges about $40 for nine 50 mg pills. A physician blogger, Leslie Ramirez MD, who is particularly interested in pharmaceutical costs reported that Costco, an American wholesale warehouse club had drug prices that were at pharmacy cost plus 14% dispensing fee. Costco's price for nine 50 mg sumatriptan pills was about $14. Although Costco requires a membership for most of their merchandise, anyone with a prescription can buy medications there. (Leslie Ramirez's website on cost comparison of drugs in the Chicago area disappeared around 2011 after this article was written about it in Forbes Magazine.)

So this person, my patient, with a raging headache, left her pharmacy empty handed because she was unable to afford a medication which had been marked up over $500 above cost, a medication which has been available as a generic for over half a decade.

I asked my pharmacist friend what this was all about, and she said that it had to do with "Average Wholesale Price" which is a number created by pharmaceutical people and distributors, originally intended to represent the actual cost of medications, allowing the price paid by insurance companies to be standardized. Since large pharmacies base their prices on AWP, that $550 for 30 sumatriptan tablets was probably pretty well set among retail pharmacies. Hiking up the AWP has various benefits to the many players in the pharmaceutical industry. Pharmacies can benefit because their costs are usually much lower than the published price (on average 14% lower, but clearly much much lower in some cases) meaning they make a handsome profit on some generic medications, wholesalers benefit since pharmacies want to buy drugs that they can sell at a substantial profit and pharmaceutical companies that produce brand name products benefit if generic drugs are kept artificially more expensive, since patients will often spring for the brand name original or be willing to pay high prices for new nearly identical brand name drugs.

It sounds like drugs are more expensive than they should be because there is collusion to overcharge for them. But it is not all based on Average Wholesale Price rigging. The cost of generic drugs to pharmacies has also gone up drastically. This article in the online version of the Wall Street Journal reports on rising costs of generic drugs, sometimes as much as 8000 percent (in the case of the antibiotic doxycycline), at least some of which was associated with rising costs to pharmacies. I wasn't able to find the actual data, but apparently lawmakers are "probing staggering price hikes" and sending letters of inquiry to the drug manufacturers asking why prices were raised, how much money they are making off of the affected drugs, who is responsible for price hikes and how costs for these medications compare overseas. The letter I linked was sent earlier this month and I don't see any information yet about a response.

Price hikes would seem to negatively affect patients, of course, but also insurance companies which pay at least some of these costs. In my review of the Affordable Care Act I haven't come across any provision that controls how much drug companies charge for their products. Old laws against price fixing do seem to apply to this sort of thing, but there is nothing new that says that Mylan, for instance, can't make a drug that cost $11 a bottle one month increase in cost to $400 a bottle a few months later. The insurance industry, though, is powerful enough that I would think they would balk at these price hikes. Medicare itself is not allowed to negotiate prices with drug companies but private insurance companies can. I'm not quite seeing what dynamic is at work here to keep them from refusing to pay for overpriced drugs.

Another thing I don't quite understand is why, when Costco charges cost plus 14% for medications consumers and insurance companies don't avail themselves of this option. I do love my corner drugstore, and realize that they survive partly by selling inexpensive drugs to insured patients for inflated prices, but I wonder why this continues to happen. It is truly valuable to have an independent pharmacist dispense medications and maintain a relationship with physician prescribers and patients, as happens in locally owned pharmacies, but we should find some way of paying for this service that is not arbitrary and subject to whimsical fluctuations.

The explanation that makes the most sense regarding drug prices is that the producers and purveyors of pharmaceuticals will charge as much for their wares as anyone will pay. Doctors cannot help because they are not aware, at the time of a patient encounter, what of their many drug options for a given condition is the best value. Unless lawmakers have the stomach to regulate the profits of big pharmaceutical companies, their most powerful lobbying entity, or enact legislation to allow market forces to reduce drug costs, prices of medications will continue to be subject to staggering increases.

*"Nearly magical " is a bit of an overstatement. It does work pretty well compared to other pain relievers, but only about a third of patients taking it have relief of their headache in an hour, and only 1 in five is headache free 24 hours later.

Sunday, October 5, 2014

Moxifloxacin for MRSA (methicillin resistant Staph aureus): Why is this not standard of care?

Moxifloxacin and MRSA. Why is this interesting?

