Click here to listen to Dr. Merli discuss the new anticoagulants

Click here to listen to Howard Bremer discuss the long history of warfarin

Click here to listen to Dr. Merli discuss the new anticoagulants

Click here to listen to Howard Bremer discuss the long history of warfarin

Click here to listen to Dr. Merli discuss the new anticoagulants

Click here to listen to Howard Bremer discuss the long history of warfarin

My name is Joe Li, MD, SFHM. I am a hospitalist who works at Beth Israel Deaconess Medical Center in Boston. I have the privilege of serving as the SHM president for the upcoming year. I thank each of you for entrusting me with this important role.

Given the trust you have shared with me, I think it is only fair that I tell you a little bit about myself. For the “birthers” in the crowd, your fears have been realized. I was not born in this country; I was not even born in Panama. Not only were my parents immigrants—like many of you or your ancestors—I am also an immigrant to this country.

While I live now in Boston, I did not grow up there. Like many of you, I grew up in the middle part of the country. Like our immediate past president, Jeff Wiese, MD, SFHM, I spent my formative years in Oklahoma. My parents were not poor, but they were far from wealthy. Like most of you, I grew up in middle-class America.

Although I now have a teaching appointment at Harvard Medical School, neither my parents nor I ever paid tuition at a private school. Like many of you, I received my schooling at a public university.

I completed my undergraduate and medical school studies at the University of Oklahoma, and I moved to Boston for my post-graduate training. In fact, I was the first hospitalist at Beth Israel Deaconess Medical Center in 1998.

I joined the National Association of Inpatient Physicians (now SHM) as a charter member, and what I have learned from spending time with many of you over the years is that my story is not unique—there is tremendous diversity throughout SHM, but we are held together by the shared vision of improving care for our hospitalized patients.

For those of you new to SHM, I want to make it clear that SHM is not run by the president but by its collective members. Yes, we do have an organizational structure, with an elected board of directors and an elected executive committee made up of the president-elect, the president, the immediate past president, and the CEO. This past year, I served as the president-elect and had the wonderful opportunity of working with Dr. Wiese, our immediate past president, Scott Flanders, MD, SFHM, and our CEO, Larry Wellikson, MD, SFHM.

I can tell you firsthand about the tireless work that each of these physicians puts in to serve our patients and our profession. Each is an incredible leader, and I thank them, as well as our board of directors and SHM staff members, for all the work they put in to keep our organization running smoothly. While their work is not obvious day to day, there are many others who serve our organization in an invisible role. These include our committee leaders and committee members. I also thank you for all of your hard work for SHM and for our profession.

Despite all the hard work that has been done, there is still so much to do. I am personally asking each one of you to serve HM in your own way. Being an SHM member and attending SHM meetings are ways of serving, but I challenge each of you to do more. For some of you, it could be setting an expectation that all of your hospitalists join SHM and attend SHM meetings. For others, it could be helping to organize and lead your local chapters.

There is a role for each of us in HM, and I believe strongly that if we are to improve the care of our patients, each of us must take responsibility by serving our profession. There is no role too small. Each one of us must lead in our own way.

I look forward to the opportunity this year of speaking with each of you, not only as I travel the country to the various SHM meetings and chapter events, but also through this monthly column. I hope to share with you my observations of the happenings throughout hospital medicine. I expect to see and hear remarkable work being done by hospitalists across the country in our continued effort to bring increasing healthcare value to our patients. TH

Dr. Li is president of SHM, associate professor of medicine at Harvard Medical School in Boston, and director of the hospital medicine program and associate chief of the division of general medicine and primary care at Beth Israel Deaconess Medical Center.

My name is Joe Li, MD, SFHM. I am a hospitalist who works at Beth Israel Deaconess Medical Center in Boston. I have the privilege of serving as the SHM president for the upcoming year. I thank each of you for entrusting me with this important role.

Given the trust you have shared with me, I think it is only fair that I tell you a little bit about myself. For the “birthers” in the crowd, your fears have been realized. I was not born in this country; I was not even born in Panama. Not only were my parents immigrants—like many of you or your ancestors—I am also an immigrant to this country.

While I live now in Boston, I did not grow up there. Like many of you, I grew up in the middle part of the country. Like our immediate past president, Jeff Wiese, MD, SFHM, I spent my formative years in Oklahoma. My parents were not poor, but they were far from wealthy. Like most of you, I grew up in middle-class America.

Although I now have a teaching appointment at Harvard Medical School, neither my parents nor I ever paid tuition at a private school. Like many of you, I received my schooling at a public university.

I completed my undergraduate and medical school studies at the University of Oklahoma, and I moved to Boston for my post-graduate training. In fact, I was the first hospitalist at Beth Israel Deaconess Medical Center in 1998.

I joined the National Association of Inpatient Physicians (now SHM) as a charter member, and what I have learned from spending time with many of you over the years is that my story is not unique—there is tremendous diversity throughout SHM, but we are held together by the shared vision of improving care for our hospitalized patients.

For those of you new to SHM, I want to make it clear that SHM is not run by the president but by its collective members. Yes, we do have an organizational structure, with an elected board of directors and an elected executive committee made up of the president-elect, the president, the immediate past president, and the CEO. This past year, I served as the president-elect and had the wonderful opportunity of working with Dr. Wiese, our immediate past president, Scott Flanders, MD, SFHM, and our CEO, Larry Wellikson, MD, SFHM.

I can tell you firsthand about the tireless work that each of these physicians puts in to serve our patients and our profession. Each is an incredible leader, and I thank them, as well as our board of directors and SHM staff members, for all the work they put in to keep our organization running smoothly. While their work is not obvious day to day, there are many others who serve our organization in an invisible role. These include our committee leaders and committee members. I also thank you for all of your hard work for SHM and for our profession.

Despite all the hard work that has been done, there is still so much to do. I am personally asking each one of you to serve HM in your own way. Being an SHM member and attending SHM meetings are ways of serving, but I challenge each of you to do more. For some of you, it could be setting an expectation that all of your hospitalists join SHM and attend SHM meetings. For others, it could be helping to organize and lead your local chapters.

There is a role for each of us in HM, and I believe strongly that if we are to improve the care of our patients, each of us must take responsibility by serving our profession. There is no role too small. Each one of us must lead in our own way.

I look forward to the opportunity this year of speaking with each of you, not only as I travel the country to the various SHM meetings and chapter events, but also through this monthly column. I hope to share with you my observations of the happenings throughout hospital medicine. I expect to see and hear remarkable work being done by hospitalists across the country in our continued effort to bring increasing healthcare value to our patients. TH

Dr. Li is president of SHM, associate professor of medicine at Harvard Medical School in Boston, and director of the hospital medicine program and associate chief of the division of general medicine and primary care at Beth Israel Deaconess Medical Center.

My name is Joe Li, MD, SFHM. I am a hospitalist who works at Beth Israel Deaconess Medical Center in Boston. I have the privilege of serving as the SHM president for the upcoming year. I thank each of you for entrusting me with this important role.

Given the trust you have shared with me, I think it is only fair that I tell you a little bit about myself. For the “birthers” in the crowd, your fears have been realized. I was not born in this country; I was not even born in Panama. Not only were my parents immigrants—like many of you or your ancestors—I am also an immigrant to this country.

While I live now in Boston, I did not grow up there. Like many of you, I grew up in the middle part of the country. Like our immediate past president, Jeff Wiese, MD, SFHM, I spent my formative years in Oklahoma. My parents were not poor, but they were far from wealthy. Like most of you, I grew up in middle-class America.

Although I now have a teaching appointment at Harvard Medical School, neither my parents nor I ever paid tuition at a private school. Like many of you, I received my schooling at a public university.

I completed my undergraduate and medical school studies at the University of Oklahoma, and I moved to Boston for my post-graduate training. In fact, I was the first hospitalist at Beth Israel Deaconess Medical Center in 1998.

I joined the National Association of Inpatient Physicians (now SHM) as a charter member, and what I have learned from spending time with many of you over the years is that my story is not unique—there is tremendous diversity throughout SHM, but we are held together by the shared vision of improving care for our hospitalized patients.

For those of you new to SHM, I want to make it clear that SHM is not run by the president but by its collective members. Yes, we do have an organizational structure, with an elected board of directors and an elected executive committee made up of the president-elect, the president, the immediate past president, and the CEO. This past year, I served as the president-elect and had the wonderful opportunity of working with Dr. Wiese, our immediate past president, Scott Flanders, MD, SFHM, and our CEO, Larry Wellikson, MD, SFHM.

I can tell you firsthand about the tireless work that each of these physicians puts in to serve our patients and our profession. Each is an incredible leader, and I thank them, as well as our board of directors and SHM staff members, for all the work they put in to keep our organization running smoothly. While their work is not obvious day to day, there are many others who serve our organization in an invisible role. These include our committee leaders and committee members. I also thank you for all of your hard work for SHM and for our profession.

Despite all the hard work that has been done, there is still so much to do. I am personally asking each one of you to serve HM in your own way. Being an SHM member and attending SHM meetings are ways of serving, but I challenge each of you to do more. For some of you, it could be setting an expectation that all of your hospitalists join SHM and attend SHM meetings. For others, it could be helping to organize and lead your local chapters.

There is a role for each of us in HM, and I believe strongly that if we are to improve the care of our patients, each of us must take responsibility by serving our profession. There is no role too small. Each one of us must lead in our own way.

I look forward to the opportunity this year of speaking with each of you, not only as I travel the country to the various SHM meetings and chapter events, but also through this monthly column. I hope to share with you my observations of the happenings throughout hospital medicine. I expect to see and hear remarkable work being done by hospitalists across the country in our continued effort to bring increasing healthcare value to our patients. TH

Dr. Li is president of SHM, associate professor of medicine at Harvard Medical School in Boston, and director of the hospital medicine program and associate chief of the division of general medicine and primary care at Beth Israel Deaconess Medical Center.

I stare; a brimming audience stares back. Two eyeballs battling thousands. Slightly uncomfortable, I shift my weight, trying to hide behind the glass podium. Two microphones snake out of the podium slithering together inches from my mouth. The attendees squirm, sidle to the edge of their seats, restless to depart. HM11 is trying to close; only I stand in its way.

A Herculean task lies before me—summarize the annual meeting in a 10-minute wrap-up session titled “What We’ve Learned.” How do you summarize four days, eight pre-courses, nine breakout tracks, and more than 100 presentations in a few minutes? A bead of forehead sweat forms; I clear my throat. Memories of the past few days slide-show across my mind. It occurs to me that the essence of the meeting is not contained in the data, the information, or the PowerPoint slides that were presented. Rather, the story of HM11 is best told through its quotes.

Patient Caps: Your Grandmother and Professionalism

“I worry about patient caps because the next patient could be your grandmother.”

—Joe Li, MD, SFHM, new president of SHM

“Patient caps are the greatest threat to the professionalism of the field.”

—Rob Bessler, MD, CEO, Sound Inpatient Physicians

These two quotes from the opening plenary focused on the 2011 HM compensation and productivity survey particularly stuck out. The most noteworthy exchange came when Drs. Li and Bressler commented on the appropriate number of daily encounters for a hospitalist. The quotes highlight two important points about patient volume, especially in the wake of the training regulations that limit the number of resident physician encounters, which can engender a “cap mentality.” One is that it matters; there is a safe amount of encounters that shouldn’t routinely be breached. Two is that in the heat of the moment, Patient 19 is as important as Patient 11 and should be treated as such. Contingency plans are essential, but our field is built on the moorings of professionalism—the focus needs to be on humans, not numbers.

Hospitalist Compensation: Increasing but Not as Juicy

“It’s not going to get less anytime soon.”

—Dr. Bressler

In commenting on the data showing that the average community hospitalist makes about $220,000 annually—a 3% increase over last year—while producing around 4,000 work RVUs—flat over last year—and that their academic counterparts made $173,000 on about 3,400 wRVUs, Dr. Bressler opined that the laws of supply and demand would dictate that salaries would continue to rise for the near term. Although I agree with Dr. Bressler, my guess is that future salary increases will be driven more by quality than quantity (more to follow below).

“Juice-to-squeeze ratio”

—John Nelson, MD, MHM, SHM cofounder

Dr. Nelson highlighted interesting data showing that the average pay per wRVU was approximately $54. However, he noted that the compensation per wRVU tends to peak at a certain level, after which compensation per wRVU falls. In other words, after, say, 4,000 wRVUs, the amount of compensation per wRVU diminishes such that seeing more patients benefits an individual hospitalist less. That is, lots of squeeze, little juice at the high end.

Reform: Variety, Change, and Waste

“Variety is about choice; change is not.”

—Cecil Wilson, MD, AMA president

“You won’t have many more conferences where you start by talking about work RVUs.”

—Bob Kocher, MD, former special assistant to President Obama

The highlight of the conference for me was Dr. Kocher’s behind-the-scenes look at what was a very publicly muddy event—the passage of ACA. Coming from a D.C. insider, this under-the-covers peek at the machinations that went into passing the healthcare reform bill was fascinating.

The key message, summarized in this comment referring to the opening plenary about hospitalist compensation and productivity, was that the future is quality and the future is now. In the very near future, we will be measured and paid based on our ability to effect quality outcomes, not patient encounters. The message was simple: It’s about quality, not quantity.

“It costs $7.50 for a healthcare transaction, versus 2 cents for a VISA transaction.”

—Dr. Kocher

A statistic I had not heard before, this quote sums up one of the major problems with American healthcare: waste. The $7.50 transaction he was referring to was the amount of money it takes to file a healthcare claim. We certainly feel it in the challenges of documentation, billing, and denials, but the system feels it in terms of high cost of capturing what in many ways should be as simple as swiping your credit card at Starbucks.

Duty-Hour Restrictions: Harbinger of The Future?

“Don’t begrudge the ACGME—begrudge us.”

—Jeff Wiese, MD, SFHM, SHM past president

In a much-anticipated session on the impact of the new ACGME residency work-hour rules commencing in July—notably limiting intern (16-hour) and resident (28-hour) shift duration—Dr. Wiese aptly pointed out that a lot of the angst toward residency work environment regulation could have been avoided if physician leadership had better reacted to the issues of sleep deprivation and resident fatigue following Libby Zion’s death in 1984. Had we put our energy into improving work conditions rather than debate the impact of sleep deprivation on the outcome in this one case, we might be in a different place today.

I couldn’t help but wonder if the message here could also be applied to society’s push for higher quality, lower cost, and safer care. Either we regulate ourselves or someone else will. In other words, we need to embrace quality and safety, or it will be thrust upon us from external sources in ways we might not like.

A Mariner Calls

“I love you, Papi. Come home and take some baseball cuts.”

—Greyson Glasheen, future Major League Baseball shortstop

I wrote in a column leading up to the annual meeting (see “Annual Meeting Mariner,” April 2011, p. 45) that I was looking forward to the meeting because it was a professional mariner of sorts, a way for me to refresh, reset, and reinvigorate. Indeed, reflecting from the podium, it had been a fantastic meeting that served its purpose well. I had learned a ton, caught up with colleagues I hadn’t seen since the last meeting, saw old medical school friends, and met future old friends. I’d led a committee, given a talk, presented a poster, met up with a mentor, and had a reunion with past attendees of the Academic Hospitalist Academy.

Yet I was ready to get back to normalcy. On the last night of the meeting, I was therefore drawn by a different, more personal mariner—this time, a 14-second voicemail message from a 3-year-old boy waiting impatiently for Dad to come home, to make him his center, to simply play a little tee ball in the backyard. TH

Dr. Glasheen is associate professor of medicine at the University of Colorado at Denver, where he serves as director of the Hospital Medicine Program and the Hospitalist Training Program, and as associate program director of the Internal Medicine Residency Program.

I stare; a brimming audience stares back. Two eyeballs battling thousands. Slightly uncomfortable, I shift my weight, trying to hide behind the glass podium. Two microphones snake out of the podium slithering together inches from my mouth. The attendees squirm, sidle to the edge of their seats, restless to depart. HM11 is trying to close; only I stand in its way.

A Herculean task lies before me—summarize the annual meeting in a 10-minute wrap-up session titled “What We’ve Learned.” How do you summarize four days, eight pre-courses, nine breakout tracks, and more than 100 presentations in a few minutes? A bead of forehead sweat forms; I clear my throat. Memories of the past few days slide-show across my mind. It occurs to me that the essence of the meeting is not contained in the data, the information, or the PowerPoint slides that were presented. Rather, the story of HM11 is best told through its quotes.

Patient Caps: Your Grandmother and Professionalism

“I worry about patient caps because the next patient could be your grandmother.”

—Joe Li, MD, SFHM, new president of SHM

“Patient caps are the greatest threat to the professionalism of the field.”

—Rob Bessler, MD, CEO, Sound Inpatient Physicians

These two quotes from the opening plenary focused on the 2011 HM compensation and productivity survey particularly stuck out. The most noteworthy exchange came when Drs. Li and Bressler commented on the appropriate number of daily encounters for a hospitalist. The quotes highlight two important points about patient volume, especially in the wake of the training regulations that limit the number of resident physician encounters, which can engender a “cap mentality.” One is that it matters; there is a safe amount of encounters that shouldn’t routinely be breached. Two is that in the heat of the moment, Patient 19 is as important as Patient 11 and should be treated as such. Contingency plans are essential, but our field is built on the moorings of professionalism—the focus needs to be on humans, not numbers.

Hospitalist Compensation: Increasing but Not as Juicy

“It’s not going to get less anytime soon.”

—Dr. Bressler

In commenting on the data showing that the average community hospitalist makes about $220,000 annually—a 3% increase over last year—while producing around 4,000 work RVUs—flat over last year—and that their academic counterparts made $173,000 on about 3,400 wRVUs, Dr. Bressler opined that the laws of supply and demand would dictate that salaries would continue to rise for the near term. Although I agree with Dr. Bressler, my guess is that future salary increases will be driven more by quality than quantity (more to follow below).

“Juice-to-squeeze ratio”

—John Nelson, MD, MHM, SHM cofounder

Dr. Nelson highlighted interesting data showing that the average pay per wRVU was approximately $54. However, he noted that the compensation per wRVU tends to peak at a certain level, after which compensation per wRVU falls. In other words, after, say, 4,000 wRVUs, the amount of compensation per wRVU diminishes such that seeing more patients benefits an individual hospitalist less. That is, lots of squeeze, little juice at the high end.

Reform: Variety, Change, and Waste

“Variety is about choice; change is not.”

—Cecil Wilson, MD, AMA president

“You won’t have many more conferences where you start by talking about work RVUs.”

