Diabetic Neuropathy: Interpreting New Guidelines
Diabetic Neuropathy: Interpreting New Guidelines
Hi, my name is Dr. Charles Argoff. I'm Professor of Neurology and Director of the Comprehensive Pain Center at Albany Medical College and Albany Medical Center in Albany, New York. I just came back from the American Academy of Neurology (AAN) meeting in Hawaii. Many of you may know that at that meeting, the guidelines for the management of painful diabetic neuropathy, a tremendously collaborative effort involving the AAN and other societies, were released preliminarily. I was very proud to have been a member of the Quality Standards Subcommittee of the AAN for 6 years, and it's important to understand how these guidelines were developed. For those of you who don't know, the AAN has published a manual for how their guidelines are developed, including how proposals should be submitted for a guideline.
There is a very thorough, logical, and methodical way in which the literature is reviewed and rated on the basis of the quality of the evidence. There is also a very thorough and logical -- or at least well thought-out -- methodology by which the treatments are rated, in terms of "should be offered," "should probably not be considered," and so on. What level of evidence exists to make this a strong or less strong recommendation? This is unparalleled, because when I've been involved in other guideline processes (and you can see if you look at the guidelines that were published in the Mayo Clinic Proceedings in 2005 or if you look at some of the guidelines that have been developed by the neuropathic pain special-interest group of the International Association for the Study of Pain) that those guideline processes were different and did not follow this prescribed approach. This is not to say they weren't well thought out, but they are different. Some of you, if you are familiar with the AAN guidelines, might wonder how the only medication that came out with a level A recommendation was pregabalin.
Some of the medications that had been given the highest level of recommendation from other guidelines, such as controlled-release oxycodone, extended-release morphine, duloxetine, and tricyclic antidepressants, for example, were given level B recommendations because according to the AAN guideline process, if a study had a dropout rate, for any reason, of more than 20%, the quality of that study is demoted by 1 level. To have a level A recommendation, you must have 2 of the highest-class studies that showed the benefit of the treatment. Duloxetine, for example, had 3 class I studies, but 2 of them had discontinuation rates of more than 20%. Those 2 studies were dropped down to a lower level, and hence the recommendation also had to be less strong. In clinical practice, it matters at the end of the day that we have options with which to treat people.
The way that the AAN guidelines come across may not reflect the fact that you still have a wide variety of available medical treatments (as well as nonmedical treatments) when considering treatment for someone who has painful diabetic neuropathy. The take-home message should not be that a single medication is going to be effective for all people, but that we still must have a clinical context in using these guidelines. The truth is that we don't have many head-to-head studies; we don't have many guidelines that are able to tell us, with any degree of certainty, which medication is going to work for the individual in front of you on a given day in your office. Please take these guidelines for what they are. They are very strong and they have a great background and foundation in the way that they were developed, but at the end of the day, we still are clinicians and have to use our best clinical judgment and the best available evidence together. Thank you.
Hi, my name is Dr. Charles Argoff. I'm Professor of Neurology and Director of the Comprehensive Pain Center at Albany Medical College and Albany Medical Center in Albany, New York. I just came back from the American Academy of Neurology (AAN) meeting in Hawaii. Many of you may know that at that meeting, the guidelines for the management of painful diabetic neuropathy, a tremendously collaborative effort involving the AAN and other societies, were released preliminarily. I was very proud to have been a member of the Quality Standards Subcommittee of the AAN for 6 years, and it's important to understand how these guidelines were developed. For those of you who don't know, the AAN has published a manual for how their guidelines are developed, including how proposals should be submitted for a guideline.
There is a very thorough, logical, and methodical way in which the literature is reviewed and rated on the basis of the quality of the evidence. There is also a very thorough and logical -- or at least well thought-out -- methodology by which the treatments are rated, in terms of "should be offered," "should probably not be considered," and so on. What level of evidence exists to make this a strong or less strong recommendation? This is unparalleled, because when I've been involved in other guideline processes (and you can see if you look at the guidelines that were published in the Mayo Clinic Proceedings in 2005 or if you look at some of the guidelines that have been developed by the neuropathic pain special-interest group of the International Association for the Study of Pain) that those guideline processes were different and did not follow this prescribed approach. This is not to say they weren't well thought out, but they are different. Some of you, if you are familiar with the AAN guidelines, might wonder how the only medication that came out with a level A recommendation was pregabalin.
Some of the medications that had been given the highest level of recommendation from other guidelines, such as controlled-release oxycodone, extended-release morphine, duloxetine, and tricyclic antidepressants, for example, were given level B recommendations because according to the AAN guideline process, if a study had a dropout rate, for any reason, of more than 20%, the quality of that study is demoted by 1 level. To have a level A recommendation, you must have 2 of the highest-class studies that showed the benefit of the treatment. Duloxetine, for example, had 3 class I studies, but 2 of them had discontinuation rates of more than 20%. Those 2 studies were dropped down to a lower level, and hence the recommendation also had to be less strong. In clinical practice, it matters at the end of the day that we have options with which to treat people.
The way that the AAN guidelines come across may not reflect the fact that you still have a wide variety of available medical treatments (as well as nonmedical treatments) when considering treatment for someone who has painful diabetic neuropathy. The take-home message should not be that a single medication is going to be effective for all people, but that we still must have a clinical context in using these guidelines. The truth is that we don't have many head-to-head studies; we don't have many guidelines that are able to tell us, with any degree of certainty, which medication is going to work for the individual in front of you on a given day in your office. Please take these guidelines for what they are. They are very strong and they have a great background and foundation in the way that they were developed, but at the end of the day, we still are clinicians and have to use our best clinical judgment and the best available evidence together. Thank you.
Source...