Search Annals: Advanced search

Podcast Transcript - February 19, 2008

[ Return to Podcast Home Page ]

Hello, and welcome to this weeks Annals of Internal Medicine Audio Summary for our February 19, 2008 issue. I’m Michael Berkwits, Deputy Editor at Annals.

We have another great issue for you this week, with articles on glucosamine for hip osteoarthritis; the apparent effects of universal health coverage on life expectancy in Taiwan; the pathophysiology of pulmonary arterial hypertension; and new CONSORT guidelines for conducting a reporting trials of non-pharmacologic interventions. Plus, I’ll summarize all the other articles in this week’s issue.

But first, here’s an in-depth summary of this week’s featured article.

Drugs Used To Prevent Contrast-Induced Nephropathy (Time: 00:36)

Our feature article this week is a systematic review and meta-analysis of drugs used to prevent contrast-induced nephropathy. Contrast nephropathy is generally defined in the literature an increase in serum creatinine greater than 0.5 mg/dL or 25% within 3 days of receiving intravascular contrast if there is no other identifiable cause. Risk factors for contrast nephropathy are pretty much what you’d expect: impaired renal function and comorbidities that predispose patients to impaired renal function, such as atherosclerotic disease, diabetes, and heart failure; but it is also thought to develop in up to 10% of patients with normal renal function, and risk for the complication is also influenced by amount and type of contrast agent used.

In this week’s article, lead author Aine Kelly and her colleagues from the University of Michigan performed a systematic review of randomized trials studying the effect of drugs used to prevent contrast nephropathy.

They found 41 trials conducted through November of 2006. Thirty of these studied N-acetylcysteine, which has a number of actions that might make it effective for preventing contrast nephropathy, but the authors also found 6 trials of theophylline, 3 of dopamine, and 1-2 trials each for fenoldopam, furosemide, iloprost, statins, vitamin C, bicarbonate, and mannitol. Patients in nearly all the trials also received background IV hydration, which was alone was the comparator agent; trial patients tended to be older than age 65 and have baseline renal insufficiency, and the trials tended to be of middling quality.

Of all the studied agents, only 3 were associated with a statistically significant reduction in risk for contrast nephropathy: oral N-acetylcysteine, oral ascorbic acid, and IV sodium bicarbonate. N-acetylcysteine, which reduced risk by about 1/3, clearly took the prize in the analysis because evidence for its effect came from so many trials; in contrast, the favorable effects of ascorbic acid and sodium bicarbonate were based on a single small trial each, with wide confidence intervals that weren’t too far from 1.0. Theophylline seemed to cut risk by about half, but its confidence interval just exceeded one.

Based on these findings, the authors conclude that the findings support the use of oral N-acetylcysteine for preventing contrast nephropathy given its efficacy, safety, wide availability, and low cost. But they acknowledge that the trials compared each agent to IV hydration, not to each other, so their claim for the superiority of N-acetylcysteine is indirect and requires evaluation in head-to-head trials until it can be fully justified. They also acknowledge that the outcome used in all the trials, a bump in serum creatinine, is really a surrogate measure; given that contrast-induced changes in creatinine usually resolve within days or a few weeks -- and that a bump in creatinine of point-5 for a patient with a baseline creatinine of 1.3 means something very different in terms of GFR than the same bump in a patients with a creatinine of 4 -- its clinical significance is uncertain….although a single small trial did demonstrate a reduction in length of stay for patients taking N-acetylcysteine. Finally, while there’s no reason not to believe the authors when they say that N-acetylcysteine is safe, this report doesn’t say anything about the relative incidence of adverse effects among the various agents, maybe in part because the trials were small and inconsistently reported safety outcomes themselves. Still, this is the kind of meta-analysis we editors like to see. If you look at the forest plot in Figure 2 of the print article you’ll see that the point estimates and confidence intervals for the effects of N-acetylcysteine fall below, above, and overlap no effect, so this pooled analysis really does seem to marshal the evidence towards a more precise estimate of effect, and so moves the literature forward.

So this is good stuff to know about, but as an internist I’m usually calling the diagnosis of contrast nephropathy after the fact, and I wasn’t sure what these data mean to me: should I be prescribing N-acetylcysteine to my patients? I called the lead author of the study and asked her that and a few other questions. Dr. Aine Kelly is a Clinical Assistant Professor of Radiology in the Division of Cardiothoracic Radiology at the VAMC and the University of Michigan in Ann Arbor, and she was good enough to respond.

Q: Dr. Kelly, thanks for talking to me.