—Bob Kocher, MD, former special assistant to President Obama

The highlight of the conference for me was Dr. Kocher’s behind-the-scenes look at what was a very publicly muddy event—the passage of ACA. Coming from a D.C. insider, this under-the-covers peek at the machinations that went into passing the healthcare reform bill was fascinating.

The key message, summarized in this comment referring to the opening plenary about hospitalist compensation and productivity, was that the future is quality and the future is now. In the very near future, we will be measured and paid based on our ability to effect quality outcomes, not patient encounters. The message was simple: It’s about quality, not quantity.

“It costs $7.50 for a healthcare transaction, versus 2 cents for a VISA transaction.”

—Dr. Kocher

A statistic I had not heard before, this quote sums up one of the major problems with American healthcare: waste. The $7.50 transaction he was referring to was the amount of money it takes to file a healthcare claim. We certainly feel it in the challenges of documentation, billing, and denials, but the system feels it in terms of high cost of capturing what in many ways should be as simple as swiping your credit card at Starbucks.

Duty-Hour Restrictions: Harbinger of The Future?

“Don’t begrudge the ACGME—begrudge us.”

—Jeff Wiese, MD, SFHM, SHM past president

In a much-anticipated session on the impact of the new ACGME residency work-hour rules commencing in July—notably limiting intern (16-hour) and resident (28-hour) shift duration—Dr. Wiese aptly pointed out that a lot of the angst toward residency work environment regulation could have been avoided if physician leadership had better reacted to the issues of sleep deprivation and resident fatigue following Libby Zion’s death in 1984. Had we put our energy into improving work conditions rather than debate the impact of sleep deprivation on the outcome in this one case, we might be in a different place today.

I couldn’t help but wonder if the message here could also be applied to society’s push for higher quality, lower cost, and safer care. Either we regulate ourselves or someone else will. In other words, we need to embrace quality and safety, or it will be thrust upon us from external sources in ways we might not like.

A Mariner Calls

“I love you, Papi. Come home and take some baseball cuts.”

—Greyson Glasheen, future Major League Baseball shortstop

I wrote in a column leading up to the annual meeting (see “Annual Meeting Mariner,” April 2011, p. 45) that I was looking forward to the meeting because it was a professional mariner of sorts, a way for me to refresh, reset, and reinvigorate. Indeed, reflecting from the podium, it had been a fantastic meeting that served its purpose well. I had learned a ton, caught up with colleagues I hadn’t seen since the last meeting, saw old medical school friends, and met future old friends. I’d led a committee, given a talk, presented a poster, met up with a mentor, and had a reunion with past attendees of the Academic Hospitalist Academy.

Yet I was ready to get back to normalcy. On the last night of the meeting, I was therefore drawn by a different, more personal mariner—this time, a 14-second voicemail message from a 3-year-old boy waiting impatiently for Dad to come home, to make him his center, to simply play a little tee ball in the backyard. TH

Dr. Glasheen is associate professor of medicine at the University of Colorado at Denver, where he serves as director of the Hospital Medicine Program and the Hospitalist Training Program, and as associate program director of the Internal Medicine Residency Program.

I stare; a brimming audience stares back. Two eyeballs battling thousands. Slightly uncomfortable, I shift my weight, trying to hide behind the glass podium. Two microphones snake out of the podium slithering together inches from my mouth. The attendees squirm, sidle to the edge of their seats, restless to depart. HM11 is trying to close; only I stand in its way.

A Herculean task lies before me—summarize the annual meeting in a 10-minute wrap-up session titled “What We’ve Learned.” How do you summarize four days, eight pre-courses, nine breakout tracks, and more than 100 presentations in a few minutes? A bead of forehead sweat forms; I clear my throat. Memories of the past few days slide-show across my mind. It occurs to me that the essence of the meeting is not contained in the data, the information, or the PowerPoint slides that were presented. Rather, the story of HM11 is best told through its quotes.

Patient Caps: Your Grandmother and Professionalism

“I worry about patient caps because the next patient could be your grandmother.”

—Joe Li, MD, SFHM, new president of SHM

“Patient caps are the greatest threat to the professionalism of the field.”

—Rob Bessler, MD, CEO, Sound Inpatient Physicians

These two quotes from the opening plenary focused on the 2011 HM compensation and productivity survey particularly stuck out. The most noteworthy exchange came when Drs. Li and Bressler commented on the appropriate number of daily encounters for a hospitalist. The quotes highlight two important points about patient volume, especially in the wake of the training regulations that limit the number of resident physician encounters, which can engender a “cap mentality.” One is that it matters; there is a safe amount of encounters that shouldn’t routinely be breached. Two is that in the heat of the moment, Patient 19 is as important as Patient 11 and should be treated as such. Contingency plans are essential, but our field is built on the moorings of professionalism—the focus needs to be on humans, not numbers.

Hospitalist Compensation: Increasing but Not as Juicy

“It’s not going to get less anytime soon.”

—Dr. Bressler

In commenting on the data showing that the average community hospitalist makes about $220,000 annually—a 3% increase over last year—while producing around 4,000 work RVUs—flat over last year—and that their academic counterparts made $173,000 on about 3,400 wRVUs, Dr. Bressler opined that the laws of supply and demand would dictate that salaries would continue to rise for the near term. Although I agree with Dr. Bressler, my guess is that future salary increases will be driven more by quality than quantity (more to follow below).

“Juice-to-squeeze ratio”

—John Nelson, MD, MHM, SHM cofounder

Dr. Nelson highlighted interesting data showing that the average pay per wRVU was approximately $54. However, he noted that the compensation per wRVU tends to peak at a certain level, after which compensation per wRVU falls. In other words, after, say, 4,000 wRVUs, the amount of compensation per wRVU diminishes such that seeing more patients benefits an individual hospitalist less. That is, lots of squeeze, little juice at the high end.

Reform: Variety, Change, and Waste

“Variety is about choice; change is not.”

—Cecil Wilson, MD, AMA president

“You won’t have many more conferences where you start by talking about work RVUs.”

—Bob Kocher, MD, former special assistant to President Obama

The highlight of the conference for me was Dr. Kocher’s behind-the-scenes look at what was a very publicly muddy event—the passage of ACA. Coming from a D.C. insider, this under-the-covers peek at the machinations that went into passing the healthcare reform bill was fascinating.

The key message, summarized in this comment referring to the opening plenary about hospitalist compensation and productivity, was that the future is quality and the future is now. In the very near future, we will be measured and paid based on our ability to effect quality outcomes, not patient encounters. The message was simple: It’s about quality, not quantity.

“It costs $7.50 for a healthcare transaction, versus 2 cents for a VISA transaction.”

—Dr. Kocher

A statistic I had not heard before, this quote sums up one of the major problems with American healthcare: waste. The $7.50 transaction he was referring to was the amount of money it takes to file a healthcare claim. We certainly feel it in the challenges of documentation, billing, and denials, but the system feels it in terms of high cost of capturing what in many ways should be as simple as swiping your credit card at Starbucks.

Duty-Hour Restrictions: Harbinger of The Future?

“Don’t begrudge the ACGME—begrudge us.”

—Jeff Wiese, MD, SFHM, SHM past president

In a much-anticipated session on the impact of the new ACGME residency work-hour rules commencing in July—notably limiting intern (16-hour) and resident (28-hour) shift duration—Dr. Wiese aptly pointed out that a lot of the angst toward residency work environment regulation could have been avoided if physician leadership had better reacted to the issues of sleep deprivation and resident fatigue following Libby Zion’s death in 1984. Had we put our energy into improving work conditions rather than debate the impact of sleep deprivation on the outcome in this one case, we might be in a different place today.

I couldn’t help but wonder if the message here could also be applied to society’s push for higher quality, lower cost, and safer care. Either we regulate ourselves or someone else will. In other words, we need to embrace quality and safety, or it will be thrust upon us from external sources in ways we might not like.

A Mariner Calls

“I love you, Papi. Come home and take some baseball cuts.”

—Greyson Glasheen, future Major League Baseball shortstop

I wrote in a column leading up to the annual meeting (see “Annual Meeting Mariner,” April 2011, p. 45) that I was looking forward to the meeting because it was a professional mariner of sorts, a way for me to refresh, reset, and reinvigorate. Indeed, reflecting from the podium, it had been a fantastic meeting that served its purpose well. I had learned a ton, caught up with colleagues I hadn’t seen since the last meeting, saw old medical school friends, and met future old friends. I’d led a committee, given a talk, presented a poster, met up with a mentor, and had a reunion with past attendees of the Academic Hospitalist Academy.

Yet I was ready to get back to normalcy. On the last night of the meeting, I was therefore drawn by a different, more personal mariner—this time, a 14-second voicemail message from a 3-year-old boy waiting impatiently for Dad to come home, to make him his center, to simply play a little tee ball in the backyard. TH

Dr. Glasheen is associate professor of medicine at the University of Colorado at Denver, where he serves as director of the Hospital Medicine Program and the Hospitalist Training Program, and as associate program director of the Internal Medicine Residency Program.

Eight glorious months ago, my wife, Bridget, and I went to the hospital for the birth of our daughter, Livia. I remember the night clearly. It was a planned induction. Labor and delivery was quite busy, so we spent a few hours in the waiting area before our room was ready. Prominently displayed, a royal blue banner and crystal piece announced a Codman Award from the Joint Commission. Presented to only a few healthcare champions annually, this award represented a significant achievement in birth safety. I was proud to have Bridget (and Livia) there.

But had it not been for the Institute for Healthcare Improvement (IHI) Annual Forum’s plenary sessions earlier that year, I probably would have ignored the flashy cabinet, mistaking it for propaganda or a feel-good award that everyone receives if they are nice to Joint Commission inspectors. As it were, I recalled the IHI panel discussion where I had first heard the CEO of Seton Family of Hospitals describe dramatic reductions in the network’s rates of birth injury. Most contentious had been the elimination of elective labor inductions and C-sections at our hospitals before 39 weeks’ gestation.

Perhaps understandably, Bridget was not distracted by any of this. Eyes closed, she was trying to make it through one last uncomfortable night while resting sideways on four chairs pushed together. I knew better than to force the conversation. Two weeks earlier, we had a heated discussion about whether there was any reason to induce earlier for convenience (i.e. obstetrician and grandparents-to-be schedules).

A few months later, at a meeting of physician leadership in our network, the question of whether doctors could lead transformative improvements in care in our community was raised. Thinking back to the Codman Award, I asked an obstetrician if the birth-safety initiative had increased the leadership capacity of physicians.

The reply was quick. “Not really,” she said. “The physicians felt like they were just following some rules.”

Rules? Nobody wanted to bask in the glory of a project that greatly improved outcomes and reduced costs? As I sat in silence and tried to absorb the significance of the response, I was hit from the right with another revelation. A hospital executive in the group noted that this was a very unpopular initiative amongst administrators. There were now fewer feeders and growers populating our NICUs, and this significantly and negatively impacted the bottom line of the hospitals. NICU reimbursement, of course, is a cash cow.

Thus, it came as no surprise when my editor forwarded a recent New York Times piece (www.nytimes.com/2011/03/20/us/20ttnicus.html) on this very issue of overuse in NICU care. The article even profiled my hospital network in Austin, Texas. The drama in the story was the millions of dollars lost by hospitals, potential Texas Medicaid crackdowns on NICU care, and the move away from convenience care.

But a much more important point was missed … value.

The Only Goal

Simply stated, value in healthcare is quality outcomes divided by total costs of care. The real storyline here is that a multidisciplinary team within Seton has greatly improved the single most important metric in healthcare—value. The numerator is healthy deliveries. The denominator is total costs of care. Quality outcomes will drive costs lower, and maximizing this equation should be the only goal we work toward. And yet, routine discussions of how we achieve value are all but absent in our daily conversations.

I suppose it’s only natural that we are distracted. The media will always focus on the dramatic aspect of the story. Political strategists spend days in fluorescently lit rooms devising new ways to keep us misdirected (think death panels). Our academic research agenda continues to prioritize technological advances over efficient healthcare delivery. And our fragmented payment systems all but guarantee that care providers will waste their time on the wrong financial analyses. “Perverse” is an oft-used term to describe our reimbursement system; it aptly describes my experience with “performance” data. How is it that I am regularly subjected to financial reports detailing every bit of billing and coding minutiae, but it takes an act of Congress for me to find simple clinical outcomes data, let alone costs of care? Value is the forgotten stepchild of healthcare reform rhetoric.

Thus, the publicizing of overuse in NICUs is a microcosm of the quagmire that we find ourselves in today. Healthcare spending is a tsunami projected to devastate the shores of our national economy in as little as five years. In the shadow of this rapidly receding financial wave, competing interest groups stand barefoot on the beach debating whether the clinical waste surrounding us is really pollution (one person’s waste is another’s income, as the saying goes). It’s as if we’re all frozen by the spectacle, unable to move toward higher-level value solutions.

All sides will agree, however, that we are quickly running out of time. Continued inaction will condemn us to a crash financial evacuation of cholera-like proportions.

Simple Solution: HM

How do we avert such a natural disaster? I see front-line clinicians—yes, hospitalists—leading the way. Hospitals and healthcare networks are actively mobilizing to create accountable-care organizations (ACOs) in preparation for payment reform almost certain to resurrect some form of capitation or bundling. The finance department of these organizations can only do so much. As they feel the tremors of financial instability, they will cling to what they know—increasing revenue through new services and budget line-item reductions (e.g. decreased funding for hospitalists).

These are short-term solutions at best, and your HM group might already be experiencing the after-effects of such activity.

Hospital administrators will tighten the financial belts, but they cannot improve clinical quality by reducing waste. To paraphrase Atul Gawande, doctors must cap their own pens if we are to reduce waste in the system. Value, then, can only be defined at the bedside in the context of a healthy physician-patient relationship. And as hospitalists, we are at the bedside of the most expensive decisions in medicine.

Although the future landscape might seem bleak, opportunities for HM are aglow with promise. We have the best view of how the system might make the biggest gains. We have been raised with a focus on quality. Scores of improvement success stories are told annually at our national meetings. If we can shift our conversations to improving quality while lowering costs, I believe that defining value will prove to be our field of dreams. TH

Dr. Shen is medical director of hospital medicine at Dell Children’s Hospital in Austin, Texas, and The Hospitalist’s pediatric editor.

Eight glorious months ago, my wife, Bridget, and I went to the hospital for the birth of our daughter, Livia. I remember the night clearly. It was a planned induction. Labor and delivery was quite busy, so we spent a few hours in the waiting area before our room was ready. Prominently displayed, a royal blue banner and crystal piece announced a Codman Award from the Joint Commission. Presented to only a few healthcare champions annually, this award represented a significant achievement in birth safety. I was proud to have Bridget (and Livia) there.

But had it not been for the Institute for Healthcare Improvement (IHI) Annual Forum’s plenary sessions earlier that year, I probably would have ignored the flashy cabinet, mistaking it for propaganda or a feel-good award that everyone receives if they are nice to Joint Commission inspectors. As it were, I recalled the IHI panel discussion where I had first heard the CEO of Seton Family of Hospitals describe dramatic reductions in the network’s rates of birth injury. Most contentious had been the elimination of elective labor inductions and C-sections at our hospitals before 39 weeks’ gestation.

Perhaps understandably, Bridget was not distracted by any of this. Eyes closed, she was trying to make it through one last uncomfortable night while resting sideways on four chairs pushed together. I knew better than to force the conversation. Two weeks earlier, we had a heated discussion about whether there was any reason to induce earlier for convenience (i.e. obstetrician and grandparents-to-be schedules).

A few months later, at a meeting of physician leadership in our network, the question of whether doctors could lead transformative improvements in care in our community was raised. Thinking back to the Codman Award, I asked an obstetrician if the birth-safety initiative had increased the leadership capacity of physicians.

The reply was quick. “Not really,” she said. “The physicians felt like they were just following some rules.”

Rules? Nobody wanted to bask in the glory of a project that greatly improved outcomes and reduced costs? As I sat in silence and tried to absorb the significance of the response, I was hit from the right with another revelation. A hospital executive in the group noted that this was a very unpopular initiative amongst administrators. There were now fewer feeders and growers populating our NICUs, and this significantly and negatively impacted the bottom line of the hospitals. NICU reimbursement, of course, is a cash cow.

Thus, it came as no surprise when my editor forwarded a recent New York Times piece (www.nytimes.com/2011/03/20/us/20ttnicus.html) on this very issue of overuse in NICU care. The article even profiled my hospital network in Austin, Texas. The drama in the story was the millions of dollars lost by hospitals, potential Texas Medicaid crackdowns on NICU care, and the move away from convenience care.

But a much more important point was missed … value.

The Only Goal

Simply stated, value in healthcare is quality outcomes divided by total costs of care. The real storyline here is that a multidisciplinary team within Seton has greatly improved the single most important metric in healthcare—value. The numerator is healthy deliveries. The denominator is total costs of care. Quality outcomes will drive costs lower, and maximizing this equation should be the only goal we work toward. And yet, routine discussions of how we achieve value are all but absent in our daily conversations.

I suppose it’s only natural that we are distracted. The media will always focus on the dramatic aspect of the story. Political strategists spend days in fluorescently lit rooms devising new ways to keep us misdirected (think death panels). Our academic research agenda continues to prioritize technological advances over efficient healthcare delivery. And our fragmented payment systems all but guarantee that care providers will waste their time on the wrong financial analyses. “Perverse” is an oft-used term to describe our reimbursement system; it aptly describes my experience with “performance” data. How is it that I am regularly subjected to financial reports detailing every bit of billing and coding minutiae, but it takes an act of Congress for me to find simple clinical outcomes data, let alone costs of care? Value is the forgotten stepchild of healthcare reform rhetoric.

Thus, the publicizing of overuse in NICUs is a microcosm of the quagmire that we find ourselves in today. Healthcare spending is a tsunami projected to devastate the shores of our national economy in as little as five years. In the shadow of this rapidly receding financial wave, competing interest groups stand barefoot on the beach debating whether the clinical waste surrounding us is really pollution (one person’s waste is another’s income, as the saying goes). It’s as if we’re all frozen by the spectacle, unable to move toward higher-level value solutions.

All sides will agree, however, that we are quickly running out of time. Continued inaction will condemn us to a crash financial evacuation of cholera-like proportions.

Simple Solution: HM

How do we avert such a natural disaster? I see front-line clinicians—yes, hospitalists—leading the way. Hospitals and healthcare networks are actively mobilizing to create accountable-care organizations (ACOs) in preparation for payment reform almost certain to resurrect some form of capitation or bundling. The finance department of these organizations can only do so much. As they feel the tremors of financial instability, they will cling to what they know—increasing revenue through new services and budget line-item reductions (e.g. decreased funding for hospitalists).

