COVID-19: More hydroxychloroquine data from France, more questions

A controversial study led by Didier Raoult, MD, PhD, on the combination of hydroxychloroquine and azithromycin in patients with COVID-19 was published March 20. The latest results from the same Marseille team, which involve 80 patients, were reported on March 27.

The investigators report a significant reduction in the viral load (83% patients had negative results on quantitative polymerase chain reaction testing at day 7, and 93% had negative results on day 8). There was a “clinical improvement compared to the natural progression.” One death occurred, and three patients were transferred to intensive care units.

If the data seem encouraging, the lack of a control arm in the study leaves clinicians perplexed, however.

Benjamin Davido, MD, an infectious disease specialist at Raymond-Poincaré Hospital in Garches, Paris, spoke in an interview about the implications of these new results.
 

What do you think about the new results presented by Prof. Raoult’s team? Do they confirm the effectiveness of hydroxychloroquine?

These results are complementary [to the original results] but don’t offer any new information or new statistical evidence. They are absolutely superimposable and say overall that, between 5 and 7 days [of treatment], very few patients shed the virus. But that is not the question that everyone is asking.

Even if we don’t necessarily have to conduct a randomized study, we should at least compare the treatment, either against another therapy – which could be hydroxychloroquine monotherapy, or just standard of care. It needed an authentic control arm.

To recruit 80 patients so quickly, the researchers probably took people with essentially ambulatory forms of the disease (there was a call for screening in the south of France) – therefore, by definition, less severe cases.

But to describe such a population of patients as going home and saying, “There were very few hospitalizations and it is going well,” does not in any way prove that the treatment reduces hospitalizations.
 

The argument for not having a control arm in this study was that it would be unethical. What do you think?

I agree with this argument when it comes to patients presenting with risk factors or who are starting to develop pneumonia.

But I don’t think this is the case at the beginning of the illness. Of course, you don’t want to wait to have severe disease or for the patient to be in intensive care to start treatment. In these cases, it is indeed very difficult to find a control arm.
 

In the ongoing Discovery trial, which involves more than 3,000 patients in Europe, including 800 in France, the patients have severe disease, and there are five treatment arms. Moreover, hydroxychloroquine is given without azithromycin. What do you think of this?

I think it’s a mistake. It will not answer the question of the effectiveness of hydroxychloroquine in COVID-19, especially as they’re not studying azithromycin in a situation where the compound seems necessary for the effectiveness of the treatment.

In addition, Discovery reinforces the notion of studying Kaletra [lopinavir/ritonavir, AbbVie] again, while Chinese researchers have shown that it does not work, the argument being that Kaletra was given too late (N Engl J Med. 2020 Mar 18. doi: 10.1056/NEJMoa2001282). Therefore, if we make the same mistakes from a methodological point of view, we will end up with negative results.
 

What should have been done in the Marseille study?

The question is: Are there more or fewer hospitalizations when we treat a homogeneous population straight away?

The answer could be very clear, as a control already exists! They are the patients that flow into our hospitals every day – ironically, these 80 patients [in the latest results, presented March 27] could be among the 80% who had a form similar to nasopharyngitis and resolved.

In this illness, we know that there are 80% spontaneous recoveries and 20% so-called severe forms. Therefore, with 80 patients, we are very underpowered. The cohort is too small for a disease in which 80% of the evolution is benign.

It would take 1,000 patients, and then, even without a control arm, we would have an answer.

On March 26, Didier Raoult’s team also announced having already treated 700 patients with hydroxychloroquine, with only one death. Therefore, if this cohort increases significantly in Marseille and we see that, on the map, there are fewer issues with patient flow and saturation in Marseille and that there are fewer patients in intensive care, you will have to wonder about the effect of hydroxychloroquine.

We will find out very quickly. If it really works, and they treat all the patients presenting at Timone Hospital, we will soon have the answer. It will be a real-life study.
 

What are the other studies on hydroxychloroquine that could give us answers?

There was a Chinese study that did not show a difference in effectiveness between hydroxychloroquine and placebo, but that was, again, conducted in only around 20 patients (J Zhejiang Univ (Med Sci). 2020. doi: 10.3785/j.issn.1008-9292.2020.03.03). This cohort is too small and tells us nothing; it cannot show anything. We must wait for the results of larger trials being conducted in China.

