The mark of any great society is balance—balance between the production realized today and the preservation of “production capacity” to ensure the same or greater production in the future. HM is not exempt from this fundamental tenet. What we do now in the way of advancing quality, efficiency, and patient safety will matter little if our contributions are not sustained by the generation that follows us.

It is tempting to think that the issue of how we train residents is germane only to universities, but the reality is that it affects us all. There are 126 “university” medical school programs, but there are 384 residency programs, most of which are within community-based hospitals. The result is that most hospitalists encounter resident physicians in some capacity, and all hospitalists will encounter the results of residency training when they welcome a new recruit to their ranks.

The education and socialization of our residents will define the character of the hospitalists of the future. But the “residency” in which most of us trained does not exist anymore: The duty-hours changes and additional training requirements have dramatically changed the landscape of residency training in the past 10 years, and another series of sea changes is underway. As with all things HM, we again have a choice: Be reactive, wait for the dust to clear, and then lament the results, or be proactive and see this change for what it is—an opportunity to improve healthcare quality now, and in the future.

The ACGME

HM felt the impact of the first wave of duty-hours restrictions beginning in 2003, as many training programs opted to employ hospitalists to provide the coverage that could no longer be maintained by residents working under tighter admission caps and duty-hour restrictions. In doing so, hospitalists have provided a valuable service in preserving the integrity of training environments and fidelity to the Accreditation Council for Graduate Medical Education (ACGME) regulations (more than 85% of training programs have hospitalists working in their systems). But the model of hospitalists working solely as “resident-extenders” is not sustainable.

First, hospitalists who work solely on nonteaching services are at great risk of burning out, especially if the distribution of patients has been manipulated such that the more interesting patients are funneled away from the hospitalist’s service to the teaching service. Second, there is a risk in perception: In models in which the hospitalist is solely the “overflow cap coverage” or the night-float physician (i.e., the resident-extender), residents come to see hospitalists as the “PGY-4, 5, 6 …” physicians—that is, the physician who becomes a resident for life. The result is a serious pipeline issue for us, as the most talented resident physicians are unlikely to forego subspecialty training for a career in HM if hospitalists are perceived as perpetual residents.

The solution is simple: The hospitalist’s role in training environments has to be more than merely solving admission cap or duty-hour issues. It is fine for hospitalists to operate nonteaching services, but the hospitalist also has to be a part of the fulfillment that comes with overseeing teaching services. Further, residents have to see the hospitalist career for what it actually is: Academic or not, HM is much more than merely clinical service. HM is about the value-added services of system interventions to improve quality and patient safety; it is about developing a career as a systems architect. Getting the best and brightest residents to choose HM as a career is contingent upon residents seeing hospitalists in the training environment who are happy and fulfilled in the execution of this career goal.

The hospitalist’s plight was helped substantially on June 23, when ACGME released for comment the revised Common Program Requirements (www.acgme.org). The duty-hours changes are unlikely to substantially alter hospitalists’ lives; the only significant change was a limitation on intern shift durations to fewer than 16 hours in a row (upper-level residents still operate under the 24+6 hour rule, with increased flexibility to stay longer by volition). But the interesting part of the new requirements is an augmented focus on teaching residents transitions-of-care skills, improving direct supervision of residents, and constructing educational systems that minimize handoffs.

There is no specialty that is as suited as HM for fulfilling these unique (and, as of yet, unmet) requirements. Transitions, quality, being present on the hospital wards … this is what we do. And requiring instruction in transitions and quality is an unprecedented leverage point for HM to advance the quality of future physicians. How great it would be to attend HM20 and realize that the attendees had already learned the “Quality 101” lessons (i.e., those we are currently teaching at our annual meeting) as part of their residency? Freed from the need to do basic quality sessions, the content of the annual meeting could escalate to even higher-level principles that would result in substantial and sustainable quality improvement (QI).

MedPAC and GME Funding

Simultaneous with the ACGME changes are changes at the Medicare Payment Advisory Committee (MedPAC), the advisory organization responsible for recommending changes in the distribution of Centers for Medicare and Medicaid Services (CMS) funds to support graduate medical education. CMS is the primary funding agent for residency training. Each hospital receives direct medical expenditures to cover a resident’s salary and benefits. Each hospital has a pre-set per-resident allotment, or PRA. This number varies by hospital, but the average is $100,000 per resident. CMS reimburses the hospital a percentage of this number based upon the percentage of hospital days occupied by Medicare patients (e.g., 35% Medicare days=$35,000 per resident).