These are short-term solutions at best, and your HM group might already be experiencing the after-effects of such activity.

Hospital administrators will tighten the financial belts, but they cannot improve clinical quality by reducing waste. To paraphrase Atul Gawande, doctors must cap their own pens if we are to reduce waste in the system. Value, then, can only be defined at the bedside in the context of a healthy physician-patient relationship. And as hospitalists, we are at the bedside of the most expensive decisions in medicine.

Although the future landscape might seem bleak, opportunities for HM are aglow with promise. We have the best view of how the system might make the biggest gains. We have been raised with a focus on quality. Scores of improvement success stories are told annually at our national meetings. If we can shift our conversations to improving quality while lowering costs, I believe that defining value will prove to be our field of dreams. TH

Dr. Shen is medical director of hospital medicine at Dell Children’s Hospital in Austin, Texas, and The Hospitalist’s pediatric editor.

Eight glorious months ago, my wife, Bridget, and I went to the hospital for the birth of our daughter, Livia. I remember the night clearly. It was a planned induction. Labor and delivery was quite busy, so we spent a few hours in the waiting area before our room was ready. Prominently displayed, a royal blue banner and crystal piece announced a Codman Award from the Joint Commission. Presented to only a few healthcare champions annually, this award represented a significant achievement in birth safety. I was proud to have Bridget (and Livia) there.

But had it not been for the Institute for Healthcare Improvement (IHI) Annual Forum’s plenary sessions earlier that year, I probably would have ignored the flashy cabinet, mistaking it for propaganda or a feel-good award that everyone receives if they are nice to Joint Commission inspectors. As it were, I recalled the IHI panel discussion where I had first heard the CEO of Seton Family of Hospitals describe dramatic reductions in the network’s rates of birth injury. Most contentious had been the elimination of elective labor inductions and C-sections at our hospitals before 39 weeks’ gestation.

Perhaps understandably, Bridget was not distracted by any of this. Eyes closed, she was trying to make it through one last uncomfortable night while resting sideways on four chairs pushed together. I knew better than to force the conversation. Two weeks earlier, we had a heated discussion about whether there was any reason to induce earlier for convenience (i.e. obstetrician and grandparents-to-be schedules).

A few months later, at a meeting of physician leadership in our network, the question of whether doctors could lead transformative improvements in care in our community was raised. Thinking back to the Codman Award, I asked an obstetrician if the birth-safety initiative had increased the leadership capacity of physicians.

The reply was quick. “Not really,” she said. “The physicians felt like they were just following some rules.”

Rules? Nobody wanted to bask in the glory of a project that greatly improved outcomes and reduced costs? As I sat in silence and tried to absorb the significance of the response, I was hit from the right with another revelation. A hospital executive in the group noted that this was a very unpopular initiative amongst administrators. There were now fewer feeders and growers populating our NICUs, and this significantly and negatively impacted the bottom line of the hospitals. NICU reimbursement, of course, is a cash cow.

Thus, it came as no surprise when my editor forwarded a recent New York Times piece (www.nytimes.com/2011/03/20/us/20ttnicus.html) on this very issue of overuse in NICU care. The article even profiled my hospital network in Austin, Texas. The drama in the story was the millions of dollars lost by hospitals, potential Texas Medicaid crackdowns on NICU care, and the move away from convenience care.

But a much more important point was missed … value.

The Only Goal

Simply stated, value in healthcare is quality outcomes divided by total costs of care. The real storyline here is that a multidisciplinary team within Seton has greatly improved the single most important metric in healthcare—value. The numerator is healthy deliveries. The denominator is total costs of care. Quality outcomes will drive costs lower, and maximizing this equation should be the only goal we work toward. And yet, routine discussions of how we achieve value are all but absent in our daily conversations.

I suppose it’s only natural that we are distracted. The media will always focus on the dramatic aspect of the story. Political strategists spend days in fluorescently lit rooms devising new ways to keep us misdirected (think death panels). Our academic research agenda continues to prioritize technological advances over efficient healthcare delivery. And our fragmented payment systems all but guarantee that care providers will waste their time on the wrong financial analyses. “Perverse” is an oft-used term to describe our reimbursement system; it aptly describes my experience with “performance” data. How is it that I am regularly subjected to financial reports detailing every bit of billing and coding minutiae, but it takes an act of Congress for me to find simple clinical outcomes data, let alone costs of care? Value is the forgotten stepchild of healthcare reform rhetoric.

Thus, the publicizing of overuse in NICUs is a microcosm of the quagmire that we find ourselves in today. Healthcare spending is a tsunami projected to devastate the shores of our national economy in as little as five years. In the shadow of this rapidly receding financial wave, competing interest groups stand barefoot on the beach debating whether the clinical waste surrounding us is really pollution (one person’s waste is another’s income, as the saying goes). It’s as if we’re all frozen by the spectacle, unable to move toward higher-level value solutions.

All sides will agree, however, that we are quickly running out of time. Continued inaction will condemn us to a crash financial evacuation of cholera-like proportions.

Simple Solution: HM

How do we avert such a natural disaster? I see front-line clinicians—yes, hospitalists—leading the way. Hospitals and healthcare networks are actively mobilizing to create accountable-care organizations (ACOs) in preparation for payment reform almost certain to resurrect some form of capitation or bundling. The finance department of these organizations can only do so much. As they feel the tremors of financial instability, they will cling to what they know—increasing revenue through new services and budget line-item reductions (e.g. decreased funding for hospitalists).

These are short-term solutions at best, and your HM group might already be experiencing the after-effects of such activity.

Hospital administrators will tighten the financial belts, but they cannot improve clinical quality by reducing waste. To paraphrase Atul Gawande, doctors must cap their own pens if we are to reduce waste in the system. Value, then, can only be defined at the bedside in the context of a healthy physician-patient relationship. And as hospitalists, we are at the bedside of the most expensive decisions in medicine.

Although the future landscape might seem bleak, opportunities for HM are aglow with promise. We have the best view of how the system might make the biggest gains. We have been raised with a focus on quality. Scores of improvement success stories are told annually at our national meetings. If we can shift our conversations to improving quality while lowering costs, I believe that defining value will prove to be our field of dreams. TH

Dr. Shen is medical director of hospital medicine at Dell Children’s Hospital in Austin, Texas, and The Hospitalist’s pediatric editor.

To the Editor: In their article, “Is iron therapy for anemia harmful in the setting of infection?” in the March 2011 issue, Daoud et al¹ illustrated an interesting aspect we often encounter, especially in nephrology practice. However, I believe several points should be clarified in this context.

First, “iron therapy” and “iron stores” are quite different things when we talk about infection. As Daoud et al state, human studies involving iron therapy and infection are conflicting in their results. The explanation is likely that iron therapy per se does not always translate to increased iron stores, while iron stores do correlate with increased risk of infection and death, whether in hemodialysis patients² or in the general population.³ Intravenous iron therapy mostly gains the association with risk of infection when dosed greater than a certain amount or for an extended duration. In addition, Pieracci et al⁴ showed that oral iron therapy for anemia does not boost the infection rate during critical illness when equivalent iron markers are achieved.

This mounting evidence solidifies the view that iron stores underlie the infection susceptibility. But to prove this concept, a randomized controlled study consisting of achieving similar iron stores by component therapy or intravenous iron supplementation would be the best option.

Second, I wish to add a category of infection omitted in their article, ie, fungal infection (mucormycosis). Mucormycosis, a rare but life-threatening disease, is caused by fungi of the class Zygomycetes that spread systemically in immunocompromised hosts, with a high death rate. Iron overload, whether or not accompanied by the use of deferoxamine (Desferal), is an established risk factor for mucormycosis. These fungi possess a high-affinity iron permease and produce siderophores, both of which facilitate the uptake of iron.⁵ An abundant host iron pool further enhances their scavenging process, resulting in devastating proliferation and tissue damage. This disease category should be borne in mind when dealing with immunocompromised patients undergoing iron therapy.

To the Editor: In their article, “Is iron therapy for anemia harmful in the setting of infection?” in the March 2011 issue, Daoud et al¹ illustrated an interesting aspect we often encounter, especially in nephrology practice. However, I believe several points should be clarified in this context.

First, “iron therapy” and “iron stores” are quite different things when we talk about infection. As Daoud et al state, human studies involving iron therapy and infection are conflicting in their results. The explanation is likely that iron therapy per se does not always translate to increased iron stores, while iron stores do correlate with increased risk of infection and death, whether in hemodialysis patients² or in the general population.³ Intravenous iron therapy mostly gains the association with risk of infection when dosed greater than a certain amount or for an extended duration. In addition, Pieracci et al⁴ showed that oral iron therapy for anemia does not boost the infection rate during critical illness when equivalent iron markers are achieved.

This mounting evidence solidifies the view that iron stores underlie the infection susceptibility. But to prove this concept, a randomized controlled study consisting of achieving similar iron stores by component therapy or intravenous iron supplementation would be the best option.

Second, I wish to add a category of infection omitted in their article, ie, fungal infection (mucormycosis). Mucormycosis, a rare but life-threatening disease, is caused by fungi of the class Zygomycetes that spread systemically in immunocompromised hosts, with a high death rate. Iron overload, whether or not accompanied by the use of deferoxamine (Desferal), is an established risk factor for mucormycosis. These fungi possess a high-affinity iron permease and produce siderophores, both of which facilitate the uptake of iron.⁵ An abundant host iron pool further enhances their scavenging process, resulting in devastating proliferation and tissue damage. This disease category should be borne in mind when dealing with immunocompromised patients undergoing iron therapy.

To the Editor: In their article, “Is iron therapy for anemia harmful in the setting of infection?” in the March 2011 issue, Daoud et al¹ illustrated an interesting aspect we often encounter, especially in nephrology practice. However, I believe several points should be clarified in this context.

First, “iron therapy” and “iron stores” are quite different things when we talk about infection. As Daoud et al state, human studies involving iron therapy and infection are conflicting in their results. The explanation is likely that iron therapy per se does not always translate to increased iron stores, while iron stores do correlate with increased risk of infection and death, whether in hemodialysis patients² or in the general population.³ Intravenous iron therapy mostly gains the association with risk of infection when dosed greater than a certain amount or for an extended duration. In addition, Pieracci et al⁴ showed that oral iron therapy for anemia does not boost the infection rate during critical illness when equivalent iron markers are achieved.

This mounting evidence solidifies the view that iron stores underlie the infection susceptibility. But to prove this concept, a randomized controlled study consisting of achieving similar iron stores by component therapy or intravenous iron supplementation would be the best option.

Second, I wish to add a category of infection omitted in their article, ie, fungal infection (mucormycosis). Mucormycosis, a rare but life-threatening disease, is caused by fungi of the class Zygomycetes that spread systemically in immunocompromised hosts, with a high death rate. Iron overload, whether or not accompanied by the use of deferoxamine (Desferal), is an established risk factor for mucormycosis. These fungi possess a high-affinity iron permease and produce siderophores, both of which facilitate the uptake of iron.⁵ An abundant host iron pool further enhances their scavenging process, resulting in devastating proliferation and tissue damage. This disease category should be borne in mind when dealing with immunocompromised patients undergoing iron therapy.

In Reply: We agree that iron therapy is different than iron stores, but iron therapy should be started on the basis of depleted iron stores; otherwise, it is unjustifiable. We also agree that elevated iron stores are dangerous in the setting of infection, more than iron therapy itself. This is really an unproven theory. Most studies that showed worse outcomes of iron therapy found that elevated ferritin is a risk factor.¹ The problem, as we outlined in our paper, is that most serum markers of iron are unreliable in case of inflammation or infection or in the critically ill.² Evaluation of bone marrow stores is probably the most accurate.³

In Reply: We agree that iron therapy is different than iron stores, but iron therapy should be started on the basis of depleted iron stores; otherwise, it is unjustifiable. We also agree that elevated iron stores are dangerous in the setting of infection, more than iron therapy itself. This is really an unproven theory. Most studies that showed worse outcomes of iron therapy found that elevated ferritin is a risk factor.¹ The problem, as we outlined in our paper, is that most serum markers of iron are unreliable in case of inflammation or infection or in the critically ill.² Evaluation of bone marrow stores is probably the most accurate.³

In Reply: We agree that iron therapy is different than iron stores, but iron therapy should be started on the basis of depleted iron stores; otherwise, it is unjustifiable. We also agree that elevated iron stores are dangerous in the setting of infection, more than iron therapy itself. This is really an unproven theory. Most studies that showed worse outcomes of iron therapy found that elevated ferritin is a risk factor.¹ The problem, as we outlined in our paper, is that most serum markers of iron are unreliable in case of inflammation or infection or in the critically ill.² Evaluation of bone marrow stores is probably the most accurate.³

To the Editor: I congratulate Drs. O’Grady and Chertow for their excellent review on bloodstream infections.¹ I just want to call attention to one aspect that the authors forgot. In Figure 1, they classified patients as being mildly or moderately ill if they had no hypotension or organ failure, and subdivided this group into those having or not having high-risk factors. The high-risk factors included evidence of severe sepsis, which by definition needs dysfunction or failure of one or more organs.²

As has been demonstrated by epidemiologic studies, severe sepsis is associated with a high risk of death,³ twice as high as in patients with only catheter-related bloodstream infection.⁴ So, according to the joint guidelines of the American College of Chest Physicians and the Society of Critical Care Medicine,² severe sepsis implies dysfunction or failure of at least one organ. I believe that patients with severe sepsis should be classified in the group of seriously ill.

To the Editor: I congratulate Drs. O’Grady and Chertow for their excellent review on bloodstream infections.¹ I just want to call attention to one aspect that the authors forgot. In Figure 1, they classified patients as being mildly or moderately ill if they had no hypotension or organ failure, and subdivided this group into those having or not having high-risk factors. The high-risk factors included evidence of severe sepsis, which by definition needs dysfunction or failure of one or more organs.²

As has been demonstrated by epidemiologic studies, severe sepsis is associated with a high risk of death,³ twice as high as in patients with only catheter-related bloodstream infection.⁴ So, according to the joint guidelines of the American College of Chest Physicians and the Society of Critical Care Medicine,² severe sepsis implies dysfunction or failure of at least one organ. I believe that patients with severe sepsis should be classified in the group of seriously ill.

To the Editor: I congratulate Drs. O’Grady and Chertow for their excellent review on bloodstream infections.¹ I just want to call attention to one aspect that the authors forgot. In Figure 1, they classified patients as being mildly or moderately ill if they had no hypotension or organ failure, and subdivided this group into those having or not having high-risk factors. The high-risk factors included evidence of severe sepsis, which by definition needs dysfunction or failure of one or more organs.²

As has been demonstrated by epidemiologic studies, severe sepsis is associated with a high risk of death,³ twice as high as in patients with only catheter-related bloodstream infection.⁴ So, according to the joint guidelines of the American College of Chest Physicians and the Society of Critical Care Medicine,² severe sepsis implies dysfunction or failure of at least one organ. I believe that patients with severe sepsis should be classified in the group of seriously ill.

In Reply: We thank Dr. Dias for his careful read of our article, “Managing bloodstream infections in patients who have short-term central venous catheters,” and we acknowledge that he is correct to point out that, by definition, severe sepsis is sepsis associated with organ dysfunction, hypoperfusion, or hypotension. Given this, he is correct that patients with severe sepsis should be categorized in the “seriously ill” patient group in our Figure 1.

In effect, however, the recommendations for patients in the “high-risk-factor” group are the same as the recommendations for the “seriously ill” patient group, which are to remove the catheter, draw at least two sets of blood cultures with at least one from a peripheral vein, and start empiric antibiotic therapy.

In Reply: We thank Dr. Dias for his careful read of our article, “Managing bloodstream infections in patients who have short-term central venous catheters,” and we acknowledge that he is correct to point out that, by definition, severe sepsis is sepsis associated with organ dysfunction, hypoperfusion, or hypotension. Given this, he is correct that patients with severe sepsis should be categorized in the “seriously ill” patient group in our Figure 1.

In effect, however, the recommendations for patients in the “high-risk-factor” group are the same as the recommendations for the “seriously ill” patient group, which are to remove the catheter, draw at least two sets of blood cultures with at least one from a peripheral vein, and start empiric antibiotic therapy.

In Reply: We thank Dr. Dias for his careful read of our article, “Managing bloodstream infections in patients who have short-term central venous catheters,” and we acknowledge that he is correct to point out that, by definition, severe sepsis is sepsis associated with organ dysfunction, hypoperfusion, or hypotension. Given this, he is correct that patients with severe sepsis should be categorized in the “seriously ill” patient group in our Figure 1.

In effect, however, the recommendations for patients in the “high-risk-factor” group are the same as the recommendations for the “seriously ill” patient group, which are to remove the catheter, draw at least two sets of blood cultures with at least one from a peripheral vein, and start empiric antibiotic therapy.

Depression of the ST segment and inversion of the T wave are common electrocardiographic abnormalities. Knowing the various ischemic and nonischemic morphologic features is critical for a timely diagnosis of high-risk myocardial ischemia and electrolyte- or drug-related abnormalities. Moreover, it is important to recognize that true posterior infarction or subtle ST-segment elevation infarction may masquerade as ST-segment depression ischemia, and that pulmonary embolism may masquerade as anterior ischemia. These common electrocardiographic abnormalities are summarized in Table 1.

THE ST SEGMENT AND THE T WAVE: A PRIMER

The ST segment corresponds to the plateau phase of ventricular repolarization (phase 2 of the action potential), while the T wave corresponds to the phase of rapid ventricular repolarization (phase 3). ST-segment or T-wave changes may be secondary to abnormalities of depolarization, ie, pre-excitation or abnormalities of QRS voltage or duration.

On the other hand, ST-segment and T-wave abnormalities may be unrelated to any QRS abnormality, in which case they are called primary repolarization abnormalities. These are caused by ischemia, pericarditis, myocarditis, drugs (digoxin, antiarrhythmic drugs), and electrolyte abnormalities, particularly potassium abnormalities.