It surprises me that, today, we still do not have Italian data on the use of chloroquine-type drugs ... perhaps because they have a care pathway that means there is no outpatient treatment and that they arrive already with severe disease. The Italian recommendations nevertheless indicate the use of hydroxychloroquine.

I also wonder about the lack of studies of cohorts where, in retrospect, we could have followed people previously treated with hydroxychloroquine for chronic diseases (e.g., rheumatoid arthritis, lupus, etc.). Or we could identify all those patients on the health insurance system who had prescriptions.

That is how we discovered the AIDS epidemic in San Francisco: There was an increase in the number of prescriptions for trimethoprim/sulfamethoxazole (Bactrim) that corresponded to a population subtype (homosexual), and we realized that it was for a disease that resembled pneumocystosis. We discovered that via the drug!

If hydroxychloroquine is effective, it is enough to look at people who took it before the epidemic and see how they fared. And there, we do not need a control arm. This could give us some direction. The March 26 decree of the new Véran Law states that community pharmacies can dispense to patients with a previous prescription, so we can find these individuals.
 

Do you think that the lack of, or difficulty in setting up, studies on hydroxychloroquine in France is linked to decisions that are more political than scientific?

Perhaps the contaminated blood scandal still casts a shadow in France, and there is a great deal of anxiety over the fact that we are already in a crisis, and we do not want a second one. I can understand that.

However, just a week ago, access to this drug (and others with market approval that have been on the market for several years) was blocked in hospital central pharmacies, while we are the medical specialists with the authorization! It was unacceptable.

It was sorted out 48 hours ago: hydroxychloroquine is now available in the hospital, and to my knowledge, we no longer have a problem obtaining it.

It took time to alleviate doubts over the major health risks with this drug. [Officials] seemed almost like amateurs in their hesitation; I think they lacked foresight. We have forgotten that the treatment advocated by Prof. Didier Raoult is not chloroquine but rather hydroxychloroquine, and we know that the adverse effects are less [with hydroxychloroquine] than with chloroquine.
 

You yourself have treated patients with chloroquine, despite the risk for toxicity highlighted by some.

Initially, when we first started treating patients, we thought of chloroquine because we did not have data on hydroxychloroquine, only Chinese data with chloroquine. We therefore prescribed chloroquine several days before prescribing hydroxychloroquine.

The question of the toxicity of chloroquine was not unjustified, but I think we took far too much time to decide on the toxicity of hydroxychloroquine. Is [the latter] political? I don’t know. It was widely publicized, which amazes me for a drug that is already available.

On the other hand, everyone was talking at the same time about the toxicity of NSAIDs. ... One has the impression it was to create a diversion. I think there were double standards at play and a scapegoat was needed to gain some time and ask questions.

What is sure is that it is probably not for financial reasons, as hydroxychloroquine costs nothing. That’s to say there were probably pharmaceutical issues at stake for possible competitors of hydroxychloroquine; I do not want to get into this debate, and it doesn’t matter, as long as we have an answer.

Today, the only thing we have advanced on is the “safety” of hydroxychloroquine, the low risk to the general population. ... On the other hand, we have still not made any progress on the evidence of efficacy, compared with other treatments.

Personally, I really believe in hydroxychloroquine. It would nevertheless be a shame to think we had found the fountain of youth and realize, in 4 weeks, that we have the same number of deaths. That is the problem. I hope that we will soon have solid data so we do not waste time focusing solely on hydroxychloroquine.
 

What are the other avenues of research that grab your attention?

The Discovery trial will probably give an answer on remdesivir [GS-5734, Gilead], which is a direct antiviral and could be interesting. But there are other studies being conducted currently in China.

There is also favipiravir [T-705, Avigan, Toyama Chemical], which is an anti-influenza drug used in Japan, which could explain, in part, the control of the epidemic in that country. There are effects in vitro on coronavirus. But it is not at all studied in France at the moment. Therefore, we should not focus exclusively on hydroxychloroquine; we must keep a close eye on other molecules, in particular the “old” drugs, like this antiviral.

The study was supported by the Institut Hospitalo-Universitaire (IHU) Méditerranée Infection, the National Research Agency, under the Investissements d’avenir program, Région Provence Alpes Côte d’Azur, and European funding FEDER PRIMI. The authors have disclosed no relevant financial relationships.