A: Thank you.

Q: Let’s review the basics. What do we know about the mechanism for contrast-induced nephropathy?

A: Well, iodinated contrast agents, they’re thought to cause alterations in the hemodynamic mechanisms in the kidney. Specifically, they’re thought to cause increased vascular resistance, and a decrease in glomerular filtration rate. These changes are found within 24 hours, but the effects last a little longer. And, suspected mediators would include endothelin and adenosine, and some of the agents that have been used to prevent contrast-induced nephropathy are actually antagonists of these. But also implicated is altered mechanisms of modulation of prostaglandins and nitric oxide, so reductions in prostaglandins and nitric oxide have been associated with increased risk of contrast-induced nephropathy, and similarly, increased adenosine and endothelin are implicated in increased risk of nephropathy.

Q: Are there specific tests or procedures or CT protocols for example that require especially high dye loads and that put patients most at risk?

A: There are. There are procedures known as CT angiography whereby we inject larger volumes of contrast, though that is getting better now that we’ve multidetector CT. But in general, if we do a CT angiogram we’re injecting a fairly large dose and imaging at the arterial phase. Hence they’re called CT arteriograms. We do these in various parts of the body such as the circle of Willis in the brain, pulmonary arteries in the chest, or coronary arteries, and then in the abdomen we might do it for the aorta or the mesenteric vessels. So CT angiograms would be one culprit. Another would be interventional procedures which are quite prolonged, so examples would be coiling or embolization where it requires multiple injections of contrast to check patency and then to check afterwards that the aneurysm or the neoplasm, the circulation has been occluded. So any long or protracted interventional procedure would also bring with it large volumes of contrast agents. This would also include cardiac procedures as well, where there are multiple injections into individual coronary arteries and branches. So those would be the procedures that are implicated. Specific patient groups that would be susceptible would be patients with preexisting renal impairment and patients with diabetes who may also have renal impairment.

Q: Are patients with diabetes who don’t have apparent renal disease also at risk?

A: I believe they would be at a small increased risk based upon having diabetes and probably some microvascular changes in the kidneys, but it’s debatable whether we should approach these patients any differently. I think we should probably just use caution in any case.

Q: Now, there are iso-osmolar, low osmolar, and non-osmolar contrast agents that are supposed to decrease the risk for renal injury. Can you remind us of what the difference is between those categories and are those the radiology department’s call, so that ordering providers like internists don’t need to think about them or are there times that office providers should be reminding radiology departments on their imaging orders to pick an agent or class of contrast agent given an individual’s apparent risk?

A: There are two different things. Firstly, there are high or low osmolar. Most of the contrast agents contain iodine bound in a very large molecule so they have a very high osmolarity compared to blood osmolarity so they’re very viscous with a syrupy consistency. We have developed smaller molecules with less iodine concentration so these become less osmolar. All of the contrast agents to date still have an osmolarity higher than that of blood, but we are approaching osmolarity of blood with some of the very low osmolar agents. The lower osmolar contrast agents are preferable in patients who do have impaired renal function and will cause less side effects including less side effects, including less of a risk of contrast nephropathy. An example would be iopamadol. And contrast agents are also divided into ionic and non-ionic, based upon whether they have positive or negative ions. And in general, the ionic ones have more side-effect profile associated with them. Nowadays, we’re rarely using non-ionic so you can have non-ionic contrast agents which have different osmolarity. Some of them are low and some of them are very low or iso-osmolar. So we’re kind of trying to go towards non-ionic, low, or iso-osmolar agents. Regarding who should be looking at which contrast agents to be used as radiologists, we are prescribing an examination and also prescribing the contrast, which itself is a drug. So I believe anytime we prescribe we should be familiar with the drug, its indication, side effects and then relate that to the particular patient. It’s no harm for the clinician or the internist to be aware that we do use different contrast agents, and occasionally people will call me and ask, you know they’ve described a particular patient scenario and I will advise them, “please write this on the form.” So although we encourage the internist or the clinician to indicate this as a fail safe mechanism, I believe the illness is mainly upon the radiologist.

Q: N-acetylcysteine comes out the winner in your analysis, mainly because it’s inexpensive, and it’s been most studied, and most of the evidence points to a reduction in nephropathy. But it doesn’t seem like it’s used that often. I’ve never heard of anyone prescribing it for their patients. And there seems to be interest in finding additional preventive agents. For example, I saw several additional trials published in the past year or so since you’ve completed your review suggesting that sodium bicarb is effective. So I’m wondering if you can interpret for us why there’s continued uncertainty about N-acetylcysteine, and what seems like ferment in the literature about finding an agent to prevent nephropathy if N-acetylcysteine is available and effective.