Moxifloxacin
The drug company Bayer applied for a patent on yet another drug in the flouroquinolone category of antibiotics in 1989 and received approval by the FDA (Food and Drug Administration) in 1999 for Avelox, the brand name they gave to moxifloxacin. A Japanese company had discovered in the 1970's that adding a flourine to relatively ineffective antibiotics in the quinolone family, such as nalidixic acid, made them dramatically more active, thus creating flouroquinolones. That discovery led to the development of norfloxacin, then ciprofloxacin and levofloxacin which have become mainstays of antibiotic therapy. Ciprofloxacin is extremely useful for treating urinary infections and a variety of other serious infections including anthrax and traveler's diarrhea. Levofloxacin has become one of our drugs of choice for treating pneumonia and is especially useful because it achieves the same levels when given by mouth as it does by intravenous injection. Moxifloxacin hasn't really caught on to the same extent, even though it also is absorbed extremely well when taken orally and achieves particularly high levels in the lungs. It is also more effective for treating infections caused by gram positive organisms than ciprofloxacin or levofloxacin, including resistant Strep pneumoniae and Staph aureus. It is approved for treating skin, lung and abdominal infections caused by susceptible organisms and in some hospitals (like the VA, I hear) it is the least expensive flouroquinolone option due to deals with the manufacturer, so it is used more often. It just became available in a generic form in the US this year (2014).

MRSA
Methicillin resistant Staphylococcus aureus has grabbed headlines as it has become more common, both as a bug acquired in the hospital and now in the community, that is to say outside of hospitals. In some places resistant staph infections are now more common than the ones that are sensitive to the antibiotics we use most often. Staph aureus is usually quite a virulent bug, spreading aggressively in infected tissue and often seeding the bloodstream and even establishing itself on heart valves. It can cause particularly severe pneumonia, especially in already ill hospitalized patients and patients from nursing homes. It has become an especially big problem among intravenous drug abusers who are some of our sickest patients anyway, with coexisting issues like HIV infection and lack of adequate medical care. We have struggled to find antibiotics which work for MRSA and have turned to older and sometimes less effective antibiotics as well as newer and absurdly expensive ones.

A few weeks ago, while treating a patient with a MRSA infected wound, a colleague who is a wound care doctor suggested using moxifloxacin to treat her infection. I thought he was maybe just a little stupid, not to know that MRSA is usually resistant to flouroquinolones. I told him as much, except the stupid part and he told me that I was wrong, that he had just heard a talk at a wound care meeting and that moxifloxacin was good for MRSA. I checked the microbiology sensitivity sheet for my patient's MRSA to see what antibiotics it was sensitive to, and it was, indeed, resistant to levofloxacin and ciprofloxacin and our lab did not even test for moxifloxacin. I started poking around in the literature to find out what supported his claim that moxifloxacin was good for MRSA. There wasn't much, but there was an article that showed that, using MRSA from 12 patients who acquired it in the community or the hospital, moxifloxacin was more effective in killing the staph than trimethoprim sulfamethoxazole, linezolid or clindamycin. Another article showed that moxifloxacin was more effective than vancomycin, a standard treatment for MRSA, in treating MRSA in experimental biofilms, like the bacterial mats that characterize infected wounds. A third one looked at the effectiveness of vancomycin, ciprofloxacin and moxifloxacin at curing experimental heart infections (endocarditis) in rats and found that moxifloxacin was more effective than vancomycin and that ciprofloxacin didn't work at all. There were no human studies comparing moxifloxacin, head to head, with other standard antibiotics such as vancomycin for MRSA. So I guess he was right and I was wrong.

Standard of Care
This week's JAMA (Journal of the American Medical Association) featured an article entitled Clinical Management of Staphylococcus aureus Bacteremia, A Review, by Thomas L. Holland MD et al. The article concluded that vancomycin and daptomycin (a moderately toxic and very expensive new antibiotic for MRSA) are the first line antibiotic choices for MRSA bacteremia , that is infection found in the blood. This was based on 81 studies, none of which looked at moxifloxacin. The antibiotics studied were pretty much all the newer, recently released, very expensive and usually intravenous antibiotics. Studies involving humans are very expensive to perform, and funding is usually from pharmaceutical companies attempting to show that their drug works, which will make back the money they spend in research if all goes as planned. To give the article credit, the final conclusion was that well-designed studies to address the management of S. aureus bacteremia are needed.

Sepsis and Pneumonia
The standard of care in the hospitals where I have practiced is to use vancomycin (along with other broad spectrum antibiotic coverage) for patient who are seriously ill, in whom MRSA is suspected. Vancomycin is a difficult antibiotic to use, requiring measurement of levels to assure it is effective but not reaching toxic levels. It can cause kidney failure and hearing loss and if it is given quickly can cause "red man syndrome" which is what it sounds like, and quite disconcerting, though not deadly. Vancomycin must be given slowly which is a bit of an issue when a person is dying of rapidly progressive infection. But that's not actually the whole problem. We tend to use vancomycin when we suspect that there may be resistant staph in the lungs, but vancomycin actually has poor lung penetration and, even at standard doses, falls to what are probably ineffective levels during treatment.