The hospital also receives indirect medical expenditures, or IME. IME is not a distinct payment to the hospital, but rather an “inflator” of the clinical-care payments the hospital receives from CMS. IME is paid to the hospital under the presumption that a typical training facility incurs greater cost due to higher patient severity, a higher indigent care percentage, and has higher resource utilization due to residents’ excessive testing, etc. The final presumption is that support is needed for the educational infrastructure (i.e., supervision and teaching).

IME is not inconsequential to a hospital; depending upon the payor mix, a 200-bed hospital might have from $4 million to $8 million in annual IME payments. CMS’ total IME payments to hospitals is more than $6 billion a year. Each hospital’s IME revenue can be found at www.graham-center.org/online/graham/home/tools-resources/data-tables/dt001-gme-2007.html.

The game-changing event occurred in April, when MedPAC announced its intent to reassess the mechanisms of IME funding, with a vision of IME funding eventually being linked to a hospital’s training programs’ ability to demonstrate substantial improvement in quality and patient safety. And here is the leverage point that is a unique opportunity for hospitalists in the training environment. For many hospitalists, especially if employed directly by the hospital, there is little financial incentive to engaging on a teaching service. The ACGME caps limit the service size, and this in turn limits the possible RVUs. Up until now, asking the hospital to compensate for teaching time (i.e., EVUs) was a pipe dream. But the linking of IME funding to quality outcomes (and quality instruction to residents) could change all of that.

If you put the two together: ACGME calling for instruction in quality and transitions, plus MedPAC calling for payments linked to resident outcomes in quality and patient safety, you have one inescapable conclusion—the residency of the future will hinge upon having supervisors with the necessary expertise to ensure that residents participate in, and understand the principles of, patient safety and quality as a part of the residency curriculum. And the people who can ensure that goal are likely to be in a position to warrant compensation for doing so.

Who is better to do this than the hospitalist?

SHM’s Proactive Strategy

This is the opportune time for HM to advance its stature as a profession and to ensure its future via a pipeline of residents adequately training in quality and patient safety. But it is not enough to merely wish for this to happen. There are real barriers that have kept hospitalists from being more intimately involved in physician training, the first of which is age.

HM is a young specialty (the average hospitalist is 37; the average HM leader is 41), and its youth makes it hard to compete with older subspecialists/generalists who have more experience in education. But deficits in experience can be compensated by additional training.

The Academic Hospitalist Academy (AHA)—cosponsored by SHM, the Society of General Internal Medicine (SGIM), and the Association of Chiefs and Leaders of General Internal Medicine (ACLGIM)—is the key to the strategy of catching up quickly. The academy will convene this month outside of Atlanta, and it is very important that each training facility think about sending one of its hospitalists to receive the advanced training in education necessary to compensate for not having years of experience in medical education. Academy details are available at http://academichospitalist.org.

SHM’s initiatives on this front do not stop with the academy. Over the past three months, Kevin O’Leary, MD, and his Quality Improvement Education Committee have been furiously building a “Quality and Patient Safety” curriculum, with a target audience of new hospitalists and resident physicians. The vision is to create a Web-based, interactive curriculum that teaches resident physicians the basics of quality and patient safety, design projects with their colleagues (under the supervision of their hospitalist mentor), and track their data to see real-time results.

Unlike other curricula on the market, the SHM Quality Curriculum for residents will be dynamic, requiring participating institutions commit to SHM’s modus operandi of mentored implementation by sponsoring a hospitalist to receive the training necessary to put the curriculum in motion. To this end, SHM has collaborated with the Alliance for Internal Medicine (AIM) in co-sponsoring the Quality Academy, with a focus on how to teach quality and patient safety. Jen Meyers, MD, FHM, and Jeff Glasheen, MD, SFHM, will be leading the team responsible for the development of this Quality Training Course, which should emerge in the fall of 2011.

As this project proceeds, Paul Grant, MD, chair of the Early Career Hospitalist Committee, and Cheryl O’Malley, MD, chair of the Pipeline Committee, will provide counsel. Both of these groups will continue efforts to improve the process by which residents transition from residency to HM practice, and supporting young physicians with distance mentoring.

The SHM vision of our production capacity is simple: Bring in the best and brightest hospitalists who are interested in teaching quality and patient safety, train them in the fundamentals of medical education, provide them with an “off the net” curriculum for how to teach quality, then return them to their respective training environments to coach residents on the principles of quality.