A: N-acetylcysteine is the most studied agent, but if we were to look at all the individual studies we would find that a lot of them came down on the side of favoring N-acetylcysteine, and others were inconclusive, and others yet again would say that it increased the incidence of contrast nephropathy. So I think there was a lot of uncertainty. Hence, people were probably reluctant to adopt the agent for patients who were having contrast injected. So I think a meta-analysis or large studies like ours might help clarify some of that uncertainty. I think it may be a little while before people are ready to embrace it or take it on, which is surprising given that it is convenient, it’s cheap, it’s an oral agent, it has a very low side effect profile. But I do think there was a lot of confusion in the literature as more and more trials were being done. But I think we now have quite a large body of evidence to suggest that it is beneficial. There is still a lot of searching for other agents despite the fact that we’ve demonstrated this. But I guess the evidence is still accumulating, so people are searching for alternative agents. And you mentioned sodium bicarbonate as an alternative agent. I think when we look at other agents there’s less evidence out there and that would be a similar situation with theophylline and some of the other agents, dopamine; there are less trials and a lot of the trials again are giving conflicting results, so I think we just don’t have enough of a body of evidence for those agents to make a firm conclusion at this point.

Q: If providers, though, and radiologists are running an IV anyways to hydrate their patients, which was background therapy in almost all of the trials that were in your review, would it in theory be easier just to give somebody an ample of sodium bicarb if it’s proven to be effective, than to have to prescribe them two 600 mg tablets of N-acetylcysteine?

A: In some ways it would be convenient, I guess, if they already have an intravenous line inserted and they are receiving fluids. On the other hand, we would need to look at the side-effect profile of sodium bicarbonate. Although we are giving it for a short time, there may be side-effects associated with that, or there may interactions with other medications that the patient is taking.

Again, N-acetylcysteine has very low interaction rate in the limited dosage that we would use for nephropathy prevention. So in answer to your question, if

bicarbonate did show itself to be effective in preventing nephropathy, we would have to examine the side-effect profile and then look at the convenience. I guess if they have a line-up that’s fair enough.

Q: So would it be fair to say that it’s your opinion that this review settles the case and N-acetylcysteine should be used routinely as the agent of choice?

A: I would like to say so, but I think the terminology I would use would be to strengthen the case, because people will argue quite reasonably that demonstrating a decrease in nephropathy: does this translate to a clinical affect of decreasing the incidence of renal failure or decreasing the need for dialysis? So I think it strengthens the evidence, but I don’t think were 100% of the way there, because we are looking really at a surrogate marker.

Q: And so I wonder if you can tell listeners who you think should be managing this problem? Is contrast nephropathy something internists should be thinking about so that when they order their contrast studies and at risk patients they should be ordering N-acetylcysteine tablets the way that some providers write prescriptions for bowel preps when they arrange a colonoscopy, or is this something, again, that internists can trust radiology techs and physicians to identify and take care of?

A: I’d like to say that one side could be trusted over another. I think there’s always the danger that patients will fall between two stones. Ideally, I think both the referring physician and the performing radiologist should be responsible. I guess ultimately we are giving the drug, so we should really be responsible for ensuring that it’s handled appropriately. It’s a bit difficult because you do need to hydrate and start the medication before you receive the contrast, so I think it will fall on both parties to ensure that the patient receives the prescription and adequate instructions to commence the therapy before they get the actual contrast.

Q: And for listeners who are interested in taking on that responsibility for their patients, what are the correct instructions to give N-acetylcysteine to patients who are at risk for contrast nephropathy?

A: Different studies use different amounts, but in general they use 600 mg twice a day for two days. And that would commence on the day before the test. So if the contrast was to be given on Tuesday, the patient should take this on Monday and then again on Tuesday, twice a day on the day of the contrast agent. So it would be 600 mg orally twice a day for two days. That’s not something that’s widely known, so when I now talk to clinicians and they consult me, I will advise them of this dosage.

B: Dr. Kelly thanks for talking to me.

A: Thank you very much.

That was Dr. Aine Kelly, Clinical Assistant Professor of Radiology at the VAMC and the University of Michigan in Ann Arbor, and the lead author of this week’s meta-analysis, entitled “Effectiveness of Drug for Preventing Contrast-Induced Nephropathy.”