ST-segment deviation is usually measured at its junction with the end of the QRS complex, ie, the J point, and is referenced against the TP or PR segment.¹ But some prefer to measure the magnitude of the ST-segment deviation 40 to 80 ms after the J point, when all myocardial fibers are expected to have reached the same level of membrane potential and to form an isoelectric ST segment; at the very onset of repolarization, small differences in membrane potential may normally be seen and may cause deviation of the J point and of the early portion of the ST segment.²

Although a diagnosis of ST-segment elevation myocardial infarction (STEMI) that mandates emergency reperfusion therapy requires ST-segment elevation greater than 1 mm in at least two contiguous leads,³ any ST-segment depression or elevation (≥ 0.5 mm, using the usual standard of 1.0 mV = 10 mm) may be abnormal, particularly when the clinical context or the shape of the ST segment suggests ischemia, or when other ischemic signs such as T-wave abnormalities, Q waves, or reciprocal ST-segment changes are concomitantly present. On the other hand, ST-segment depression of up to 0.5 mm in leads V₂ and V₃ and 1 mm in the other leads may be normal.¹

In adults, the T wave normally is inverted in lead aVR; is upright or inverted in leads aVL, III, and V₁; and is upright in leads I, II, aVF, and V₂ through V₆. The T wave is considered inverted when it is deeper than 1 mm; it is considered flat when its peak amplitude is between 1.0 mm and −1.0 mm.¹

As we will discuss, certain features allow the various causes of ST-segment and T-wave abnormalities to be distinguished from one another.

SECONDARY ST-SEGMENT AND T-WAVE ABNORMALITIES

Modified with permission from Hanna EB, Quintal R, Jain N. Cardiology: Handbook for Clinicians. Arlington, VA: Scrubhill Press; 2009:328–354.

• The ST segment and T wave are directed opposite to the QRS: this is called discordance between the QRS complex and the ST-T abnormalities. In the case of right bundle branch block, the ST and T are directed opposite to the terminal portion of the QRS, ie, the part of the QRS deformed by the conduction abnormality.
• The ST segment and T wave are both abnormal and deviate in the same direction, ie, the ST segment is down-sloping and the T wave is inverted in leads with an upright QRS complex, which gives the ST-T complex a “reverse checkmark” asymmetric morphology.
• The ST and T abnormalities are not dynamic, ie, they do not change in the course of several hours to several days.

Thus, in cases of left ventricular hypertrophy or left bundle branch block, since the QRS complex is upright in the left lateral leads I, aVL, V₅, and V₆, the ST segment is characteristically depressed and the T wave is inverted in these leads (Figure 2). In cases of right ventricular hypertrophy or right bundle branch block, T waves are characteristically inverted in the right precordial leads V₁, V₂, and V₃.

Left bundle branch block is always associated with secondary ST-T abnormalities, the absence of which suggests associated ischemia. Left and right ventricular hypertrophy, on the other hand, are not always associated with ST-T abnormalities, but when these are present, they correlate with more severe hypertrophy or ventricular systolic dysfunction,⁴ and have been called strain pattern. In addition, while these morphologic features are consistent with secondary abnormalities, they do not rule out ischemia in a patient with angina.

Some exceptions to these typical morphologic features:

• Right ventricular hypertrophy and right bundle branch block may be associated with isolated T-wave inversion without ST-segment depression in precordial leads V₁, V₂, and V₃.
• Left ventricular hypertrophy may be associated with symmetric T-wave inversion without ST-segment depression or with a horizontally depressed ST segment. This may be the case in up to one-third of ST-T abnormalities secondary to left ventricular hypertrophy and is seen in hypertrophic cardiomyopathy, particularly the apical variant, in leads V₃ through V₆.⁵

ISCHEMIC ST-SEGMENT DEPRESSION, T-WAVE INVERSION, OR BOTH

ST-segment depression or T-wave inversion is consistent with ischemia if any of the following is true:

• The ST-segment depression or T-wave inversion is directed in the same direction as the QRS complex: this is called concordance between the QRS complex and the ST or T abnormality (Figure 1B).
• The ST segment is depressed but the T wave is upright (Figure 1C).
• The T wave has a positive-negative biphasic pattern (Figure 1D).
• The T wave is symmetrically inverted and has a pointed configuration, while the ST segment is not deviated or is upwardly bowed (coved) or horizontally depressed (Figure 1E).
• The magnitude of ST-segment depression progresses or regresses on serial tracings, or ST-segment depression progresses to T-wave abnormality during ischemia-free intervals (dynamic ST-segment depression).

Unlike ST-segment elevation, ST-segment depression does not localize ischemia.⁶ However, the extent and the magnitude of ST-segment depression correlate with the extent and the severity of ischemia. In fact, ST-segment depression in eight or more leads, combined with ST-segment elevation in leads aVR and V₁ and occurring during ischemic pain, is associated with a 75% predictive accuracy for left main coronary artery or three-vessel disease (Figure 3).^7,8 This finding may also be seen in cases of tight proximal stenosis of the left anterior descending coronary artery.⁹

Wellens syndrome

Wellens and his colleagues showed that 75% of patients who developed these T-wave abnormalities and who were treated medically without angiographic investigation went on to develop extensive anterior wall myocardial infarction within a mean of 8.5 days.¹⁰

In a later investigation of 1,260 patients presenting with unstable angina, 180 patients (14%) had this characteristic T-wave pattern.¹¹ All of the latter patients had stenosis of 50% or more in the proximal left anterior descending artery, and 18% had total occlusion of the left anterior descending artery.

Thus, although medical management may provide symptomatic improvement at first, early coronary angiography and revascularization should be strongly considered in anyone with Wellens syndrome because it usually predicts impending anterior myocardial infarction.

Wellens syndrome is characterized by two patterns of T-wave changes. In 75% of cases, T waves are deeply (≥ 5 mm) and symmetrically inverted in leads V₂ through V₄ (Figures 1E, 4B). In 25% of cases, the T wave has a characteristic positive-negative biphasic morphology in leads V₂ through V₄ (Figures 1D, 4A).¹⁰ In both patterns, the ST segment is isoelectric or minimally elevated (< 1 mm) with a straight or convex morphology, the down-slope of the T wave is sharp, and the QT interval is often prolonged. These abnormalities are characteristically seen hours to days after the ischemic chest pain resolves. In fact, the ischemic episode is usually associated with transient ST-segment elevation or depression that progresses to the T-wave abnormality after the pain subsides.¹¹

In Wellens’ original description, only 12% of patients had increases in their creatine kinase levels, and these were small. Therefore, the electrocardiogram may be the only indication of an impending large anterior infarction in a chest-pain-free patient.¹²

T waves that are symmetrically but less deeply inverted than Wellens-type T waves may still represent ischemia. However, this finding is less specific for ischemia and is associated with better outcomes than Wellens syndrome or ST-segment deviation, particularly when the T wave is less than 3 mm deep.¹⁴ In fact, one prospective cohort study found that isolated mild T-wave inversion in patients presenting with acute coronary syndrome is associated with a favorable long-term outcome, similar to that in patients with no electrocardiographic changes.¹⁵

FREQUENTLY MISSED DIAGNOSES MANIFESTING AS ST-SEGMENT DEPRESSION OR T-WAVE INVERSION

True posterior ST-segment elevation myocardial infarction

When accompanied by inferior STEMI, posterior infarction is easily recognized, but it can be difficult to diagnose when it occurs alone, the so-called true posterior STEMI.

In most cases of posterior infarction, the posterior chest leads V₇, V₈, and V₉ reveal ST-segment elevation.¹⁹ One study found that ST-segment depression in the anterior precordial leads was as sensitive as ST-segment elevation in leads V₇ through V₉ in identifying posterior myocardial infarction (sensitivity 80%),²⁰ while other studies found that ST-segment deviation on standard 12-lead electrocardiography has a lower sensitivity (about 60%) in identifying posterior infarction.^18,21

Tall or wide (≥ 0.04-s) R waves in leads V₁ or V₂, particularly when associated with upright T waves, suggest posterior infarction and may further corroborate this diagnosis, but this finding may take up to 24 hours to manifest and is seen in only about 50% of patients with posterior infarction.²¹

Studies have shown that ST-segment elevation on standard 12-lead electrocardiography is found in fewer than 50% of patients with acute left circumflex occlusion and inferoposterior infarction,¹⁸ yet these are cases of “missed” STEMI that indeed benefit from emergency angiography and reperfusion. In addition, studies of non–ST-segment elevation acute coronary syndrome consistently identify patients who have epicardial vessel occlusion (about 15%–20% of cases),¹⁸ yet their initial angiography is usually delayed for hours or days after the initial presentation.

A subgroup analysis from TRITON–TIMI 38 (Trial to Assess Improvement in Therapeutic Outcomes by Optimizing Platelet Inhibition With Prasugrel Thrombolysis in Myocardial Infarction 38) evaluated patients with isolated anterior ST-segment depression. An occluded “culprit” artery was found 26% of the time, most often the left circumflex artery. Moreover, those patients had a significantly higher rate of death or myocardial infarction at 30-day follow-up than patients without a culprit artery, probably related to delayed revascularization.²²

Recognizing that ST-segment depression that is greatest in leads V₁, V₂, or V₃ represents posterior infarction helps identify a portion of the missed STEMIs in a timely fashion. In addition, in cases of anterior ST-segment depression and in cases of chest pain with nondiagnostic electrocardiography, the recording of ST elevation in leads V₇, V₈, and V₉ is highly sensitive for detecting a true posterior injury.

An anterior ischemic pattern of symmetric T-wave inversion in the precordial leads V₁ through V₄ may also be a sign of acute or chronic right ventricular strain, particularly acute pulmonary embolism. Sinus tachycardia is usually present, but other signs of pulmonary embolism, such as right ventricular hypertrophy and right bundle branch block, may be absent. In fact, T-wave inversion in leads V₁ through V₄ is noted in 19% of patients with nonmassive pulmonary embolism and in 85% of patients with massive pulmonary embolism, and is the most sensitive and specific electrocardiographic finding in massive pulmonary embolism.²³

In addition, acute pulmonary embolism may be associated with T-wave inversion in leads III and aVF,²⁴ and changes of concomitant anterior and inferior ischemia should always raise the question of this diagnosis.

In one retrospective study of patients with acute pulmonary embolism, nonspecific ST-segment or T-wave changes were the most common finding on electrocardiography, noted in 49%.²⁵ Rapid regression of these changes on serial tracings favors pulmonary embolism rather than myocardial infarction.

ST-segment depression reciprocal to a subtle ST-segment elevation

When ST-segment elevation occurs in two contiguous standard leads while ST-segment depression occurs in other leads, and when the ST-segment and T-wave abnormalities are ischemic rather than secondary to depolarization abnormalities, ST-segment elevation is considered the primary ischemic abnormality whereas ST-segment depression is often considered a reciprocal “mirror image” change. This “reciprocal” change may also represent remote ischemia in a distant territory in patients with multivessel coronary disease.^26,27

Reciprocal ST-segment depression is present in all patients with inferior myocardial infarction and in 70% of patients with anterior myocardial infarction.²⁸

Hypokalemia and digitalis effect

DIFFUSE (GLOBAL) T-WAVE INVERSION

Reproduced with permission from Glancy DL, et al. Global T-wave inversion in a 77-year-old woman. Proc (Bayl Univ Med Cent) 2009; 22:81–82.

Walder and Spodick³⁶ have found this pattern to be caused most often by myocardial ischemia or neurologic events, particularly intracranial hemorrhage, and it seems more prevalent in women. Other causes include hypertrophic cardiomyopathy, stress-induced cardiomyopathy (takotsubo cardiomyopathy), cocaine abuse, pericarditis, pulmonary embolism, and advanced or complete atrioventricular block.^36,37

The prognosis in patients with global T-wave inversion is determined by the underlying disease, and the striking T-wave changes per se do not imply a poor prognosis.³⁸

OTHER CAUSES OF T-WAVE INVERSION OR ST-SEGMENT DEPRESSION

Various other entities may cause T-wave inversion, notably acute pericarditis or myocarditis, ^41,42 memory T-wave phenomenon,^43,44 and normal variants of repolarization (Table 1, Figure 9).⁴⁵ Additionally, a nonpathologic junctional ST-segment depression may be seen in tachycardia (Figure 10).

Depression of the ST segment and inversion of the T wave are common electrocardiographic abnormalities. Knowing the various ischemic and nonischemic morphologic features is critical for a timely diagnosis of high-risk myocardial ischemia and electrolyte- or drug-related abnormalities. Moreover, it is important to recognize that true posterior infarction or subtle ST-segment elevation infarction may masquerade as ST-segment depression ischemia, and that pulmonary embolism may masquerade as anterior ischemia. These common electrocardiographic abnormalities are summarized in Table 1.

THE ST SEGMENT AND THE T WAVE: A PRIMER

The ST segment corresponds to the plateau phase of ventricular repolarization (phase 2 of the action potential), while the T wave corresponds to the phase of rapid ventricular repolarization (phase 3). ST-segment or T-wave changes may be secondary to abnormalities of depolarization, ie, pre-excitation or abnormalities of QRS voltage or duration.

On the other hand, ST-segment and T-wave abnormalities may be unrelated to any QRS abnormality, in which case they are called primary repolarization abnormalities. These are caused by ischemia, pericarditis, myocarditis, drugs (digoxin, antiarrhythmic drugs), and electrolyte abnormalities, particularly potassium abnormalities.

ST-segment deviation is usually measured at its junction with the end of the QRS complex, ie, the J point, and is referenced against the TP or PR segment.¹ But some prefer to measure the magnitude of the ST-segment deviation 40 to 80 ms after the J point, when all myocardial fibers are expected to have reached the same level of membrane potential and to form an isoelectric ST segment; at the very onset of repolarization, small differences in membrane potential may normally be seen and may cause deviation of the J point and of the early portion of the ST segment.²

Although a diagnosis of ST-segment elevation myocardial infarction (STEMI) that mandates emergency reperfusion therapy requires ST-segment elevation greater than 1 mm in at least two contiguous leads,³ any ST-segment depression or elevation (≥ 0.5 mm, using the usual standard of 1.0 mV = 10 mm) may be abnormal, particularly when the clinical context or the shape of the ST segment suggests ischemia, or when other ischemic signs such as T-wave abnormalities, Q waves, or reciprocal ST-segment changes are concomitantly present. On the other hand, ST-segment depression of up to 0.5 mm in leads V₂ and V₃ and 1 mm in the other leads may be normal.¹

In adults, the T wave normally is inverted in lead aVR; is upright or inverted in leads aVL, III, and V₁; and is upright in leads I, II, aVF, and V₂ through V₆. The T wave is considered inverted when it is deeper than 1 mm; it is considered flat when its peak amplitude is between 1.0 mm and −1.0 mm.¹

As we will discuss, certain features allow the various causes of ST-segment and T-wave abnormalities to be distinguished from one another.

SECONDARY ST-SEGMENT AND T-WAVE ABNORMALITIES

Modified with permission from Hanna EB, Quintal R, Jain N. Cardiology: Handbook for Clinicians. Arlington, VA: Scrubhill Press; 2009:328–354.

• The ST segment and T wave are directed opposite to the QRS: this is called discordance between the QRS complex and the ST-T abnormalities. In the case of right bundle branch block, the ST and T are directed opposite to the terminal portion of the QRS, ie, the part of the QRS deformed by the conduction abnormality.
• The ST segment and T wave are both abnormal and deviate in the same direction, ie, the ST segment is down-sloping and the T wave is inverted in leads with an upright QRS complex, which gives the ST-T complex a “reverse checkmark” asymmetric morphology.
• The ST and T abnormalities are not dynamic, ie, they do not change in the course of several hours to several days.

Thus, in cases of left ventricular hypertrophy or left bundle branch block, since the QRS complex is upright in the left lateral leads I, aVL, V₅, and V₆, the ST segment is characteristically depressed and the T wave is inverted in these leads (Figure 2). In cases of right ventricular hypertrophy or right bundle branch block, T waves are characteristically inverted in the right precordial leads V₁, V₂, and V₃.

Left bundle branch block is always associated with secondary ST-T abnormalities, the absence of which suggests associated ischemia. Left and right ventricular hypertrophy, on the other hand, are not always associated with ST-T abnormalities, but when these are present, they correlate with more severe hypertrophy or ventricular systolic dysfunction,⁴ and have been called strain pattern. In addition, while these morphologic features are consistent with secondary abnormalities, they do not rule out ischemia in a patient with angina.

Some exceptions to these typical morphologic features:

• Right ventricular hypertrophy and right bundle branch block may be associated with isolated T-wave inversion without ST-segment depression in precordial leads V₁, V₂, and V₃.
• Left ventricular hypertrophy may be associated with symmetric T-wave inversion without ST-segment depression or with a horizontally depressed ST segment. This may be the case in up to one-third of ST-T abnormalities secondary to left ventricular hypertrophy and is seen in hypertrophic cardiomyopathy, particularly the apical variant, in leads V₃ through V₆.⁵

ISCHEMIC ST-SEGMENT DEPRESSION, T-WAVE INVERSION, OR BOTH

ST-segment depression or T-wave inversion is consistent with ischemia if any of the following is true:

• The ST-segment depression or T-wave inversion is directed in the same direction as the QRS complex: this is called concordance between the QRS complex and the ST or T abnormality (Figure 1B).
• The ST segment is depressed but the T wave is upright (Figure 1C).
• The T wave has a positive-negative biphasic pattern (Figure 1D).
• The T wave is symmetrically inverted and has a pointed configuration, while the ST segment is not deviated or is upwardly bowed (coved) or horizontally depressed (Figure 1E).
• The magnitude of ST-segment depression progresses or regresses on serial tracings, or ST-segment depression progresses to T-wave abnormality during ischemia-free intervals (dynamic ST-segment depression).

Unlike ST-segment elevation, ST-segment depression does not localize ischemia.⁶ However, the extent and the magnitude of ST-segment depression correlate with the extent and the severity of ischemia. In fact, ST-segment depression in eight or more leads, combined with ST-segment elevation in leads aVR and V₁ and occurring during ischemic pain, is associated with a 75% predictive accuracy for left main coronary artery or three-vessel disease (Figure 3).^7,8 This finding may also be seen in cases of tight proximal stenosis of the left anterior descending coronary artery.⁹

Wellens syndrome

Wellens and his colleagues showed that 75% of patients who developed these T-wave abnormalities and who were treated medically without angiographic investigation went on to develop extensive anterior wall myocardial infarction within a mean of 8.5 days.¹⁰

In a later investigation of 1,260 patients presenting with unstable angina, 180 patients (14%) had this characteristic T-wave pattern.¹¹ All of the latter patients had stenosis of 50% or more in the proximal left anterior descending artery, and 18% had total occlusion of the left anterior descending artery.

Thus, although medical management may provide symptomatic improvement at first, early coronary angiography and revascularization should be strongly considered in anyone with Wellens syndrome because it usually predicts impending anterior myocardial infarction.