Training programs that invest in this vision will reap the rewards of fidelity to the new ACGME requirements. Hospitals that support such a vision will receive assurances, should MedPAC’s recommendation come to fruition, that DME and IME funding is secure. Hospitalists investing in this vision will find a fulfilling career in quality education.

And all of us will find assurances that, for as good as things are right now for HM, the future will be even better. TH

Dr. Wiese is president of SHM.

The mark of any great society is balance—balance between the production realized today and the preservation of “production capacity” to ensure the same or greater production in the future. HM is not exempt from this fundamental tenet. What we do now in the way of advancing quality, efficiency, and patient safety will matter little if our contributions are not sustained by the generation that follows us.

It is tempting to think that the issue of how we train residents is germane only to universities, but the reality is that it affects us all. There are 126 “university” medical school programs, but there are 384 residency programs, most of which are within community-based hospitals. The result is that most hospitalists encounter resident physicians in some capacity, and all hospitalists will encounter the results of residency training when they welcome a new recruit to their ranks.

The education and socialization of our residents will define the character of the hospitalists of the future. But the “residency” in which most of us trained does not exist anymore: The duty-hours changes and additional training requirements have dramatically changed the landscape of residency training in the past 10 years, and another series of sea changes is underway. As with all things HM, we again have a choice: Be reactive, wait for the dust to clear, and then lament the results, or be proactive and see this change for what it is—an opportunity to improve healthcare quality now, and in the future.

The ACGME

HM felt the impact of the first wave of duty-hours restrictions beginning in 2003, as many training programs opted to employ hospitalists to provide the coverage that could no longer be maintained by residents working under tighter admission caps and duty-hour restrictions. In doing so, hospitalists have provided a valuable service in preserving the integrity of training environments and fidelity to the Accreditation Council for Graduate Medical Education (ACGME) regulations (more than 85% of training programs have hospitalists working in their systems). But the model of hospitalists working solely as “resident-extenders” is not sustainable.

First, hospitalists who work solely on nonteaching services are at great risk of burning out, especially if the distribution of patients has been manipulated such that the more interesting patients are funneled away from the hospitalist’s service to the teaching service. Second, there is a risk in perception: In models in which the hospitalist is solely the “overflow cap coverage” or the night-float physician (i.e., the resident-extender), residents come to see hospitalists as the “PGY-4, 5, 6 …” physicians—that is, the physician who becomes a resident for life. The result is a serious pipeline issue for us, as the most talented resident physicians are unlikely to forego subspecialty training for a career in HM if hospitalists are perceived as perpetual residents.

The solution is simple: The hospitalist’s role in training environments has to be more than merely solving admission cap or duty-hour issues. It is fine for hospitalists to operate nonteaching services, but the hospitalist also has to be a part of the fulfillment that comes with overseeing teaching services. Further, residents have to see the hospitalist career for what it actually is: Academic or not, HM is much more than merely clinical service. HM is about the value-added services of system interventions to improve quality and patient safety; it is about developing a career as a systems architect. Getting the best and brightest residents to choose HM as a career is contingent upon residents seeing hospitalists in the training environment who are happy and fulfilled in the execution of this career goal.

The hospitalist’s plight was helped substantially on June 23, when ACGME released for comment the revised Common Program Requirements (www.acgme.org). The duty-hours changes are unlikely to substantially alter hospitalists’ lives; the only significant change was a limitation on intern shift durations to fewer than 16 hours in a row (upper-level residents still operate under the 24+6 hour rule, with increased flexibility to stay longer by volition). But the interesting part of the new requirements is an augmented focus on teaching residents transitions-of-care skills, improving direct supervision of residents, and constructing educational systems that minimize handoffs.

There is no specialty that is as suited as HM for fulfilling these unique (and, as of yet, unmet) requirements. Transitions, quality, being present on the hospital wards … this is what we do. And requiring instruction in transitions and quality is an unprecedented leverage point for HM to advance the quality of future physicians. How great it would be to attend HM20 and realize that the attendees had already learned the “Quality 101” lessons (i.e., those we are currently teaching at our annual meeting) as part of their residency? Freed from the need to do basic quality sessions, the content of the annual meeting could escalate to even higher-level principles that would result in substantial and sustainable quality improvement (QI).