This article first appeared on Medscape.com.

A controversial study led by Didier Raoult, MD, PhD, on the combination of hydroxychloroquine and azithromycin in patients with COVID-19 was published March 20. The latest results from the same Marseille team, which involve 80 patients, were reported on March 27.

The investigators report a significant reduction in the viral load (83% patients had negative results on quantitative polymerase chain reaction testing at day 7, and 93% had negative results on day 8). There was a “clinical improvement compared to the natural progression.” One death occurred, and three patients were transferred to intensive care units.

If the data seem encouraging, the lack of a control arm in the study leaves clinicians perplexed, however.

Benjamin Davido, MD, an infectious disease specialist at Raymond-Poincaré Hospital in Garches, Paris, spoke in an interview about the implications of these new results.
 

What do you think about the new results presented by Prof. Raoult’s team? Do they confirm the effectiveness of hydroxychloroquine?

These results are complementary [to the original results] but don’t offer any new information or new statistical evidence. They are absolutely superimposable and say overall that, between 5 and 7 days [of treatment], very few patients shed the virus. But that is not the question that everyone is asking.

Even if we don’t necessarily have to conduct a randomized study, we should at least compare the treatment, either against another therapy – which could be hydroxychloroquine monotherapy, or just standard of care. It needed an authentic control arm.

To recruit 80 patients so quickly, the researchers probably took people with essentially ambulatory forms of the disease (there was a call for screening in the south of France) – therefore, by definition, less severe cases.

But to describe such a population of patients as going home and saying, “There were very few hospitalizations and it is going well,” does not in any way prove that the treatment reduces hospitalizations.
 

The argument for not having a control arm in this study was that it would be unethical. What do you think?

I agree with this argument when it comes to patients presenting with risk factors or who are starting to develop pneumonia.

But I don’t think this is the case at the beginning of the illness. Of course, you don’t want to wait to have severe disease or for the patient to be in intensive care to start treatment. In these cases, it is indeed very difficult to find a control arm.
 

In the ongoing Discovery trial, which involves more than 3,000 patients in Europe, including 800 in France, the patients have severe disease, and there are five treatment arms. Moreover, hydroxychloroquine is given without azithromycin. What do you think of this?

I think it’s a mistake. It will not answer the question of the effectiveness of hydroxychloroquine in COVID-19, especially as they’re not studying azithromycin in a situation where the compound seems necessary for the effectiveness of the treatment.

In addition, Discovery reinforces the notion of studying Kaletra [lopinavir/ritonavir, AbbVie] again, while Chinese researchers have shown that it does not work, the argument being that Kaletra was given too late (N Engl J Med. 2020 Mar 18. doi: 10.1056/NEJMoa2001282). Therefore, if we make the same mistakes from a methodological point of view, we will end up with negative results.
 

What should have been done in the Marseille study?

The question is: Are there more or fewer hospitalizations when we treat a homogeneous population straight away?

The answer could be very clear, as a control already exists! They are the patients that flow into our hospitals every day – ironically, these 80 patients [in the latest results, presented March 27] could be among the 80% who had a form similar to nasopharyngitis and resolved.

In this illness, we know that there are 80% spontaneous recoveries and 20% so-called severe forms. Therefore, with 80 patients, we are very underpowered. The cohort is too small for a disease in which 80% of the evolution is benign.

It would take 1,000 patients, and then, even without a control arm, we would have an answer.

On March 26, Didier Raoult’s team also announced having already treated 700 patients with hydroxychloroquine, with only one death. Therefore, if this cohort increases significantly in Marseille and we see that, on the map, there are fewer issues with patient flow and saturation in Marseille and that there are fewer patients in intensive care, you will have to wonder about the effect of hydroxychloroquine.

We will find out very quickly. If it really works, and they treat all the patients presenting at Timone Hospital, we will soon have the answer. It will be a real-life study.
 

What are the other studies on hydroxychloroquine that could give us answers?

There was a Chinese study that did not show a difference in effectiveness between hydroxychloroquine and placebo, but that was, again, conducted in only around 20 patients (J Zhejiang Univ (Med Sci). 2020. doi: 10.3785/j.issn.1008-9292.2020.03.03). This cohort is too small and tells us nothing; it cannot show anything. We must wait for the results of larger trials being conducted in China.