Other Articles In This Week’s Issue (Time: 17:03)

The lead article in this week’s issue is the one we covered a month ago about multidrug resistant MRSA infection in urban communities of men who have sex with men. The study received a fair amount of media coverage, and some pushback from the gay community, members of which thought they were unfairly stigmatized by some of the language and ideas in the authors’ press release and in the Annals report.

In this week’s issue we have an editorial, not yet finalized at the time of my last summary, that puts some of the more controversial issues raised by that report into some perspective. Authors Rachel Gorwitz, Scott Fridkin, and Kim Workowski, all of the CDC, write that the study does not clearly establish community-associated multidrug resistant MRSA as a sexually transmitted infection – which the authors never formally claimed – because studies of staph aureus infection in adults and children show that the buttocks and groin are common sites for infection, and the study couldn’t assess if sexual activity in general and genital, anal, or oral contact in particular, was implicated in transmission of the infection in the populations they studied. The editorialists make the point that incision and drainage, not antibiotics, are the primary therapy for uncomplicated MRSA skin infections, and that good hygiene and meticulous wound care, not antibiotics, are probably most important for transmitting infection. So they say that the study shouldn’t change providers’ clinical approach to patients with skin infections. Specifically, when antibiotics do seem indicated, clinicians’ should consult local epidemiology and susceptibility patterns when choosing a drug, and they should take the time to teach all patients with skin infection, regardless of their demographic, that wound care and containment are the key to preventing transmission to others.

Other articles in this week’s issue include a randomized trial suggesting that glucosamine is no better than placebo for treating the symptoms of hip arthritis, with an accompanying editorial that after carefully reviewing many of the methodologic pitfalls of studying glucosamine and osteoarthritis concludes that “the jury is still is out.”

We have a study of changes in life expectancy before and after the introduction of universal health insurance in Taiwan, suggesting that universal coverage did indeed lead to improvements in life expectancy for populations who had the worst baseline health, but that significant disparities in life expectancy remained for these groups even with improved access, possibly because of persistent unhealthy behaviors, such a smoking and betel quid chewing.

We have a contribution to our occasional Physiology in Medicine series, this one reviewing the role of transforming growth factor beta cytokines and receptor signals in vascular intimal proliferation and development of pulmonary arterial hypertension.

We’re publishing an extension of CONSORT guideline for conducting and reporting randomized trials of nonpharmacologic treatments such as surgery, implantable devices, and behavioral interventions.

And we have an interesting report in our Clinical Observations section about a case of anaphylactic shock in an amateur cyclist who the night before a competition injected himself with actovegin, a calf blood extract manufactured in Switzerland that is used by athletes to improve their performance by increasing blood oxygenation without raising their hematocrit.

On Being A Doctor: I Can’t Be Bothered (Time: 20:19)

Finally, our On Being A Doctor essay this week is entitled “I Can’t Be Bothered.” It’s read here by its author, Dr. Faith Fitzgerald.

“When I get a telephone call from a patient, it always begins with the same words: “I hate to bother you, but….”

A student calls me: “I know how busy you are, and I hate to bother you, but I’m really in trouble and need to talk to you….”

I discover on my early morning rounds that a patient, for whom I am the attending on the general medicine teaching service at my academic hospital, died several hours ago. The death was expected, but I had emphasized to the housestaff and nurses that I wanted to be notified of any significant changes (and I think death is very significant) in any of my patients. I ask both the nurse on shift and the night-float covering resident why they hadn’t called me, and their answer is the same: “It was late last night. I didn’t want to bother you.”

A secretary in my division of general medicine staggered into my academic office about 1 year ago complaining of faintness and headache. I had her lie down on the couch and took her blood pressure. It was 210/160. I got her to our emergency department and under immediate therapy for hypertensive crisis. When she was safely stabilized and in the ICU, I tried to call her personal physician, one of my former residents and an excellent internist, who was based at another hospital in our town. Because I didn’t have his direct office or pager number, I dialed his hospital’s general number and got the mechanized answering machine, which ran me through a series of numerical choices beginning with the information that if my concern was a life-threatening emergency I should hang up and call 911. An interminable series of inapplicable choices followed and ended with me on hold listening to tinny Musak interspersed with what can only be described as advertisements for the hospital. None of the choices offered the hope of connecting with a physician. I waited for a human being to talk to me.

Finally, a voice! “This is the operator. Can I help you?”

“Yes. This is Dr. Fitzgerald at the University Medical Center. I need to speak to one of your doctors about one of his patients.”

“What is your member number?”