Wellens syndrome is characterized by two patterns of T-wave changes. In 75% of cases, T waves are deeply (≥ 5 mm) and symmetrically inverted in leads V₂ through V₄ (Figures 1E, 4B). In 25% of cases, the T wave has a characteristic positive-negative biphasic morphology in leads V₂ through V₄ (Figures 1D, 4A).¹⁰ In both patterns, the ST segment is isoelectric or minimally elevated (< 1 mm) with a straight or convex morphology, the down-slope of the T wave is sharp, and the QT interval is often prolonged. These abnormalities are characteristically seen hours to days after the ischemic chest pain resolves. In fact, the ischemic episode is usually associated with transient ST-segment elevation or depression that progresses to the T-wave abnormality after the pain subsides.¹¹

In Wellens’ original description, only 12% of patients had increases in their creatine kinase levels, and these were small. Therefore, the electrocardiogram may be the only indication of an impending large anterior infarction in a chest-pain-free patient.¹²

T waves that are symmetrically but less deeply inverted than Wellens-type T waves may still represent ischemia. However, this finding is less specific for ischemia and is associated with better outcomes than Wellens syndrome or ST-segment deviation, particularly when the T wave is less than 3 mm deep.¹⁴ In fact, one prospective cohort study found that isolated mild T-wave inversion in patients presenting with acute coronary syndrome is associated with a favorable long-term outcome, similar to that in patients with no electrocardiographic changes.¹⁵

FREQUENTLY MISSED DIAGNOSES MANIFESTING AS ST-SEGMENT DEPRESSION OR T-WAVE INVERSION

True posterior ST-segment elevation myocardial infarction

When accompanied by inferior STEMI, posterior infarction is easily recognized, but it can be difficult to diagnose when it occurs alone, the so-called true posterior STEMI.

In most cases of posterior infarction, the posterior chest leads V₇, V₈, and V₉ reveal ST-segment elevation.¹⁹ One study found that ST-segment depression in the anterior precordial leads was as sensitive as ST-segment elevation in leads V₇ through V₉ in identifying posterior myocardial infarction (sensitivity 80%),²⁰ while other studies found that ST-segment deviation on standard 12-lead electrocardiography has a lower sensitivity (about 60%) in identifying posterior infarction.^18,21

Tall or wide (≥ 0.04-s) R waves in leads V₁ or V₂, particularly when associated with upright T waves, suggest posterior infarction and may further corroborate this diagnosis, but this finding may take up to 24 hours to manifest and is seen in only about 50% of patients with posterior infarction.²¹

Studies have shown that ST-segment elevation on standard 12-lead electrocardiography is found in fewer than 50% of patients with acute left circumflex occlusion and inferoposterior infarction,¹⁸ yet these are cases of “missed” STEMI that indeed benefit from emergency angiography and reperfusion. In addition, studies of non–ST-segment elevation acute coronary syndrome consistently identify patients who have epicardial vessel occlusion (about 15%–20% of cases),¹⁸ yet their initial angiography is usually delayed for hours or days after the initial presentation.

A subgroup analysis from TRITON–TIMI 38 (Trial to Assess Improvement in Therapeutic Outcomes by Optimizing Platelet Inhibition With Prasugrel Thrombolysis in Myocardial Infarction 38) evaluated patients with isolated anterior ST-segment depression. An occluded “culprit” artery was found 26% of the time, most often the left circumflex artery. Moreover, those patients had a significantly higher rate of death or myocardial infarction at 30-day follow-up than patients without a culprit artery, probably related to delayed revascularization.²²

Recognizing that ST-segment depression that is greatest in leads V₁, V₂, or V₃ represents posterior infarction helps identify a portion of the missed STEMIs in a timely fashion. In addition, in cases of anterior ST-segment depression and in cases of chest pain with nondiagnostic electrocardiography, the recording of ST elevation in leads V₇, V₈, and V₉ is highly sensitive for detecting a true posterior injury.

An anterior ischemic pattern of symmetric T-wave inversion in the precordial leads V₁ through V₄ may also be a sign of acute or chronic right ventricular strain, particularly acute pulmonary embolism. Sinus tachycardia is usually present, but other signs of pulmonary embolism, such as right ventricular hypertrophy and right bundle branch block, may be absent. In fact, T-wave inversion in leads V₁ through V₄ is noted in 19% of patients with nonmassive pulmonary embolism and in 85% of patients with massive pulmonary embolism, and is the most sensitive and specific electrocardiographic finding in massive pulmonary embolism.²³

In addition, acute pulmonary embolism may be associated with T-wave inversion in leads III and aVF,²⁴ and changes of concomitant anterior and inferior ischemia should always raise the question of this diagnosis.

In one retrospective study of patients with acute pulmonary embolism, nonspecific ST-segment or T-wave changes were the most common finding on electrocardiography, noted in 49%.²⁵ Rapid regression of these changes on serial tracings favors pulmonary embolism rather than myocardial infarction.

ST-segment depression reciprocal to a subtle ST-segment elevation

When ST-segment elevation occurs in two contiguous standard leads while ST-segment depression occurs in other leads, and when the ST-segment and T-wave abnormalities are ischemic rather than secondary to depolarization abnormalities, ST-segment elevation is considered the primary ischemic abnormality whereas ST-segment depression is often considered a reciprocal “mirror image” change. This “reciprocal” change may also represent remote ischemia in a distant territory in patients with multivessel coronary disease.^26,27

Reciprocal ST-segment depression is present in all patients with inferior myocardial infarction and in 70% of patients with anterior myocardial infarction.²⁸

Hypokalemia and digitalis effect

DIFFUSE (GLOBAL) T-WAVE INVERSION

Reproduced with permission from Glancy DL, et al. Global T-wave inversion in a 77-year-old woman. Proc (Bayl Univ Med Cent) 2009; 22:81–82.

Walder and Spodick³⁶ have found this pattern to be caused most often by myocardial ischemia or neurologic events, particularly intracranial hemorrhage, and it seems more prevalent in women. Other causes include hypertrophic cardiomyopathy, stress-induced cardiomyopathy (takotsubo cardiomyopathy), cocaine abuse, pericarditis, pulmonary embolism, and advanced or complete atrioventricular block.^36,37

The prognosis in patients with global T-wave inversion is determined by the underlying disease, and the striking T-wave changes per se do not imply a poor prognosis.³⁸

OTHER CAUSES OF T-WAVE INVERSION OR ST-SEGMENT DEPRESSION

Various other entities may cause T-wave inversion, notably acute pericarditis or myocarditis, ^41,42 memory T-wave phenomenon,^43,44 and normal variants of repolarization (Table 1, Figure 9).⁴⁵ Additionally, a nonpathologic junctional ST-segment depression may be seen in tachycardia (Figure 10).

Depression of the ST segment and inversion of the T wave are common electrocardiographic abnormalities. Knowing the various ischemic and nonischemic morphologic features is critical for a timely diagnosis of high-risk myocardial ischemia and electrolyte- or drug-related abnormalities. Moreover, it is important to recognize that true posterior infarction or subtle ST-segment elevation infarction may masquerade as ST-segment depression ischemia, and that pulmonary embolism may masquerade as anterior ischemia. These common electrocardiographic abnormalities are summarized in Table 1.

THE ST SEGMENT AND THE T WAVE: A PRIMER

The ST segment corresponds to the plateau phase of ventricular repolarization (phase 2 of the action potential), while the T wave corresponds to the phase of rapid ventricular repolarization (phase 3). ST-segment or T-wave changes may be secondary to abnormalities of depolarization, ie, pre-excitation or abnormalities of QRS voltage or duration.

On the other hand, ST-segment and T-wave abnormalities may be unrelated to any QRS abnormality, in which case they are called primary repolarization abnormalities. These are caused by ischemia, pericarditis, myocarditis, drugs (digoxin, antiarrhythmic drugs), and electrolyte abnormalities, particularly potassium abnormalities.

ST-segment deviation is usually measured at its junction with the end of the QRS complex, ie, the J point, and is referenced against the TP or PR segment.¹ But some prefer to measure the magnitude of the ST-segment deviation 40 to 80 ms after the J point, when all myocardial fibers are expected to have reached the same level of membrane potential and to form an isoelectric ST segment; at the very onset of repolarization, small differences in membrane potential may normally be seen and may cause deviation of the J point and of the early portion of the ST segment.²

Although a diagnosis of ST-segment elevation myocardial infarction (STEMI) that mandates emergency reperfusion therapy requires ST-segment elevation greater than 1 mm in at least two contiguous leads,³ any ST-segment depression or elevation (≥ 0.5 mm, using the usual standard of 1.0 mV = 10 mm) may be abnormal, particularly when the clinical context or the shape of the ST segment suggests ischemia, or when other ischemic signs such as T-wave abnormalities, Q waves, or reciprocal ST-segment changes are concomitantly present. On the other hand, ST-segment depression of up to 0.5 mm in leads V₂ and V₃ and 1 mm in the other leads may be normal.¹

In adults, the T wave normally is inverted in lead aVR; is upright or inverted in leads aVL, III, and V₁; and is upright in leads I, II, aVF, and V₂ through V₆. The T wave is considered inverted when it is deeper than 1 mm; it is considered flat when its peak amplitude is between 1.0 mm and −1.0 mm.¹

As we will discuss, certain features allow the various causes of ST-segment and T-wave abnormalities to be distinguished from one another.

SECONDARY ST-SEGMENT AND T-WAVE ABNORMALITIES

Modified with permission from Hanna EB, Quintal R, Jain N. Cardiology: Handbook for Clinicians. Arlington, VA: Scrubhill Press; 2009:328–354.

• The ST segment and T wave are directed opposite to the QRS: this is called discordance between the QRS complex and the ST-T abnormalities. In the case of right bundle branch block, the ST and T are directed opposite to the terminal portion of the QRS, ie, the part of the QRS deformed by the conduction abnormality.
• The ST segment and T wave are both abnormal and deviate in the same direction, ie, the ST segment is down-sloping and the T wave is inverted in leads with an upright QRS complex, which gives the ST-T complex a “reverse checkmark” asymmetric morphology.
• The ST and T abnormalities are not dynamic, ie, they do not change in the course of several hours to several days.

Thus, in cases of left ventricular hypertrophy or left bundle branch block, since the QRS complex is upright in the left lateral leads I, aVL, V₅, and V₆, the ST segment is characteristically depressed and the T wave is inverted in these leads (Figure 2). In cases of right ventricular hypertrophy or right bundle branch block, T waves are characteristically inverted in the right precordial leads V₁, V₂, and V₃.

Left bundle branch block is always associated with secondary ST-T abnormalities, the absence of which suggests associated ischemia. Left and right ventricular hypertrophy, on the other hand, are not always associated with ST-T abnormalities, but when these are present, they correlate with more severe hypertrophy or ventricular systolic dysfunction,⁴ and have been called strain pattern. In addition, while these morphologic features are consistent with secondary abnormalities, they do not rule out ischemia in a patient with angina.

Some exceptions to these typical morphologic features:

• Right ventricular hypertrophy and right bundle branch block may be associated with isolated T-wave inversion without ST-segment depression in precordial leads V₁, V₂, and V₃.
• Left ventricular hypertrophy may be associated with symmetric T-wave inversion without ST-segment depression or with a horizontally depressed ST segment. This may be the case in up to one-third of ST-T abnormalities secondary to left ventricular hypertrophy and is seen in hypertrophic cardiomyopathy, particularly the apical variant, in leads V₃ through V₆.⁵

ISCHEMIC ST-SEGMENT DEPRESSION, T-WAVE INVERSION, OR BOTH

ST-segment depression or T-wave inversion is consistent with ischemia if any of the following is true:

• The ST-segment depression or T-wave inversion is directed in the same direction as the QRS complex: this is called concordance between the QRS complex and the ST or T abnormality (Figure 1B).
• The ST segment is depressed but the T wave is upright (Figure 1C).
• The T wave has a positive-negative biphasic pattern (Figure 1D).
• The T wave is symmetrically inverted and has a pointed configuration, while the ST segment is not deviated or is upwardly bowed (coved) or horizontally depressed (Figure 1E).
• The magnitude of ST-segment depression progresses or regresses on serial tracings, or ST-segment depression progresses to T-wave abnormality during ischemia-free intervals (dynamic ST-segment depression).

Unlike ST-segment elevation, ST-segment depression does not localize ischemia.⁶ However, the extent and the magnitude of ST-segment depression correlate with the extent and the severity of ischemia. In fact, ST-segment depression in eight or more leads, combined with ST-segment elevation in leads aVR and V₁ and occurring during ischemic pain, is associated with a 75% predictive accuracy for left main coronary artery or three-vessel disease (Figure 3).^7,8 This finding may also be seen in cases of tight proximal stenosis of the left anterior descending coronary artery.⁹

Wellens syndrome

Wellens and his colleagues showed that 75% of patients who developed these T-wave abnormalities and who were treated medically without angiographic investigation went on to develop extensive anterior wall myocardial infarction within a mean of 8.5 days.¹⁰

In a later investigation of 1,260 patients presenting with unstable angina, 180 patients (14%) had this characteristic T-wave pattern.¹¹ All of the latter patients had stenosis of 50% or more in the proximal left anterior descending artery, and 18% had total occlusion of the left anterior descending artery.

Thus, although medical management may provide symptomatic improvement at first, early coronary angiography and revascularization should be strongly considered in anyone with Wellens syndrome because it usually predicts impending anterior myocardial infarction.

Wellens syndrome is characterized by two patterns of T-wave changes. In 75% of cases, T waves are deeply (≥ 5 mm) and symmetrically inverted in leads V₂ through V₄ (Figures 1E, 4B). In 25% of cases, the T wave has a characteristic positive-negative biphasic morphology in leads V₂ through V₄ (Figures 1D, 4A).¹⁰ In both patterns, the ST segment is isoelectric or minimally elevated (< 1 mm) with a straight or convex morphology, the down-slope of the T wave is sharp, and the QT interval is often prolonged. These abnormalities are characteristically seen hours to days after the ischemic chest pain resolves. In fact, the ischemic episode is usually associated with transient ST-segment elevation or depression that progresses to the T-wave abnormality after the pain subsides.¹¹

In Wellens’ original description, only 12% of patients had increases in their creatine kinase levels, and these were small. Therefore, the electrocardiogram may be the only indication of an impending large anterior infarction in a chest-pain-free patient.¹²

T waves that are symmetrically but less deeply inverted than Wellens-type T waves may still represent ischemia. However, this finding is less specific for ischemia and is associated with better outcomes than Wellens syndrome or ST-segment deviation, particularly when the T wave is less than 3 mm deep.¹⁴ In fact, one prospective cohort study found that isolated mild T-wave inversion in patients presenting with acute coronary syndrome is associated with a favorable long-term outcome, similar to that in patients with no electrocardiographic changes.¹⁵

FREQUENTLY MISSED DIAGNOSES MANIFESTING AS ST-SEGMENT DEPRESSION OR T-WAVE INVERSION

True posterior ST-segment elevation myocardial infarction

When accompanied by inferior STEMI, posterior infarction is easily recognized, but it can be difficult to diagnose when it occurs alone, the so-called true posterior STEMI.

In most cases of posterior infarction, the posterior chest leads V₇, V₈, and V₉ reveal ST-segment elevation.¹⁹ One study found that ST-segment depression in the anterior precordial leads was as sensitive as ST-segment elevation in leads V₇ through V₉ in identifying posterior myocardial infarction (sensitivity 80%),²⁰ while other studies found that ST-segment deviation on standard 12-lead electrocardiography has a lower sensitivity (about 60%) in identifying posterior infarction.^18,21

Tall or wide (≥ 0.04-s) R waves in leads V₁ or V₂, particularly when associated with upright T waves, suggest posterior infarction and may further corroborate this diagnosis, but this finding may take up to 24 hours to manifest and is seen in only about 50% of patients with posterior infarction.²¹

Studies have shown that ST-segment elevation on standard 12-lead electrocardiography is found in fewer than 50% of patients with acute left circumflex occlusion and inferoposterior infarction,¹⁸ yet these are cases of “missed” STEMI that indeed benefit from emergency angiography and reperfusion. In addition, studies of non–ST-segment elevation acute coronary syndrome consistently identify patients who have epicardial vessel occlusion (about 15%–20% of cases),¹⁸ yet their initial angiography is usually delayed for hours or days after the initial presentation.

A subgroup analysis from TRITON–TIMI 38 (Trial to Assess Improvement in Therapeutic Outcomes by Optimizing Platelet Inhibition With Prasugrel Thrombolysis in Myocardial Infarction 38) evaluated patients with isolated anterior ST-segment depression. An occluded “culprit” artery was found 26% of the time, most often the left circumflex artery. Moreover, those patients had a significantly higher rate of death or myocardial infarction at 30-day follow-up than patients without a culprit artery, probably related to delayed revascularization.²²

Recognizing that ST-segment depression that is greatest in leads V₁, V₂, or V₃ represents posterior infarction helps identify a portion of the missed STEMIs in a timely fashion. In addition, in cases of anterior ST-segment depression and in cases of chest pain with nondiagnostic electrocardiography, the recording of ST elevation in leads V₇, V₈, and V₉ is highly sensitive for detecting a true posterior injury.

An anterior ischemic pattern of symmetric T-wave inversion in the precordial leads V₁ through V₄ may also be a sign of acute or chronic right ventricular strain, particularly acute pulmonary embolism. Sinus tachycardia is usually present, but other signs of pulmonary embolism, such as right ventricular hypertrophy and right bundle branch block, may be absent. In fact, T-wave inversion in leads V₁ through V₄ is noted in 19% of patients with nonmassive pulmonary embolism and in 85% of patients with massive pulmonary embolism, and is the most sensitive and specific electrocardiographic finding in massive pulmonary embolism.²³

In addition, acute pulmonary embolism may be associated with T-wave inversion in leads III and aVF,²⁴ and changes of concomitant anterior and inferior ischemia should always raise the question of this diagnosis.

In one retrospective study of patients with acute pulmonary embolism, nonspecific ST-segment or T-wave changes were the most common finding on electrocardiography, noted in 49%.²⁵ Rapid regression of these changes on serial tracings favors pulmonary embolism rather than myocardial infarction.