It is often difficult to exclude pneumonia as a cause of serious infection in a patient who presents with sepsis, and the usual approach is to clobber them with broad spectrum antibiotics to cover whatever they might have. We try to get the antibiotics in to the patient as soon as humanly possible, ideally within an hour of arrival. Sometimes, however, it is difficult to get an intravenous line started and so a central venous catheter is placed, which must wait for a physician to do it, usually. Then there is a chest x-ray done to make sure that the line is in the right place and there is no lung collapse complicating the procedure. Then come the antibiotics. It can be agonizingly slow to get that first dose of life-saving antibiotics into a patient. Moxifloxacin can be given orally. "Here, take this." Bloop. Done. Or it can be given intravenously, if gut function is questionable, but quickly. Moxifloxacin covers most gram negative and gram positive organisms as well as atypical lung pathogens that cause serious infection including MRSA. Moxifloxacin dose is 400mg once daily and need not be adjusted for kidney or liver function.

So what is the catch?
What is wrong with moxifloxacin and why are we not using it more commonly? Moxifloxacin does not reach adequate levels in the urinary tract to treat urinary tract infections, which can be the cause of sepsis. But we can evaluate the urine quite quickly, in minutes actually, and adequately rule out urinary tract infection. Moxifloxacin can cause liver failure and serious skin rashes, but liver failure is extremely rare and all antibiotics cause skin reactions in some patients. It can cause tendons to rupture, similar to other fluoroquinolones, though that is also pretty rare. Moxifloxacin isn't cheap, somewhere between 5 and 20 dollars a pill. But that is compared to $8 a day plus high administration and monitoring costs for vancomycin and about $300 for daptomycin, plus administration costs. And moxifloxacin is now generic and produced by over 30 companies worldwide so its cost will likely become negligible. The biggest issue is that it hasn't been adequately studied in the setting of serious infection and isn't likely to be studied because it will make nobody money to do the expensive research. There is some evidence that staph aureus can become resistant to moxifloxacin in cultures, but, although this would certainly limit its usefulness, there has been no studies that I can find showing that this is actually relevant clinically, and some evidence implies that it might not be.

If, by some chance, it were to be studied and found to be superior to our present goofy standard of care, it would make some pretty profound changes in the way we do things. If moxifloxacin could be used to treat Staph aureus bacteremia then patients would not have to remain in the hospital or have outpatient intravenous antibiotics for 2 weeks, or 4-6 weeks in the case of complicated infections. It is incredibly inconvenient and dangerous to have patients on intravenous antibiotics for a prolonged amount of time. Intravenous drug abusers cannot be allowed to go home with an intravenous catheter in place because they will use it to inject drugs and the catheter will become infected. Those patients end up becoming fixtures in our hospital wards, often bored and disruptive, as they finish their prolonged treatments. When they leave against medical advice without completing their course of treatment a significant number will return, gravely ill, with a recurrence of their infection. The intravenous lines themselves, in addition to being very expensive, can cause infections and blood clots. Moxifloxacin achieves nearly identical levels when given orally as it does when given intravenously, so there would be no need for IV lines for 2-6 weeks.

In light of this information, what now?
I am not prepared to go against the standard of care at this point and use oral moxifloxacin for Staph aureus bacteremia, except in patients for whom intravenous therapy is impossible or likely to cause harm. I am, however, likely to use it for sepsis, when the urinary tract is not the source, in place of vancomycin plus other empiric gram negative and atypical organism coverage. I am also likely to choose it for treatment of wounds in which Staph aureus and gram negative organisms are identified or suspected. It is more than about time that adequate research was done to determine how we should use this drug for Staph, especially MRSA bacteremia.


Thursday, September 18, 2014

Emergency room doctors can safely use bedside ultrasound to diagnose kidney stones, saving billions of dollars and preventing some radiation induced cancers

I have been following the progress of bedside ultrasound (using ultrasound as a diagnostic tool during physical exam of patients) as it gets a foothold in standard medical practice. It has been part of my practice for almost 3 years now, during which time I have been repeatedly amazed by how helpful it is for guiding my clinical decisions. There is good research showing how useful it is for all sorts of applications, from heart problems to intestinal obstruction, but it is still slow to catch on.