It surprises me that, today, we still do not have Italian data on the use of chloroquine-type drugs ... perhaps because they have a care pathway that means there is no outpatient treatment and that they arrive already with severe disease. The Italian recommendations nevertheless indicate the use of hydroxychloroquine.

I also wonder about the lack of studies of cohorts where, in retrospect, we could have followed people previously treated with hydroxychloroquine for chronic diseases (e.g., rheumatoid arthritis, lupus, etc.). Or we could identify all those patients on the health insurance system who had prescriptions.

That is how we discovered the AIDS epidemic in San Francisco: There was an increase in the number of prescriptions for trimethoprim/sulfamethoxazole (Bactrim) that corresponded to a population subtype (homosexual), and we realized that it was for a disease that resembled pneumocystosis. We discovered that via the drug!

If hydroxychloroquine is effective, it is enough to look at people who took it before the epidemic and see how they fared. And there, we do not need a control arm. This could give us some direction. The March 26 decree of the new Véran Law states that community pharmacies can dispense to patients with a previous prescription, so we can find these individuals.
 

Do you think that the lack of, or difficulty in setting up, studies on hydroxychloroquine in France is linked to decisions that are more political than scientific?

Perhaps the contaminated blood scandal still casts a shadow in France, and there is a great deal of anxiety over the fact that we are already in a crisis, and we do not want a second one. I can understand that.

However, just a week ago, access to this drug (and others with market approval that have been on the market for several years) was blocked in hospital central pharmacies, while we are the medical specialists with the authorization! It was unacceptable.

It was sorted out 48 hours ago: hydroxychloroquine is now available in the hospital, and to my knowledge, we no longer have a problem obtaining it.

It took time to alleviate doubts over the major health risks with this drug. [Officials] seemed almost like amateurs in their hesitation; I think they lacked foresight. We have forgotten that the treatment advocated by Prof. Didier Raoult is not chloroquine but rather hydroxychloroquine, and we know that the adverse effects are less [with hydroxychloroquine] than with chloroquine.
 

You yourself have treated patients with chloroquine, despite the risk for toxicity highlighted by some.

Initially, when we first started treating patients, we thought of chloroquine because we did not have data on hydroxychloroquine, only Chinese data with chloroquine. We therefore prescribed chloroquine several days before prescribing hydroxychloroquine.

The question of the toxicity of chloroquine was not unjustified, but I think we took far too much time to decide on the toxicity of hydroxychloroquine. Is [the latter] political? I don’t know. It was widely publicized, which amazes me for a drug that is already available.

On the other hand, everyone was talking at the same time about the toxicity of NSAIDs. ... One has the impression it was to create a diversion. I think there were double standards at play and a scapegoat was needed to gain some time and ask questions.

What is sure is that it is probably not for financial reasons, as hydroxychloroquine costs nothing. That’s to say there were probably pharmaceutical issues at stake for possible competitors of hydroxychloroquine; I do not want to get into this debate, and it doesn’t matter, as long as we have an answer.

Today, the only thing we have advanced on is the “safety” of hydroxychloroquine, the low risk to the general population. ... On the other hand, we have still not made any progress on the evidence of efficacy, compared with other treatments.

Personally, I really believe in hydroxychloroquine. It would nevertheless be a shame to think we had found the fountain of youth and realize, in 4 weeks, that we have the same number of deaths. That is the problem. I hope that we will soon have solid data so we do not waste time focusing solely on hydroxychloroquine.
 

What are the other avenues of research that grab your attention?

The Discovery trial will probably give an answer on remdesivir [GS-5734, Gilead], which is a direct antiviral and could be interesting. But there are other studies being conducted currently in China.

There is also favipiravir [T-705, Avigan, Toyama Chemical], which is an anti-influenza drug used in Japan, which could explain, in part, the control of the epidemic in that country. There are effects in vitro on coronavirus. But it is not at all studied in France at the moment. Therefore, we should not focus exclusively on hydroxychloroquine; we must keep a close eye on other molecules, in particular the “old” drugs, like this antiviral.

The study was supported by the Institut Hospitalo-Universitaire (IHU) Méditerranée Infection, the National Research Agency, under the Investissements d’avenir program, Région Provence Alpes Côte d’Azur, and European funding FEDER PRIMI. The authors have disclosed no relevant financial relationships.