ST-segment depression reciprocal to a subtle ST-segment elevation

When ST-segment elevation occurs in two contiguous standard leads while ST-segment depression occurs in other leads, and when the ST-segment and T-wave abnormalities are ischemic rather than secondary to depolarization abnormalities, ST-segment elevation is considered the primary ischemic abnormality whereas ST-segment depression is often considered a reciprocal “mirror image” change. This “reciprocal” change may also represent remote ischemia in a distant territory in patients with multivessel coronary disease.^26,27

Reciprocal ST-segment depression is present in all patients with inferior myocardial infarction and in 70% of patients with anterior myocardial infarction.²⁸

Hypokalemia and digitalis effect

DIFFUSE (GLOBAL) T-WAVE INVERSION

Reproduced with permission from Glancy DL, et al. Global T-wave inversion in a 77-year-old woman. Proc (Bayl Univ Med Cent) 2009; 22:81–82.

Walder and Spodick³⁶ have found this pattern to be caused most often by myocardial ischemia or neurologic events, particularly intracranial hemorrhage, and it seems more prevalent in women. Other causes include hypertrophic cardiomyopathy, stress-induced cardiomyopathy (takotsubo cardiomyopathy), cocaine abuse, pericarditis, pulmonary embolism, and advanced or complete atrioventricular block.^36,37

The prognosis in patients with global T-wave inversion is determined by the underlying disease, and the striking T-wave changes per se do not imply a poor prognosis.³⁸

OTHER CAUSES OF T-WAVE INVERSION OR ST-SEGMENT DEPRESSION

Various other entities may cause T-wave inversion, notably acute pericarditis or myocarditis, ^41,42 memory T-wave phenomenon,^43,44 and normal variants of repolarization (Table 1, Figure 9).⁴⁵ Additionally, a nonpathologic junctional ST-segment depression may be seen in tachycardia (Figure 10).

Our patient, a 56-year-old woman, presents with proximal muscle weakness in all four limbs. It started a few months ago and has gradually become severe, so that she now has difficulty rising from a seated position and has trouble opening jars. She has fallen several times. She says she has no muscle pain, difficulty swallowing, or difficulty breathing.

She sought medical attention at another hospital and was found to be hypothyroid, with a thyrotropin (thyroid-stimulating hormone [TSH]) level of 38 μU/mL (reference range 0.4–5.5), for which she was started on levothyroxine (Synthroid) 100 μg daily. She also had a low serum potassium level, for which potassium supplements and spironolactone (Aldactone) were started. She was taking furosemide (Lasix) 20 mg/day at the time.

Despite the thyroid replacement therapy, she continued to become weaker and had more falls. She also noticed a new, nonpainful rash on her lower abdomen.

Review of systems

• Night sweats
• Leg swelling
• Puffiness and discoloration around the eyes, with easy bruisability.

Medical history

• Diabetes mellitus
• Seizures in the 1970s
• Resection of a thymic tumor in 2003 (the exact pathology is unknown)
• Cirrhosis of unknown etiology
• No known history of hypertension
• No history of alcohol or intravenous drug use
• Quit smoking many years ago
• Coronary artery bypass surgery in 2003
• One sibling with myasthenia gravis.

Medications

• Levothyroxine
• Rosuvastatin (Crestor)
• Omeprazole (Prilosec)
• Spironolactone
• Furosemide
• Potassium chloride
• Metoprolol tartrate (Lopressor)
• Metformin (Glucophage)
• Ramipril (Altace).

Physical examination

She is hemodynamically stable and is not hypertensive. Her thyroid is not enlarged. Her lungs are clear to auscultation. Her heart sounds are normal, except for a nonradiating pansystolic murmur most audible at the apex.

Her abdomen is soft and is not distended. Her abdominal rash has a dermatomal distribution consistent with an L1 distribution, with vesicles over an erythematous base. Purpuric lesions are noted over her lower extremities.

Her leg strength is 3 on a scale of 5 on both sides; her arm strength is normal. Ankle and knee reflexes are absent bilaterally.

Initial laboratory analysis

PROGRESSIVE MUSCLE WEAKNESS

1. What are possible causes of her muscle weakness?

• Myasthenia gravis
• Hypothyroidism
• Dermatomyositis-polymyositis
• Drug-induced myopathy
• Cushing syndrome
• All of the above

All of these are potential causes of muscle weakness.

Myasthenia gravis

Myasthenia gravis, an autoimmune disease, can affect people of all ages and either sex. It presents with muscle weakness and fatigability, which characteristically fluctuate during the day. Some patients present in crisis with respiratory failure, which may require ventilatory support.^1,2

Myasthenia gravis is characterized by auto-antibodies against the postsynaptic membrane of the neuromuscular junction. Most patients have antibodies to the extracellular portion of the acetylcholine receptor; a small number of patients have antibodies against a muscle-specific tyrosine kinase that interacts with this receptor.

About 15% of patients with myasthenia gravis have a thymoma thought to be involved in the pathogenesis of the disease. Treatments include immune suppressive therapy and thymectomy.

Our patient has a history of thymic lesion resection, but her antibody workup for myasthenia gravis was negative.

Hypothyroidism

Hypothyroidism, the most common disorder of the thyroid gland, is especially prevalent in women.³ Its common symptoms include fatigue, exercise intolerance, muscle weakness, cramps, and stiffness.

Both the TSH and the free thyroxine (T₄) level must be measured to diagnose hypothyroidism. This information can also help differentiate primary hypothyroidism (ie, due to a defect in the thyroid gland) from secondary hypothyroidism (ie, due to a defect in the pituitary gland). Elevated TSH with low free T₄ levels indicates primary thyroid failure, whereas the combination of a normal or low TSH and a low free T₄ usually indicates pituitary failure. Subclinical hypothyroidism is characterized by mildly to moderately elevated TSH, but total T₄ and free T₄ values are still within the reference range. Replacement therapy is with levothyroxine.^3–6

Our patient has a history of hypothyroidism, which could explain her muscle weakness, but she is currently on replacement therapy, and her TSH level on admission was normal.

Dermatomyositis-polymyositis is characterized by proximal muscle weakness, creatine kinase elevation, erythema on sun-exposed skin, heliotrope rash, and Gottron papules. It occurs mostly in women after the second decade of life. Some medications have been implicated in its pathogenesis, such as statins, fibrates, hydroxyurea, penicillamine, and omeprazole (Prilosec).⁷

In a middle-aged patient, this diagnosis should prompt a search for cancer, especially of the gastrointestinal system, breast, and lung.⁸ Cancer can arise up to 3 years after the diagnosis of dermatomyositis or polymyositis.

Antisynthetase antibody syndrome is suspected if the patient is positive for antisynthetase antibody and has the following manifestations: acute onset of disease, constitutional symptoms, interstitial lung disease, inflammatory arthritis, mechanic’s hands (thickened, cracked skin on the palmar aspect of the thumb and index finger), and Raynaud phenomenon.^4,8,9

The diagnosis is made by a thorough clinical evaluation. Electromyography can show an inflammatory pattern of myopathy. The gold standard test for this diagnosis is muscle biopsy.

Our patient has a normal creatine kinase level, which excludes the diagnosis of dermatomyositis-polymyositis.

Statin-induced myopathy

Up to 10% of patients taking statins develop myalgia. Rhabdomyolysis, the extreme form of myopathy, is rare.

The exact mechanism of statin-induced myopathy remains unclear; mitochondrial dysfunction, cholesterol composition of cell membranes, and coenzyme Q10 deficiency have been proposed.

Risk factors for statin-induced myopathy include female sex, older age, higher doses of statins, a family history of statin-induced myopathy, and hypothyroidism. Drugs that increase the risk include fibric acid derivatives, macrolides, and amiodarone (Cordarone). If a statin and any of the above drugs are both required, certain statins—ie, pravastatin (Pravachol) and rosuvastatin—are recommended, since they are the statins least likely to cause rhabdomyolysis.^5,7,10–12

The combination of fluvastatin (Lescol) and gemfibrozil (Lopid) has also been found to be safe.¹³ In a crossover study in 17 patients, no significant difference was seen in the area under the curve for plasma concentration over time, in the maximum plasma concentration, or in the time to maximum concentration with the combination vs with each drug alone.¹³

Our patient is taking a statin and has hypothyroidism, which increases the risk of statin-induced myopathy. However, her creatine kinase level is normal.

Cushing syndrome

Cushing syndrome (hypercortisolism) is one of the most challenging endocrine diseases to diagnose. Most of its clinical features overlap with those of common diseases, and some patients have an atypical clinical presentation with only isolated symptoms. Further, its presentation can be subtle, with weight gain, amenorrhea, muscle weakness, and easy bruisability. Acne, moon facies, plethora, abdominal striae, and purpura are other common signs. It is three to 10 times more common in women than in men.

Cushing syndrome can be classified according to whether or not the excess cortisol secretion depends on corticotropin (formerly called adrenocorticotropic hormone or ACTH) (Figure  1). In corticotropin-dependent cases, the most common cause is pituitary adenoma. (When Cushing syndrome is due to excessive pituitary secretion of corticotropin, which in turn stimulates the adrenal glands to secrete excessive amounts of cortisol, it is called Cushing disease). Other causes of corticotropin-dependent Cushing syndrome are ectopic corticotropin-producing tumors such as carcinoid tumors or medullary thyroid cancers. Corticotropin-independent Cushing syndrome can be caused by adrenal adenomas, adrenal carcinoma, and bilateral primary micronodular or macronodular adrenocortical hyperplasia.^14–17

However, the most common cause of Cushing syndrome is glucocorticoid therapy.

BACK TO OUR PATIENT: HER CONDITION DETERIORATES

Our patient’s physical condition deteriorates, she develops respiratory distress, and she is admitted to the medical intensive care unit. Her mental status also deteriorates, and she becomes lethargic and unresponsive.

She is intubated to protect her airway. After this, she develops hypotension that does not respond to fluid resuscitation and that requires vasopressors. Her condition continues to worsen as she develops acute kidney injury and disseminated intravascular coagulation. Her vesicular rash becomes more widespread, involving the entire trunk.

A workup for sepsis is initiated, but her initial blood and urine cultures are negative. Chest radiography does not reveal any infiltrates. No other source of an infection is found.

Varicella zoster is isolated on viral culture of a specimen obtained from the rash, and a polymerase chain reaction test of her blood shows cytomegalovirus DNA (64,092 copies per mL). Immune suppression is suspected, so a CD4 count is ordered (Table 2). Serologic tests for human immunodeficiency virus are negative.

What could have caused our patient to have muscle weakness in addition to disseminated zoster with cytomegalovirus viremia?

The diagnosis here is Cushing syndrome.

HOW TO TEST FOR CUSHING SYNDROME

2. In any practice, you may meet many perimenopausal women who have complaints of weight gain, amenorrhea, and acne. How can you determine if this is Cushing syndrome? What are the screening tests?

• 24-Hour urinary cortisol excretion
• A late-night salivary cortisol level
• A low-dose dexamethasone suppression test
• All of the above
• None of the above

Any of the tests listed here can be used to determine whether this is truly Cushing syndrome.

24-Hour urinary cortisol excretion has a reference range of 20 to 100 μg/24 hours. However, results may be falsely high in patients who are depressed or who abuse alcohol.

The late-night salivary cortisol level is another useful test.^14,16,18 Patients with Cushing syndrome are found to have high late-night salivary cortisol levels as compared with normal people, indicating the loss of natural circadian rhythm.^14,16,18

The low-dose dexamethasone suppression test, as first described by Liddle in 1960,¹⁹ involved giving dexamethasone 0.5 mg by mouth every 6 hours for 48 hours and measuring the serum cortisol level 6 hours after the last dose. In healthy people, this low dose of dexamethasone suppresses the production of corticotropin by the pituitary gland and in turn the production of cortisol, but in patients with Cushing syndrome the cortisol level remains high. An alternative is the overnight 1-mg dexamethasone suppression test—ie, giving 1 mg of dexamethasone at 11:00 pm and measuring the serum cortisol level early the next morning. Failure of the cortisol level to drop to less than 1.8 μg/dL suggests Cushing syndrome and warrants a complete evaluation for it.

Confirmatory testing is sometimes needed if patients have mild abnormalities in their screening tests. A combination low-dose dexamethasone suppression test and corticotropin-releasing hormone test can be used to differentiate Cushing syndrome from pseudo-Cushing syndrome. This is performed by giving dexamethasone orally 0.5 mg every 6 hours for 48 hours and then giving ovine-sequence corticotropin-releasing hormone 1 μg/kg intravenously 2 hours after the last dose of dexamethasone. The plasma cortisol value 15 minutes after the dose of corticotropin-releasing hormone is greater than 1.4 μg/dL (38 nmol/L) in patients with Cushing syndrome but remains low in patients with pseudo-Cushing syndrome.

Is this corticotropin-dependent or corticotropin-independent?

Once Cushing syndrome is diagnosed by one of the screening methods described above, the source of the excess glucocorticoids needs to be determined. Measuring the serum corticotropin level early in the morning would be the next step.

A low corticotropin level (< 10 pg/mL) indicates a corticotropin-independent source, most likely in the adrenal glands. Hence, computed tomography or magnetic resonance imaging (MRI) of the adrenal glands is warranted. Of note: adrenal incidentalomas are quite common, present in 5% of the general population, and a lesion on the adrenal gland does not prove that the patient has primary adrenal disease.^16,20

IS THE EXCESS CORTICOTROPIN FROM A PITUITARY OR AN ECTOPIC SOURCE?

3. If the corticotropin level is elevated, how can you determine if it is from the pituitary or from an ectopic source?

• MRI of the pituitary gland
• High-dose dexamethasone suppression test
• Corticotropin-releasing hormone stimulation test
• Bilateral inferior petrosal sinus sampling

If the corticotropin level is high (> 10 pg/mL), it is of paramount importance to determine whether the corticotropin comes from the pituitary gland or from an ectopic source.

MRI of the pituitary gland should be done in patients with suspected corticotropin-dependent Cushing syndrome. However, MRI may be negative in 50% of patients with Cushing disease, and it should therefore not be used for screening. In addition, 10% of the population may have pituitary incidentalomas on MRI.

Most cases of corticotropin-dependent Cushing syndrome are caused by microadenomas (smaller than 1 cm), while a few cases are caused by macroadenomas (larger than 1 cm). If a microadenoma is found on MRI, further testing with bilateral inferior petrosal sinus sampling is recommended (described below); if a macroadenoma is found, then no further testing is required.^21,22 In fact, patients who have biochemical findings compatible with Cushing disease (ie, due to an overactive pituitary) and who have an adenoma larger than 6 mm do not require further evaluation.²³

A high-dose dexamethasone suppression test involves giving 8 mg of dexamethasone in the evening and measuring the cortisol level the next morning. If the cortisol level declines to 50% of the baseline level after this dose, this suggests a pituitary cause.

Corticotropin-releasing hormone stimulation testing. In most cases of pituitary tumors and a few cases of ectopic corticotropin-secreting tumors, giving corticotropin-releasing hormone leads to an increase in serum corticotropin and cortisol levels. In contrast, these levels do not respond to corticotropin-releasing hormone stimulation if the problem is in the adrenal gland. The test is performed by giving 1 μg/kg or 100 μg synthetic or human corticotropin-releasing hormone. A 35% to 50% increase above baseline in corticotropin suggests a pituitary cause.²³

Bilateral inferior petrosal sinus sampling can be used to confirm a pituitary source, as it is the gold standard for differentiating ectopic from pituitary corticotropin production. Once this is confirmed, a neurosurgical consult is warranted.^16,18

This procedure is usually done by advancing a sheath from the femoral vein to reach the inferior petrosal sinuses. Blood samples are obtained from both the inferior petrosal sinuses and from a peripheral vein to measure corticotropin levels before and after giving corticotropin-releasing hormone (1 μg/kg). Before corticotropin-releasing hormone is given, a gradient of central-peripheral corticotropin levels of 2.0 or greater indicates a pituitary source. With ectopic corticotropin production, the corticotropin gradient is usually less than 1.5. Corticotropin-releasing hormone is given to increase the sensitivity: after it is given, a gradient of 3.0 or greater is considered indicative of Cushing disease.²⁴

If the corticotropin level is elevated and the above tests indicate ectopic production, the source should be sought. The most common site of ectopic corticotropin production is the chest. Common causes are bronchial, thymic, and pancreatic carcinoid tumors. Other causes are small-cell lung cancer, medullary cell cancer, and pheochromocytoma.^15,18,25

BACK TO OUR PATIENT

Our patient’s further laboratory results are listed in Table 3.

She has elevated 24-hour urinary cortisol excretion, consistent with Cushing syndrome. Her corticotropin level is elevated, which rules out an adrenal cause. Her 5-HIAA (a serotonin breakdown product) and calcitonin levels are also elevated, suggesting either medullary thyroid cancer or a carcinoid tumor. She also has a mild elevation of dehydroepiandrosterone sulfate, which is consistent with corticotropin-dependent Cushing syndrome.

Our patient’s elevated levels of cortisol were the cause of her muscle weakness and severe immune deficiency, which in turn led to cytomegalovirus viremia and sepsis. Cushing syndrome usually causes hypertension, especially in cases of ectopic corticotropin production. However, our patient was normotensive on admission and then developed cytomegalovirus sepsis, which led to hypotension and shock.

Immune suppression is a well-known effect of glucocorticoids.^26–28 Kronfol et al²⁸ found that CD4 and CD8 counts and the CD4-to-CD8 ratio were low in patients with Cushing syndrome, and natural killer cell activity was suppressed. Opportunistic infections have been described in patients with Cushing syndrome.^26,27,29

MANAGEMENT OF CUSHING SYNDROME

Management of Cushing syndrome should be tailored after determining its source.

A neurosurgical consultation is warranted in cases of pituitary adenoma, with surgical resection of the adrenal source or ectopic tumor if feasible.²⁵

Medical management is recommended if surgical resection is not possible.^30,31 Several drugs can be used to inhibit cortisol synthesis in this situation.^30,32

Adrenal-acting agents

Aminoglutethimide (Cytadren) acts by blocking the conversion of cholesterol to pregnenolone, a precursor of cortisol. The dosage is 250 mg twice or three times a day. This drug is no longer available in the United States.

Ketoconazole (Nizoral) inhibits side-chain cleavage, 11-beta hydroxylase, and 17-alpha hydroxylase, thus inhibiting cortisol synthesis; it also inhibits corticotropin secretion. The dosage is 200 to 400 mg three times a day.

Metyrapone (Metopirone) blocks 11-beta-hydroxylation of deoxycortisol, the reaction that produces cortisol. The dosage is 500 to 750 mg three times a day. This drug can be obtained only from the manufacturer and only on a named-patient basis.

Etomidate (Amidate), an anesthetic drug, also blocks 11-beta-hydroxylation of deoxycortisol. It is given intravenously at a rate of 0.3 mg/kg per hour.