“I don’t have a member number. I’m a doctor. I’ve just admitted your doctor’s patient to our ICU and want to tell him what happened.”

“Are you transferring the patient to our hospital?”

“No. I’m trying to tell her doctor what happened. She’s in our ICU.”

“What doctor?”

I told her. Some moments passed, then she said, “He’s scheduled to be in his clinic today.”

“Fine. Can you put me through to his clinic?”

“We can’t do that.”

“Why not?”

“We just don’t do that. He’s very busy. Would you like to talk to the telephone advice nurse?”

“I don’t need advice. I need to tell the doctor that his patient is very sick and in our ICU.”

“Why don’t you page him?”

“Fine—can you give me his pager number?”

“We don’t give those out. Sorry.”

“How can I reach him?”

“I can transfer you to the overhead page operator. One moment please.” Another 5 minutes passed, again with Musak and ads about health maintenance programs. Then came a real person’s voice on the line, a familiar voice.

“This is the operator. Can I help you?”

“This is Dr. Fitzgerald. Didn’t I just talk to you? You were going to put me through to the page operator.”

“Oh, yes. Sorry. I’ll try again.” Then more Musak, recorded repetitive messages about periodic mammography, lipid screening, and the virtues of exercise. Minutes passed slowly.

“Page operator. Can I help you?”

“Oh, glory! Yes.” I again explained who I was and why I wanted to find my patient’s doctor. “Can you call or page him for me? I was told he’s in his clinic.”

“I can page him overhead, but if he’s in his clinic he won’t hear it.”

“What?”

“The overhead system can be heard only in the hospital, not the clinics.”

“Why page him overhead if it won’t work?”

“It’s all I can do.”

“Can’t you beep or call him in his clinic?”

“We don’t do that. He’s very busy. Would you like to speak to the advice nurse?”

“Fine.”

Again, an endless interlude of Musak and ads; I began to slump. Then a voice again.

“Hospital operator. Can I help you?”

“It’s Dr. Fitzgerald again. Can you put me through to the advice nurse?”

“Just a moment.” Musak, messages, then a dial tone: They’d hung up on me.

I gave up, thinking that if ancient Rome had had such defenses, it never would have fallen to the barbarians. The patient was transferred to her own hospital several days later. I don’t know if the system ever told her doctor I had called.

Patients in my own hospital and clinic system tell me that their attempts to reach their doctors here are equally frustrating, and I’ve had similar difficulties in contacting my own hospital colleagues in their clinics and offices. I have, however—like cunning housestaff—experientially learned how to circumnavigate the system. Patients can’t.

I am a teacher-clinician in an academic center, so I also do administrative work for my medical school and hospital. An administrator’s assistant telephones me: “The Committee meets at 10 a.m., Tuesday.” There is in this phone notification no apology for bothering me, simply the assumption that I will be there. Committees are important.

I tell my patients, residents, and students that they should call me if they need me. They are not an interruption to my work; they are my work. In this sense, I can’t be “bothered” by them. But a system and a culture designed to protect doctors from their patients assume I am bothered, and so give that same impression to those trying to reach me. This really bothers me.”

That was Dr. Faith Fitzgerald, Professor of Medicine and Associate Dean of Bioethics and Humanities at UC-Davis Medical Center. She’s a Regent on the College’s Board of Regents, a frequent contributor to Annals, and the author of this week’s On Being A Doctor essay, entitled “I Can’t Be Bothered.”

Well, that’s it for today.

For full details of these and all other articles, consult your print journal, or go to www.annals.org.

We’ll say good bye today over stride piano master Thomas Wright Waller, performing as Fats Waller And His Rhythm. From the 1997 Giants of Jazz label compilation Lulu’s Back in Town, here’s Fats with a little help from Dr. Fitzgerald, from 1935, with Don’t Let It Bother You.

Our opening theme music is by Brian Poole and Kwesi Marles.

Thanks to listeners who have e-mailed me with your comments and encouragement - you know who you are. We’d love to hear from the rest of you; send comments, criticisms, feedback, and suggestions about these summaries and the journal, to podcast{at}annals.org.

Technical support for this summary was provided by Andrew Langman, Neil Kohl, and Beth Jenkinson.

Our next, regularly scheduled March 4, 2008 issue and its accompanying audio summary will be good ones, so don’t miss ‘em, and tell your colleagues how much fun we’re having, and they’re missing out on.

I’m Michael Berkwits; have a great 2 weeks, and thanks for listening.

[ Return to Podcast Home Page ]