This article first appeared on Medscape.com.

A controversial study led by Didier Raoult, MD, PhD, on the combination of hydroxychloroquine and azithromycin in patients with COVID-19 was published March 20. The latest results from the same Marseille team, which involve 80 patients, were reported on March 27.

The investigators report a significant reduction in the viral load (83% patients had negative results on quantitative polymerase chain reaction testing at day 7, and 93% had negative results on day 8). There was a “clinical improvement compared to the natural progression.” One death occurred, and three patients were transferred to intensive care units.

If the data seem encouraging, the lack of a control arm in the study leaves clinicians perplexed, however.

Benjamin Davido, MD, an infectious disease specialist at Raymond-Poincaré Hospital in Garches, Paris, spoke in an interview about the implications of these new results.
 

What do you think about the new results presented by Prof. Raoult’s team? Do they confirm the effectiveness of hydroxychloroquine?

These results are complementary [to the original results] but don’t offer any new information or new statistical evidence. They are absolutely superimposable and say overall that, between 5 and 7 days [of treatment], very few patients shed the virus. But that is not the question that everyone is asking.

Even if we don’t necessarily have to conduct a randomized study, we should at least compare the treatment, either against another therapy – which could be hydroxychloroquine monotherapy, or just standard of care. It needed an authentic control arm.

To recruit 80 patients so quickly, the researchers probably took people with essentially ambulatory forms of the disease (there was a call for screening in the south of France) – therefore, by definition, less severe cases.

But to describe such a population of patients as going home and saying, “There were very few hospitalizations and it is going well,” does not in any way prove that the treatment reduces hospitalizations.
 

The argument for not having a control arm in this study was that it would be unethical. What do you think?

I agree with this argument when it comes to patients presenting with risk factors or who are starting to develop pneumonia.

But I don’t think this is the case at the beginning of the illness. Of course, you don’t want to wait to have severe disease or for the patient to be in intensive care to start treatment. In these cases, it is indeed very difficult to find a control arm.
 

In the ongoing Discovery trial, which involves more than 3,000 patients in Europe, including 800 in France, the patients have severe disease, and there are five treatment arms. Moreover, hydroxychloroquine is given without azithromycin. What do you think of this?

I think it’s a mistake. It will not answer the question of the effectiveness of hydroxychloroquine in COVID-19, especially as they’re not studying azithromycin in a situation where the compound seems necessary for the effectiveness of the treatment.

In addition, Discovery reinforces the notion of studying Kaletra [lopinavir/ritonavir, AbbVie] again, while Chinese researchers have shown that it does not work, the argument being that Kaletra was given too late (N Engl J Med. 2020 Mar 18. doi: 10.1056/NEJMoa2001282). Therefore, if we make the same mistakes from a methodological point of view, we will end up with negative results.
 

What should have been done in the Marseille study?

The question is: Are there more or fewer hospitalizations when we treat a homogeneous population straight away?

The answer could be very clear, as a control already exists! They are the patients that flow into our hospitals every day – ironically, these 80 patients [in the latest results, presented March 27] could be among the 80% who had a form similar to nasopharyngitis and resolved.

In this illness, we know that there are 80% spontaneous recoveries and 20% so-called severe forms. Therefore, with 80 patients, we are very underpowered. The cohort is too small for a disease in which 80% of the evolution is benign.

It would take 1,000 patients, and then, even without a control arm, we would have an answer.

On March 26, Didier Raoult’s team also announced having already treated 700 patients with hydroxychloroquine, with only one death. Therefore, if this cohort increases significantly in Marseille and we see that, on the map, there are fewer issues with patient flow and saturation in Marseille and that there are fewer patients in intensive care, you will have to wonder about the effect of hydroxychloroquine.

We will find out very quickly. If it really works, and they treat all the patients presenting at Timone Hospital, we will soon have the answer. It will be a real-life study.
 

What are the other studies on hydroxychloroquine that could give us answers?

There was a Chinese study that did not show a difference in effectiveness between hydroxychloroquine and placebo, but that was, again, conducted in only around 20 patients (J Zhejiang Univ (Med Sci). 2020. doi: 10.3785/j.issn.1008-9292.2020.03.03). This cohort is too small and tells us nothing; it cannot show anything. We must wait for the results of larger trials being conducted in China.