Centrally acting agents

Cabergoline (Dostinex). It is believed that corticotropin-producing pituitary tumors express D2 receptors. Cabergoline is a dopamine agonist that has been used in patients with Cushing disease. The dosage is 0.5 to 7 mg/week.

Pasireotide is still investigational. It is a somatostatin receptor agonist given subcutaneously for 15 consecutive days to patients with Cushing disease.

Glucocorticoid receptor antagonist

Mifepristone (Mifeprex) is a progesterone receptor and glucocorticoid II receptor antagonist that is being investigated in the treatment of persistent or recurrent Cushing disease. It is not yet approved by the US Food and Drug Administration for this indication.

BACK TO OUR PATIENT

The patient was too ill to undergo additional imaging, including octreotide scanning to identify an ectopic corticotropin-secreting tumor. She was medically treated with intravenous etomidate to reduce her cortisol level.^30,31

Unfortunately, our patient died of multiorgan failure. The exact site of her ectopic corticotropin-producing tumor was never identified, and no autopsy was done.

Our patient, a 56-year-old woman, presents with proximal muscle weakness in all four limbs. It started a few months ago and has gradually become severe, so that she now has difficulty rising from a seated position and has trouble opening jars. She has fallen several times. She says she has no muscle pain, difficulty swallowing, or difficulty breathing.

She sought medical attention at another hospital and was found to be hypothyroid, with a thyrotropin (thyroid-stimulating hormone [TSH]) level of 38 μU/mL (reference range 0.4–5.5), for which she was started on levothyroxine (Synthroid) 100 μg daily. She also had a low serum potassium level, for which potassium supplements and spironolactone (Aldactone) were started. She was taking furosemide (Lasix) 20 mg/day at the time.

Despite the thyroid replacement therapy, she continued to become weaker and had more falls. She also noticed a new, nonpainful rash on her lower abdomen.

Review of systems

• Night sweats
• Leg swelling
• Puffiness and discoloration around the eyes, with easy bruisability.

Medical history

• Diabetes mellitus
• Seizures in the 1970s
• Resection of a thymic tumor in 2003 (the exact pathology is unknown)
• Cirrhosis of unknown etiology
• No known history of hypertension
• No history of alcohol or intravenous drug use
• Quit smoking many years ago
• Coronary artery bypass surgery in 2003
• One sibling with myasthenia gravis.

Medications

• Levothyroxine
• Rosuvastatin (Crestor)
• Omeprazole (Prilosec)
• Spironolactone
• Furosemide
• Potassium chloride
• Metoprolol tartrate (Lopressor)
• Metformin (Glucophage)
• Ramipril (Altace).

Physical examination

She is hemodynamically stable and is not hypertensive. Her thyroid is not enlarged. Her lungs are clear to auscultation. Her heart sounds are normal, except for a nonradiating pansystolic murmur most audible at the apex.

Her abdomen is soft and is not distended. Her abdominal rash has a dermatomal distribution consistent with an L1 distribution, with vesicles over an erythematous base. Purpuric lesions are noted over her lower extremities.

Her leg strength is 3 on a scale of 5 on both sides; her arm strength is normal. Ankle and knee reflexes are absent bilaterally.

Initial laboratory analysis

PROGRESSIVE MUSCLE WEAKNESS

1. What are possible causes of her muscle weakness?

• Myasthenia gravis
• Hypothyroidism
• Dermatomyositis-polymyositis
• Drug-induced myopathy
• Cushing syndrome
• All of the above

All of these are potential causes of muscle weakness.

Myasthenia gravis

Myasthenia gravis, an autoimmune disease, can affect people of all ages and either sex. It presents with muscle weakness and fatigability, which characteristically fluctuate during the day. Some patients present in crisis with respiratory failure, which may require ventilatory support.^1,2

Myasthenia gravis is characterized by auto-antibodies against the postsynaptic membrane of the neuromuscular junction. Most patients have antibodies to the extracellular portion of the acetylcholine receptor; a small number of patients have antibodies against a muscle-specific tyrosine kinase that interacts with this receptor.

About 15% of patients with myasthenia gravis have a thymoma thought to be involved in the pathogenesis of the disease. Treatments include immune suppressive therapy and thymectomy.

Our patient has a history of thymic lesion resection, but her antibody workup for myasthenia gravis was negative.

Hypothyroidism

Hypothyroidism, the most common disorder of the thyroid gland, is especially prevalent in women.³ Its common symptoms include fatigue, exercise intolerance, muscle weakness, cramps, and stiffness.

Both the TSH and the free thyroxine (T₄) level must be measured to diagnose hypothyroidism. This information can also help differentiate primary hypothyroidism (ie, due to a defect in the thyroid gland) from secondary hypothyroidism (ie, due to a defect in the pituitary gland). Elevated TSH with low free T₄ levels indicates primary thyroid failure, whereas the combination of a normal or low TSH and a low free T₄ usually indicates pituitary failure. Subclinical hypothyroidism is characterized by mildly to moderately elevated TSH, but total T₄ and free T₄ values are still within the reference range. Replacement therapy is with levothyroxine.^3–6

Our patient has a history of hypothyroidism, which could explain her muscle weakness, but she is currently on replacement therapy, and her TSH level on admission was normal.

Dermatomyositis-polymyositis is characterized by proximal muscle weakness, creatine kinase elevation, erythema on sun-exposed skin, heliotrope rash, and Gottron papules. It occurs mostly in women after the second decade of life. Some medications have been implicated in its pathogenesis, such as statins, fibrates, hydroxyurea, penicillamine, and omeprazole (Prilosec).⁷

In a middle-aged patient, this diagnosis should prompt a search for cancer, especially of the gastrointestinal system, breast, and lung.⁸ Cancer can arise up to 3 years after the diagnosis of dermatomyositis or polymyositis.

Antisynthetase antibody syndrome is suspected if the patient is positive for antisynthetase antibody and has the following manifestations: acute onset of disease, constitutional symptoms, interstitial lung disease, inflammatory arthritis, mechanic’s hands (thickened, cracked skin on the palmar aspect of the thumb and index finger), and Raynaud phenomenon.^4,8,9

The diagnosis is made by a thorough clinical evaluation. Electromyography can show an inflammatory pattern of myopathy. The gold standard test for this diagnosis is muscle biopsy.

Our patient has a normal creatine kinase level, which excludes the diagnosis of dermatomyositis-polymyositis.

Statin-induced myopathy

Up to 10% of patients taking statins develop myalgia. Rhabdomyolysis, the extreme form of myopathy, is rare.

The exact mechanism of statin-induced myopathy remains unclear; mitochondrial dysfunction, cholesterol composition of cell membranes, and coenzyme Q10 deficiency have been proposed.

Risk factors for statin-induced myopathy include female sex, older age, higher doses of statins, a family history of statin-induced myopathy, and hypothyroidism. Drugs that increase the risk include fibric acid derivatives, macrolides, and amiodarone (Cordarone). If a statin and any of the above drugs are both required, certain statins—ie, pravastatin (Pravachol) and rosuvastatin—are recommended, since they are the statins least likely to cause rhabdomyolysis.^5,7,10–12

The combination of fluvastatin (Lescol) and gemfibrozil (Lopid) has also been found to be safe.¹³ In a crossover study in 17 patients, no significant difference was seen in the area under the curve for plasma concentration over time, in the maximum plasma concentration, or in the time to maximum concentration with the combination vs with each drug alone.¹³

Our patient is taking a statin and has hypothyroidism, which increases the risk of statin-induced myopathy. However, her creatine kinase level is normal.

Cushing syndrome

Cushing syndrome (hypercortisolism) is one of the most challenging endocrine diseases to diagnose. Most of its clinical features overlap with those of common diseases, and some patients have an atypical clinical presentation with only isolated symptoms. Further, its presentation can be subtle, with weight gain, amenorrhea, muscle weakness, and easy bruisability. Acne, moon facies, plethora, abdominal striae, and purpura are other common signs. It is three to 10 times more common in women than in men.

Cushing syndrome can be classified according to whether or not the excess cortisol secretion depends on corticotropin (formerly called adrenocorticotropic hormone or ACTH) (Figure  1). In corticotropin-dependent cases, the most common cause is pituitary adenoma. (When Cushing syndrome is due to excessive pituitary secretion of corticotropin, which in turn stimulates the adrenal glands to secrete excessive amounts of cortisol, it is called Cushing disease). Other causes of corticotropin-dependent Cushing syndrome are ectopic corticotropin-producing tumors such as carcinoid tumors or medullary thyroid cancers. Corticotropin-independent Cushing syndrome can be caused by adrenal adenomas, adrenal carcinoma, and bilateral primary micronodular or macronodular adrenocortical hyperplasia.^14–17

However, the most common cause of Cushing syndrome is glucocorticoid therapy.

BACK TO OUR PATIENT: HER CONDITION DETERIORATES

Our patient’s physical condition deteriorates, she develops respiratory distress, and she is admitted to the medical intensive care unit. Her mental status also deteriorates, and she becomes lethargic and unresponsive.

She is intubated to protect her airway. After this, she develops hypotension that does not respond to fluid resuscitation and that requires vasopressors. Her condition continues to worsen as she develops acute kidney injury and disseminated intravascular coagulation. Her vesicular rash becomes more widespread, involving the entire trunk.

A workup for sepsis is initiated, but her initial blood and urine cultures are negative. Chest radiography does not reveal any infiltrates. No other source of an infection is found.

Varicella zoster is isolated on viral culture of a specimen obtained from the rash, and a polymerase chain reaction test of her blood shows cytomegalovirus DNA (64,092 copies per mL). Immune suppression is suspected, so a CD4 count is ordered (Table 2). Serologic tests for human immunodeficiency virus are negative.

What could have caused our patient to have muscle weakness in addition to disseminated zoster with cytomegalovirus viremia?

The diagnosis here is Cushing syndrome.

HOW TO TEST FOR CUSHING SYNDROME

2. In any practice, you may meet many perimenopausal women who have complaints of weight gain, amenorrhea, and acne. How can you determine if this is Cushing syndrome? What are the screening tests?

• 24-Hour urinary cortisol excretion
• A late-night salivary cortisol level
• A low-dose dexamethasone suppression test
• All of the above
• None of the above

Any of the tests listed here can be used to determine whether this is truly Cushing syndrome.

24-Hour urinary cortisol excretion has a reference range of 20 to 100 μg/24 hours. However, results may be falsely high in patients who are depressed or who abuse alcohol.

The late-night salivary cortisol level is another useful test.^14,16,18 Patients with Cushing syndrome are found to have high late-night salivary cortisol levels as compared with normal people, indicating the loss of natural circadian rhythm.^14,16,18

The low-dose dexamethasone suppression test, as first described by Liddle in 1960,¹⁹ involved giving dexamethasone 0.5 mg by mouth every 6 hours for 48 hours and measuring the serum cortisol level 6 hours after the last dose. In healthy people, this low dose of dexamethasone suppresses the production of corticotropin by the pituitary gland and in turn the production of cortisol, but in patients with Cushing syndrome the cortisol level remains high. An alternative is the overnight 1-mg dexamethasone suppression test—ie, giving 1 mg of dexamethasone at 11:00 pm and measuring the serum cortisol level early the next morning. Failure of the cortisol level to drop to less than 1.8 μg/dL suggests Cushing syndrome and warrants a complete evaluation for it.

Confirmatory testing is sometimes needed if patients have mild abnormalities in their screening tests. A combination low-dose dexamethasone suppression test and corticotropin-releasing hormone test can be used to differentiate Cushing syndrome from pseudo-Cushing syndrome. This is performed by giving dexamethasone orally 0.5 mg every 6 hours for 48 hours and then giving ovine-sequence corticotropin-releasing hormone 1 μg/kg intravenously 2 hours after the last dose of dexamethasone. The plasma cortisol value 15 minutes after the dose of corticotropin-releasing hormone is greater than 1.4 μg/dL (38 nmol/L) in patients with Cushing syndrome but remains low in patients with pseudo-Cushing syndrome.

Is this corticotropin-dependent or corticotropin-independent?

Once Cushing syndrome is diagnosed by one of the screening methods described above, the source of the excess glucocorticoids needs to be determined. Measuring the serum corticotropin level early in the morning would be the next step.

A low corticotropin level (< 10 pg/mL) indicates a corticotropin-independent source, most likely in the adrenal glands. Hence, computed tomography or magnetic resonance imaging (MRI) of the adrenal glands is warranted. Of note: adrenal incidentalomas are quite common, present in 5% of the general population, and a lesion on the adrenal gland does not prove that the patient has primary adrenal disease.^16,20

IS THE EXCESS CORTICOTROPIN FROM A PITUITARY OR AN ECTOPIC SOURCE?

3. If the corticotropin level is elevated, how can you determine if it is from the pituitary or from an ectopic source?

• MRI of the pituitary gland
• High-dose dexamethasone suppression test
• Corticotropin-releasing hormone stimulation test
• Bilateral inferior petrosal sinus sampling

If the corticotropin level is high (> 10 pg/mL), it is of paramount importance to determine whether the corticotropin comes from the pituitary gland or from an ectopic source.

MRI of the pituitary gland should be done in patients with suspected corticotropin-dependent Cushing syndrome. However, MRI may be negative in 50% of patients with Cushing disease, and it should therefore not be used for screening. In addition, 10% of the population may have pituitary incidentalomas on MRI.

Most cases of corticotropin-dependent Cushing syndrome are caused by microadenomas (smaller than 1 cm), while a few cases are caused by macroadenomas (larger than 1 cm). If a microadenoma is found on MRI, further testing with bilateral inferior petrosal sinus sampling is recommended (described below); if a macroadenoma is found, then no further testing is required.^21,22 In fact, patients who have biochemical findings compatible with Cushing disease (ie, due to an overactive pituitary) and who have an adenoma larger than 6 mm do not require further evaluation.²³

A high-dose dexamethasone suppression test involves giving 8 mg of dexamethasone in the evening and measuring the cortisol level the next morning. If the cortisol level declines to 50% of the baseline level after this dose, this suggests a pituitary cause.

Corticotropin-releasing hormone stimulation testing. In most cases of pituitary tumors and a few cases of ectopic corticotropin-secreting tumors, giving corticotropin-releasing hormone leads to an increase in serum corticotropin and cortisol levels. In contrast, these levels do not respond to corticotropin-releasing hormone stimulation if the problem is in the adrenal gland. The test is performed by giving 1 μg/kg or 100 μg synthetic or human corticotropin-releasing hormone. A 35% to 50% increase above baseline in corticotropin suggests a pituitary cause.²³

Bilateral inferior petrosal sinus sampling can be used to confirm a pituitary source, as it is the gold standard for differentiating ectopic from pituitary corticotropin production. Once this is confirmed, a neurosurgical consult is warranted.^16,18

This procedure is usually done by advancing a sheath from the femoral vein to reach the inferior petrosal sinuses. Blood samples are obtained from both the inferior petrosal sinuses and from a peripheral vein to measure corticotropin levels before and after giving corticotropin-releasing hormone (1 μg/kg). Before corticotropin-releasing hormone is given, a gradient of central-peripheral corticotropin levels of 2.0 or greater indicates a pituitary source. With ectopic corticotropin production, the corticotropin gradient is usually less than 1.5. Corticotropin-releasing hormone is given to increase the sensitivity: after it is given, a gradient of 3.0 or greater is considered indicative of Cushing disease.²⁴

If the corticotropin level is elevated and the above tests indicate ectopic production, the source should be sought. The most common site of ectopic corticotropin production is the chest. Common causes are bronchial, thymic, and pancreatic carcinoid tumors. Other causes are small-cell lung cancer, medullary cell cancer, and pheochromocytoma.^15,18,25

BACK TO OUR PATIENT

Our patient’s further laboratory results are listed in Table 3.

She has elevated 24-hour urinary cortisol excretion, consistent with Cushing syndrome. Her corticotropin level is elevated, which rules out an adrenal cause. Her 5-HIAA (a serotonin breakdown product) and calcitonin levels are also elevated, suggesting either medullary thyroid cancer or a carcinoid tumor. She also has a mild elevation of dehydroepiandrosterone sulfate, which is consistent with corticotropin-dependent Cushing syndrome.

Our patient’s elevated levels of cortisol were the cause of her muscle weakness and severe immune deficiency, which in turn led to cytomegalovirus viremia and sepsis. Cushing syndrome usually causes hypertension, especially in cases of ectopic corticotropin production. However, our patient was normotensive on admission and then developed cytomegalovirus sepsis, which led to hypotension and shock.

Immune suppression is a well-known effect of glucocorticoids.^26–28 Kronfol et al²⁸ found that CD4 and CD8 counts and the CD4-to-CD8 ratio were low in patients with Cushing syndrome, and natural killer cell activity was suppressed. Opportunistic infections have been described in patients with Cushing syndrome.^26,27,29

MANAGEMENT OF CUSHING SYNDROME

Management of Cushing syndrome should be tailored after determining its source.

A neurosurgical consultation is warranted in cases of pituitary adenoma, with surgical resection of the adrenal source or ectopic tumor if feasible.²⁵

Medical management is recommended if surgical resection is not possible.^30,31 Several drugs can be used to inhibit cortisol synthesis in this situation.^30,32

Adrenal-acting agents

Aminoglutethimide (Cytadren) acts by blocking the conversion of cholesterol to pregnenolone, a precursor of cortisol. The dosage is 250 mg twice or three times a day. This drug is no longer available in the United States.

Ketoconazole (Nizoral) inhibits side-chain cleavage, 11-beta hydroxylase, and 17-alpha hydroxylase, thus inhibiting cortisol synthesis; it also inhibits corticotropin secretion. The dosage is 200 to 400 mg three times a day.

Metyrapone (Metopirone) blocks 11-beta-hydroxylation of deoxycortisol, the reaction that produces cortisol. The dosage is 500 to 750 mg three times a day. This drug can be obtained only from the manufacturer and only on a named-patient basis.

Etomidate (Amidate), an anesthetic drug, also blocks 11-beta-hydroxylation of deoxycortisol. It is given intravenously at a rate of 0.3 mg/kg per hour.

Centrally acting agents

Cabergoline (Dostinex). It is believed that corticotropin-producing pituitary tumors express D2 receptors. Cabergoline is a dopamine agonist that has been used in patients with Cushing disease. The dosage is 0.5 to 7 mg/week.

Pasireotide is still investigational. It is a somatostatin receptor agonist given subcutaneously for 15 consecutive days to patients with Cushing disease.

Glucocorticoid receptor antagonist

Mifepristone (Mifeprex) is a progesterone receptor and glucocorticoid II receptor antagonist that is being investigated in the treatment of persistent or recurrent Cushing disease. It is not yet approved by the US Food and Drug Administration for this indication.