An article came out just recently in the New England Journal of Medicine, which has a large circulation and should make a bit of a splash. This multi-center study looked at the option of having patients (excluding the very obese, pregnant and critically ill) with abdominal and flank pain suspected of having kidney stones evaluated first by emergency physicians with ultrasound of the kidneys and bladder before considering getting a CT scan. Normally a patient with suspected kidney stones (crampy pain in the back or abdomen, blood in the urine, suggestive history) will be referred for an abdominal and pelvic CT scan, which costs over $3000 and carries a significant amount of radiation exposure. In perfect circumstances performing the test and getting the results takes an hour, but it can end up taking several hours due to the usual delays. Sometimes patients with kidney stone type symptoms are referred by the emergency physician for an ultrasound by the radiology department, which takes about the same amount of time as the CT which takes the same amount of time, but costs a bit less and delivers no ionizing radiation. CT scans have beautiful pictures and can often find the kidney stone, if it's in there, and not finding the stone is strongly suggestive that the diagnosis of what is causing the pain must be sought elsewhere. Ultrasound can show if the kidney is blocked by showing lack of flow into the bladder or buildup of fluid in the kidney (hydronephrosis) but rarely actually visualizes the stone. This information, however, is adequate to make the diagnosis in most cases, when combined with a good clinical history, physical exam and lab tests.

It turns out that the bedside ultrasound exam done by emergency room docs (in this study they were from multiple medical centers including University of California at San Francisco, Cook County and Rush Medical Centers in Chicago, Group Health in Seattle and many more high quality locations) is adequate in cases of abdominal or flank pain as a first evaluation to rule in or out kidney stones. It is much more focused than an ultrasound performed by the radiology department and it only takes about 5 minutes or less to perform. Since it is done by the physician examining the patient it is also a time to take more history and do more general observation, which is always a good thing. About 40% of the patients initially evaluated this way got an official radiology ultrasound or CT scan which were felt to be necessary by the ER physician to clarify what was going on. About a million patients with kidney stones visit emergency rooms each year in the US and more than 10 times that many visit ER's with symptoms that sound a bit like kidney stones and have to be evaluated for them. If all of them got bedside ultrasound as the initial evaluation of their kidneys, my back-of-the-envelope calculations suggest that multiple billions of dollars could be saved on imaging costs and lives could potentially be saved due to reduced radiation exposure. The study showed no significant increase in complications in the patients first receiving bedside ultrasound. Actual cost savings were calculated, but not reported in the study (why?)

We can't just start doing this because not all ER doctors are yet comfortable performing and interpreting bedside ultrasound of the kidneys and bladder. But they could be. It is not hard. Pretty much anybody could learn to do this in maybe an hour and could certainly be competent after doing 50 exams. The implications of this are bigger than the article points out. When ER physicians start doing regular bedside (or "point of care" as it's sometimes termed) ultrasound they are going to get better at it. They will start to use ultrasound more and develop some pattern recognition skills that can't be predicted which will likely lead to more accurate diagnoses of other diseases, and possibly less dependence on ionizing and expensive radiation in the form of CT scans.

Unfortunately CT scans for abdominal pain in the emergency room  are an important source of revenue for both radiologists and hospitals which puts a little kink in the clear path toward adopting bedside ultrasound as a diagnostic procedure of choice. It's not clear what to do with this, because we could surely use the expertise of radiologists and radiology technicians in training physicians to be good bedside ultrasonographers and presently that would be a pretty big conflict of interest for them. Still, there is so much good stuff going on in the field of high tech ultrasound that is not in the scope of bedside ultrasound that radiologists and technicians could be kept gainfully occupied by doing things that other physicians can't and shouldn't do. In the journal of the American Institute of Ultrasound in Medicine there were several articles about amazing and technically challenging imaging applications that non-radiologists might be wise not to try. There were articles about ultrasound of the midbrain to evaluate Parkinson's Disease, ultrasound of the liver to look at severity of cirrhosis, ultrasound of children with intestinal intussusception (telescoping of the bowel) to follow the success of noninvasive treatments and detailed prenatal evaluations for conditions I didn't even know existed. Ultrasound to diagnose appendicitis has become nearly standard now, but is really hard to learn and ultrasonographers and radiologists do it well (some ER physicians do it well too, but it's far from an entry level skill.) Looking at the kidneys in 5 minutes in the ER is clearly fine for evaluating possible kidney stones. An abdominal ultrasound in the radiology department with their big powerful machine with the gorgeous images combined with the stunning command of anatomy of radiology professionals is a different and differently beneficial thing. This recent article may help move us as hospitalists, ER physicians and primary care providers toward doing more bedside ultrasound, which could be a very good thing. Perhaps more radiologists will find peace with that and can bring themselves to help teach other medical staff who need to learn how to do it.