It surprises me that, today, we still do not have Italian data on the use of chloroquine-type drugs ... perhaps because they have a care pathway that means there is no outpatient treatment and that they arrive already with severe disease. The Italian recommendations nevertheless indicate the use of hydroxychloroquine.

I also wonder about the lack of studies of cohorts where, in retrospect, we could have followed people previously treated with hydroxychloroquine for chronic diseases (e.g., rheumatoid arthritis, lupus, etc.). Or we could identify all those patients on the health insurance system who had prescriptions.

That is how we discovered the AIDS epidemic in San Francisco: There was an increase in the number of prescriptions for trimethoprim/sulfamethoxazole (Bactrim) that corresponded to a population subtype (homosexual), and we realized that it was for a disease that resembled pneumocystosis. We discovered that via the drug!

If hydroxychloroquine is effective, it is enough to look at people who took it before the epidemic and see how they fared. And there, we do not need a control arm. This could give us some direction. The March 26 decree of the new Véran Law states that community pharmacies can dispense to patients with a previous prescription, so we can find these individuals.
 

Do you think that the lack of, or difficulty in setting up, studies on hydroxychloroquine in France is linked to decisions that are more political than scientific?

Perhaps the contaminated blood scandal still casts a shadow in France, and there is a great deal of anxiety over the fact that we are already in a crisis, and we do not want a second one. I can understand that.

However, just a week ago, access to this drug (and others with market approval that have been on the market for several years) was blocked in hospital central pharmacies, while we are the medical specialists with the authorization! It was unacceptable.

It was sorted out 48 hours ago: hydroxychloroquine is now available in the hospital, and to my knowledge, we no longer have a problem obtaining it.

It took time to alleviate doubts over the major health risks with this drug. [Officials] seemed almost like amateurs in their hesitation; I think they lacked foresight. We have forgotten that the treatment advocated by Prof. Didier Raoult is not chloroquine but rather hydroxychloroquine, and we know that the adverse effects are less [with hydroxychloroquine] than with chloroquine.
 

You yourself have treated patients with chloroquine, despite the risk for toxicity highlighted by some.

Initially, when we first started treating patients, we thought of chloroquine because we did not have data on hydroxychloroquine, only Chinese data with chloroquine. We therefore prescribed chloroquine several days before prescribing hydroxychloroquine.

The question of the toxicity of chloroquine was not unjustified, but I think we took far too much time to decide on the toxicity of hydroxychloroquine. Is [the latter] political? I don’t know. It was widely publicized, which amazes me for a drug that is already available.

On the other hand, everyone was talking at the same time about the toxicity of NSAIDs. ... One has the impression it was to create a diversion. I think there were double standards at play and a scapegoat was needed to gain some time and ask questions.

What is sure is that it is probably not for financial reasons, as hydroxychloroquine costs nothing. That’s to say there were probably pharmaceutical issues at stake for possible competitors of hydroxychloroquine; I do not want to get into this debate, and it doesn’t matter, as long as we have an answer.

Today, the only thing we have advanced on is the “safety” of hydroxychloroquine, the low risk to the general population. ... On the other hand, we have still not made any progress on the evidence of efficacy, compared with other treatments.

Personally, I really believe in hydroxychloroquine. It would nevertheless be a shame to think we had found the fountain of youth and realize, in 4 weeks, that we have the same number of deaths. That is the problem. I hope that we will soon have solid data so we do not waste time focusing solely on hydroxychloroquine.
 

What are the other avenues of research that grab your attention?

The Discovery trial will probably give an answer on remdesivir [GS-5734, Gilead], which is a direct antiviral and could be interesting. But there are other studies being conducted currently in China.

There is also favipiravir [T-705, Avigan, Toyama Chemical], which is an anti-influenza drug used in Japan, which could explain, in part, the control of the epidemic in that country. There are effects in vitro on coronavirus. But it is not at all studied in France at the moment. Therefore, we should not focus exclusively on hydroxychloroquine; we must keep a close eye on other molecules, in particular the “old” drugs, like this antiviral.

The study was supported by the Institut Hospitalo-Universitaire (IHU) Méditerranée Infection, the National Research Agency, under the Investissements d’avenir program, Région Provence Alpes Côte d’Azur, and European funding FEDER PRIMI. The authors have disclosed no relevant financial relationships.

This article first appeared on Medscape.com.