BACK TO OUR PATIENT

The patient was too ill to undergo additional imaging, including octreotide scanning to identify an ectopic corticotropin-secreting tumor. She was medically treated with intravenous etomidate to reduce her cortisol level.^30,31

Unfortunately, our patient died of multiorgan failure. The exact site of her ectopic corticotropin-producing tumor was never identified, and no autopsy was done.

Our patient, a 56-year-old woman, presents with proximal muscle weakness in all four limbs. It started a few months ago and has gradually become severe, so that she now has difficulty rising from a seated position and has trouble opening jars. She has fallen several times. She says she has no muscle pain, difficulty swallowing, or difficulty breathing.

She sought medical attention at another hospital and was found to be hypothyroid, with a thyrotropin (thyroid-stimulating hormone [TSH]) level of 38 μU/mL (reference range 0.4–5.5), for which she was started on levothyroxine (Synthroid) 100 μg daily. She also had a low serum potassium level, for which potassium supplements and spironolactone (Aldactone) were started. She was taking furosemide (Lasix) 20 mg/day at the time.

Despite the thyroid replacement therapy, she continued to become weaker and had more falls. She also noticed a new, nonpainful rash on her lower abdomen.

Review of systems

• Night sweats
• Leg swelling
• Puffiness and discoloration around the eyes, with easy bruisability.

Medical history

• Diabetes mellitus
• Seizures in the 1970s
• Resection of a thymic tumor in 2003 (the exact pathology is unknown)
• Cirrhosis of unknown etiology
• No known history of hypertension
• No history of alcohol or intravenous drug use
• Quit smoking many years ago
• Coronary artery bypass surgery in 2003
• One sibling with myasthenia gravis.

Medications

• Levothyroxine
• Rosuvastatin (Crestor)
• Omeprazole (Prilosec)
• Spironolactone
• Furosemide
• Potassium chloride
• Metoprolol tartrate (Lopressor)
• Metformin (Glucophage)
• Ramipril (Altace).

Physical examination

She is hemodynamically stable and is not hypertensive. Her thyroid is not enlarged. Her lungs are clear to auscultation. Her heart sounds are normal, except for a nonradiating pansystolic murmur most audible at the apex.

Her abdomen is soft and is not distended. Her abdominal rash has a dermatomal distribution consistent with an L1 distribution, with vesicles over an erythematous base. Purpuric lesions are noted over her lower extremities.

Her leg strength is 3 on a scale of 5 on both sides; her arm strength is normal. Ankle and knee reflexes are absent bilaterally.

Initial laboratory analysis

PROGRESSIVE MUSCLE WEAKNESS

1. What are possible causes of her muscle weakness?

• Myasthenia gravis
• Hypothyroidism
• Dermatomyositis-polymyositis
• Drug-induced myopathy
• Cushing syndrome
• All of the above

All of these are potential causes of muscle weakness.

Myasthenia gravis

Myasthenia gravis, an autoimmune disease, can affect people of all ages and either sex. It presents with muscle weakness and fatigability, which characteristically fluctuate during the day. Some patients present in crisis with respiratory failure, which may require ventilatory support.^1,2

Myasthenia gravis is characterized by auto-antibodies against the postsynaptic membrane of the neuromuscular junction. Most patients have antibodies to the extracellular portion of the acetylcholine receptor; a small number of patients have antibodies against a muscle-specific tyrosine kinase that interacts with this receptor.

About 15% of patients with myasthenia gravis have a thymoma thought to be involved in the pathogenesis of the disease. Treatments include immune suppressive therapy and thymectomy.

Our patient has a history of thymic lesion resection, but her antibody workup for myasthenia gravis was negative.

Hypothyroidism

Hypothyroidism, the most common disorder of the thyroid gland, is especially prevalent in women.³ Its common symptoms include fatigue, exercise intolerance, muscle weakness, cramps, and stiffness.

Both the TSH and the free thyroxine (T₄) level must be measured to diagnose hypothyroidism. This information can also help differentiate primary hypothyroidism (ie, due to a defect in the thyroid gland) from secondary hypothyroidism (ie, due to a defect in the pituitary gland). Elevated TSH with low free T₄ levels indicates primary thyroid failure, whereas the combination of a normal or low TSH and a low free T₄ usually indicates pituitary failure. Subclinical hypothyroidism is characterized by mildly to moderately elevated TSH, but total T₄ and free T₄ values are still within the reference range. Replacement therapy is with levothyroxine.^3–6

Our patient has a history of hypothyroidism, which could explain her muscle weakness, but she is currently on replacement therapy, and her TSH level on admission was normal.

Dermatomyositis-polymyositis is characterized by proximal muscle weakness, creatine kinase elevation, erythema on sun-exposed skin, heliotrope rash, and Gottron papules. It occurs mostly in women after the second decade of life. Some medications have been implicated in its pathogenesis, such as statins, fibrates, hydroxyurea, penicillamine, and omeprazole (Prilosec).⁷

In a middle-aged patient, this diagnosis should prompt a search for cancer, especially of the gastrointestinal system, breast, and lung.⁸ Cancer can arise up to 3 years after the diagnosis of dermatomyositis or polymyositis.

Antisynthetase antibody syndrome is suspected if the patient is positive for antisynthetase antibody and has the following manifestations: acute onset of disease, constitutional symptoms, interstitial lung disease, inflammatory arthritis, mechanic’s hands (thickened, cracked skin on the palmar aspect of the thumb and index finger), and Raynaud phenomenon.^4,8,9

The diagnosis is made by a thorough clinical evaluation. Electromyography can show an inflammatory pattern of myopathy. The gold standard test for this diagnosis is muscle biopsy.

Our patient has a normal creatine kinase level, which excludes the diagnosis of dermatomyositis-polymyositis.

Statin-induced myopathy

Up to 10% of patients taking statins develop myalgia. Rhabdomyolysis, the extreme form of myopathy, is rare.

The exact mechanism of statin-induced myopathy remains unclear; mitochondrial dysfunction, cholesterol composition of cell membranes, and coenzyme Q10 deficiency have been proposed.

Risk factors for statin-induced myopathy include female sex, older age, higher doses of statins, a family history of statin-induced myopathy, and hypothyroidism. Drugs that increase the risk include fibric acid derivatives, macrolides, and amiodarone (Cordarone). If a statin and any of the above drugs are both required, certain statins—ie, pravastatin (Pravachol) and rosuvastatin—are recommended, since they are the statins least likely to cause rhabdomyolysis.^5,7,10–12

The combination of fluvastatin (Lescol) and gemfibrozil (Lopid) has also been found to be safe.¹³ In a crossover study in 17 patients, no significant difference was seen in the area under the curve for plasma concentration over time, in the maximum plasma concentration, or in the time to maximum concentration with the combination vs with each drug alone.¹³

Our patient is taking a statin and has hypothyroidism, which increases the risk of statin-induced myopathy. However, her creatine kinase level is normal.

Cushing syndrome

Cushing syndrome (hypercortisolism) is one of the most challenging endocrine diseases to diagnose. Most of its clinical features overlap with those of common diseases, and some patients have an atypical clinical presentation with only isolated symptoms. Further, its presentation can be subtle, with weight gain, amenorrhea, muscle weakness, and easy bruisability. Acne, moon facies, plethora, abdominal striae, and purpura are other common signs. It is three to 10 times more common in women than in men.

Cushing syndrome can be classified according to whether or not the excess cortisol secretion depends on corticotropin (formerly called adrenocorticotropic hormone or ACTH) (Figure  1). In corticotropin-dependent cases, the most common cause is pituitary adenoma. (When Cushing syndrome is due to excessive pituitary secretion of corticotropin, which in turn stimulates the adrenal glands to secrete excessive amounts of cortisol, it is called Cushing disease). Other causes of corticotropin-dependent Cushing syndrome are ectopic corticotropin-producing tumors such as carcinoid tumors or medullary thyroid cancers. Corticotropin-independent Cushing syndrome can be caused by adrenal adenomas, adrenal carcinoma, and bilateral primary micronodular or macronodular adrenocortical hyperplasia.^14–17

However, the most common cause of Cushing syndrome is glucocorticoid therapy.

BACK TO OUR PATIENT: HER CONDITION DETERIORATES

Our patient’s physical condition deteriorates, she develops respiratory distress, and she is admitted to the medical intensive care unit. Her mental status also deteriorates, and she becomes lethargic and unresponsive.

She is intubated to protect her airway. After this, she develops hypotension that does not respond to fluid resuscitation and that requires vasopressors. Her condition continues to worsen as she develops acute kidney injury and disseminated intravascular coagulation. Her vesicular rash becomes more widespread, involving the entire trunk.

A workup for sepsis is initiated, but her initial blood and urine cultures are negative. Chest radiography does not reveal any infiltrates. No other source of an infection is found.

Varicella zoster is isolated on viral culture of a specimen obtained from the rash, and a polymerase chain reaction test of her blood shows cytomegalovirus DNA (64,092 copies per mL). Immune suppression is suspected, so a CD4 count is ordered (Table 2). Serologic tests for human immunodeficiency virus are negative.

What could have caused our patient to have muscle weakness in addition to disseminated zoster with cytomegalovirus viremia?

The diagnosis here is Cushing syndrome.

HOW TO TEST FOR CUSHING SYNDROME

2. In any practice, you may meet many perimenopausal women who have complaints of weight gain, amenorrhea, and acne. How can you determine if this is Cushing syndrome? What are the screening tests?

• 24-Hour urinary cortisol excretion
• A late-night salivary cortisol level
• A low-dose dexamethasone suppression test
• All of the above
• None of the above

Any of the tests listed here can be used to determine whether this is truly Cushing syndrome.

24-Hour urinary cortisol excretion has a reference range of 20 to 100 μg/24 hours. However, results may be falsely high in patients who are depressed or who abuse alcohol.

The late-night salivary cortisol level is another useful test.^14,16,18 Patients with Cushing syndrome are found to have high late-night salivary cortisol levels as compared with normal people, indicating the loss of natural circadian rhythm.^14,16,18

The low-dose dexamethasone suppression test, as first described by Liddle in 1960,¹⁹ involved giving dexamethasone 0.5 mg by mouth every 6 hours for 48 hours and measuring the serum cortisol level 6 hours after the last dose. In healthy people, this low dose of dexamethasone suppresses the production of corticotropin by the pituitary gland and in turn the production of cortisol, but in patients with Cushing syndrome the cortisol level remains high. An alternative is the overnight 1-mg dexamethasone suppression test—ie, giving 1 mg of dexamethasone at 11:00 pm and measuring the serum cortisol level early the next morning. Failure of the cortisol level to drop to less than 1.8 μg/dL suggests Cushing syndrome and warrants a complete evaluation for it.

Confirmatory testing is sometimes needed if patients have mild abnormalities in their screening tests. A combination low-dose dexamethasone suppression test and corticotropin-releasing hormone test can be used to differentiate Cushing syndrome from pseudo-Cushing syndrome. This is performed by giving dexamethasone orally 0.5 mg every 6 hours for 48 hours and then giving ovine-sequence corticotropin-releasing hormone 1 μg/kg intravenously 2 hours after the last dose of dexamethasone. The plasma cortisol value 15 minutes after the dose of corticotropin-releasing hormone is greater than 1.4 μg/dL (38 nmol/L) in patients with Cushing syndrome but remains low in patients with pseudo-Cushing syndrome.

Is this corticotropin-dependent or corticotropin-independent?

Once Cushing syndrome is diagnosed by one of the screening methods described above, the source of the excess glucocorticoids needs to be determined. Measuring the serum corticotropin level early in the morning would be the next step.

A low corticotropin level (< 10 pg/mL) indicates a corticotropin-independent source, most likely in the adrenal glands. Hence, computed tomography or magnetic resonance imaging (MRI) of the adrenal glands is warranted. Of note: adrenal incidentalomas are quite common, present in 5% of the general population, and a lesion on the adrenal gland does not prove that the patient has primary adrenal disease.^16,20

IS THE EXCESS CORTICOTROPIN FROM A PITUITARY OR AN ECTOPIC SOURCE?

3. If the corticotropin level is elevated, how can you determine if it is from the pituitary or from an ectopic source?

• MRI of the pituitary gland
• High-dose dexamethasone suppression test
• Corticotropin-releasing hormone stimulation test
• Bilateral inferior petrosal sinus sampling

If the corticotropin level is high (> 10 pg/mL), it is of paramount importance to determine whether the corticotropin comes from the pituitary gland or from an ectopic source.

MRI of the pituitary gland should be done in patients with suspected corticotropin-dependent Cushing syndrome. However, MRI may be negative in 50% of patients with Cushing disease, and it should therefore not be used for screening. In addition, 10% of the population may have pituitary incidentalomas on MRI.

Most cases of corticotropin-dependent Cushing syndrome are caused by microadenomas (smaller than 1 cm), while a few cases are caused by macroadenomas (larger than 1 cm). If a microadenoma is found on MRI, further testing with bilateral inferior petrosal sinus sampling is recommended (described below); if a macroadenoma is found, then no further testing is required.^21,22 In fact, patients who have biochemical findings compatible with Cushing disease (ie, due to an overactive pituitary) and who have an adenoma larger than 6 mm do not require further evaluation.²³

A high-dose dexamethasone suppression test involves giving 8 mg of dexamethasone in the evening and measuring the cortisol level the next morning. If the cortisol level declines to 50% of the baseline level after this dose, this suggests a pituitary cause.

Corticotropin-releasing hormone stimulation testing. In most cases of pituitary tumors and a few cases of ectopic corticotropin-secreting tumors, giving corticotropin-releasing hormone leads to an increase in serum corticotropin and cortisol levels. In contrast, these levels do not respond to corticotropin-releasing hormone stimulation if the problem is in the adrenal gland. The test is performed by giving 1 μg/kg or 100 μg synthetic or human corticotropin-releasing hormone. A 35% to 50% increase above baseline in corticotropin suggests a pituitary cause.²³

Bilateral inferior petrosal sinus sampling can be used to confirm a pituitary source, as it is the gold standard for differentiating ectopic from pituitary corticotropin production. Once this is confirmed, a neurosurgical consult is warranted.^16,18

This procedure is usually done by advancing a sheath from the femoral vein to reach the inferior petrosal sinuses. Blood samples are obtained from both the inferior petrosal sinuses and from a peripheral vein to measure corticotropin levels before and after giving corticotropin-releasing hormone (1 μg/kg). Before corticotropin-releasing hormone is given, a gradient of central-peripheral corticotropin levels of 2.0 or greater indicates a pituitary source. With ectopic corticotropin production, the corticotropin gradient is usually less than 1.5. Corticotropin-releasing hormone is given to increase the sensitivity: after it is given, a gradient of 3.0 or greater is considered indicative of Cushing disease.²⁴

If the corticotropin level is elevated and the above tests indicate ectopic production, the source should be sought. The most common site of ectopic corticotropin production is the chest. Common causes are bronchial, thymic, and pancreatic carcinoid tumors. Other causes are small-cell lung cancer, medullary cell cancer, and pheochromocytoma.^15,18,25

BACK TO OUR PATIENT

Our patient’s further laboratory results are listed in Table 3.

She has elevated 24-hour urinary cortisol excretion, consistent with Cushing syndrome. Her corticotropin level is elevated, which rules out an adrenal cause. Her 5-HIAA (a serotonin breakdown product) and calcitonin levels are also elevated, suggesting either medullary thyroid cancer or a carcinoid tumor. She also has a mild elevation of dehydroepiandrosterone sulfate, which is consistent with corticotropin-dependent Cushing syndrome.

Our patient’s elevated levels of cortisol were the cause of her muscle weakness and severe immune deficiency, which in turn led to cytomegalovirus viremia and sepsis. Cushing syndrome usually causes hypertension, especially in cases of ectopic corticotropin production. However, our patient was normotensive on admission and then developed cytomegalovirus sepsis, which led to hypotension and shock.

Immune suppression is a well-known effect of glucocorticoids.^26–28 Kronfol et al²⁸ found that CD4 and CD8 counts and the CD4-to-CD8 ratio were low in patients with Cushing syndrome, and natural killer cell activity was suppressed. Opportunistic infections have been described in patients with Cushing syndrome.^26,27,29

MANAGEMENT OF CUSHING SYNDROME

Management of Cushing syndrome should be tailored after determining its source.

A neurosurgical consultation is warranted in cases of pituitary adenoma, with surgical resection of the adrenal source or ectopic tumor if feasible.²⁵

Medical management is recommended if surgical resection is not possible.^30,31 Several drugs can be used to inhibit cortisol synthesis in this situation.^30,32

Adrenal-acting agents

Aminoglutethimide (Cytadren) acts by blocking the conversion of cholesterol to pregnenolone, a precursor of cortisol. The dosage is 250 mg twice or three times a day. This drug is no longer available in the United States.

Ketoconazole (Nizoral) inhibits side-chain cleavage, 11-beta hydroxylase, and 17-alpha hydroxylase, thus inhibiting cortisol synthesis; it also inhibits corticotropin secretion. The dosage is 200 to 400 mg three times a day.

Metyrapone (Metopirone) blocks 11-beta-hydroxylation of deoxycortisol, the reaction that produces cortisol. The dosage is 500 to 750 mg three times a day. This drug can be obtained only from the manufacturer and only on a named-patient basis.

Etomidate (Amidate), an anesthetic drug, also blocks 11-beta-hydroxylation of deoxycortisol. It is given intravenously at a rate of 0.3 mg/kg per hour.

Centrally acting agents

Cabergoline (Dostinex). It is believed that corticotropin-producing pituitary tumors express D2 receptors. Cabergoline is a dopamine agonist that has been used in patients with Cushing disease. The dosage is 0.5 to 7 mg/week.

Pasireotide is still investigational. It is a somatostatin receptor agonist given subcutaneously for 15 consecutive days to patients with Cushing disease.

Glucocorticoid receptor antagonist

Mifepristone (Mifeprex) is a progesterone receptor and glucocorticoid II receptor antagonist that is being investigated in the treatment of persistent or recurrent Cushing disease. It is not yet approved by the US Food and Drug Administration for this indication.

BACK TO OUR PATIENT

The patient was too ill to undergo additional imaging, including octreotide scanning to identify an ectopic corticotropin-secreting tumor. She was medically treated with intravenous etomidate to reduce her cortisol level.^30,31

Unfortunately, our patient died of multiorgan failure. The exact site of her ectopic corticotropin-producing tumor was never identified, and no autopsy was done.