MedPAC and GME Funding

Simultaneous with the ACGME changes are changes at the Medicare Payment Advisory Committee (MedPAC), the advisory organization responsible for recommending changes in the distribution of Centers for Medicare and Medicaid Services (CMS) funds to support graduate medical education. CMS is the primary funding agent for residency training. Each hospital receives direct medical expenditures to cover a resident’s salary and benefits. Each hospital has a pre-set per-resident allotment, or PRA. This number varies by hospital, but the average is $100,000 per resident. CMS reimburses the hospital a percentage of this number based upon the percentage of hospital days occupied by Medicare patients (e.g., 35% Medicare days=$35,000 per resident).

The hospital also receives indirect medical expenditures, or IME. IME is not a distinct payment to the hospital, but rather an “inflator” of the clinical-care payments the hospital receives from CMS. IME is paid to the hospital under the presumption that a typical training facility incurs greater cost due to higher patient severity, a higher indigent care percentage, and has higher resource utilization due to residents’ excessive testing, etc. The final presumption is that support is needed for the educational infrastructure (i.e., supervision and teaching).

IME is not inconsequential to a hospital; depending upon the payor mix, a 200-bed hospital might have from $4 million to $8 million in annual IME payments. CMS’ total IME payments to hospitals is more than $6 billion a year. Each hospital’s IME revenue can be found at www.graham-center.org/online/graham/home/tools-resources/data-tables/dt001-gme-2007.html.

The game-changing event occurred in April, when MedPAC announced its intent to reassess the mechanisms of IME funding, with a vision of IME funding eventually being linked to a hospital’s training programs’ ability to demonstrate substantial improvement in quality and patient safety. And here is the leverage point that is a unique opportunity for hospitalists in the training environment. For many hospitalists, especially if employed directly by the hospital, there is little financial incentive to engaging on a teaching service. The ACGME caps limit the service size, and this in turn limits the possible RVUs. Up until now, asking the hospital to compensate for teaching time (i.e., EVUs) was a pipe dream. But the linking of IME funding to quality outcomes (and quality instruction to residents) could change all of that.

If you put the two together: ACGME calling for instruction in quality and transitions, plus MedPAC calling for payments linked to resident outcomes in quality and patient safety, you have one inescapable conclusion—the residency of the future will hinge upon having supervisors with the necessary expertise to ensure that residents participate in, and understand the principles of, patient safety and quality as a part of the residency curriculum. And the people who can ensure that goal are likely to be in a position to warrant compensation for doing so.

Who is better to do this than the hospitalist?

SHM’s Proactive Strategy

This is the opportune time for HM to advance its stature as a profession and to ensure its future via a pipeline of residents adequately training in quality and patient safety. But it is not enough to merely wish for this to happen. There are real barriers that have kept hospitalists from being more intimately involved in physician training, the first of which is age.

HM is a young specialty (the average hospitalist is 37; the average HM leader is 41), and its youth makes it hard to compete with older subspecialists/generalists who have more experience in education. But deficits in experience can be compensated by additional training.

The Academic Hospitalist Academy (AHA)—cosponsored by SHM, the Society of General Internal Medicine (SGIM), and the Association of Chiefs and Leaders of General Internal Medicine (ACLGIM)—is the key to the strategy of catching up quickly. The academy will convene this month outside of Atlanta, and it is very important that each training facility think about sending one of its hospitalists to receive the advanced training in education necessary to compensate for not having years of experience in medical education. Academy details are available at http://academichospitalist.org.

SHM’s initiatives on this front do not stop with the academy. Over the past three months, Kevin O’Leary, MD, and his Quality Improvement Education Committee have been furiously building a “Quality and Patient Safety” curriculum, with a target audience of new hospitalists and resident physicians. The vision is to create a Web-based, interactive curriculum that teaches resident physicians the basics of quality and patient safety, design projects with their colleagues (under the supervision of their hospitalist mentor), and track their data to see real-time results.

Unlike other curricula on the market, the SHM Quality Curriculum for residents will be dynamic, requiring participating institutions commit to SHM’s modus operandi of mentored implementation by sponsoring a hospitalist to receive the training necessary to put the curriculum in motion. To this end, SHM has collaborated with the Alliance for Internal Medicine (AIM) in co-sponsoring the Quality Academy, with a focus on how to teach quality and patient safety. Jen Meyers, MD, FHM, and Jeff Glasheen, MD, SFHM, will be leading the team responsible for the development of this Quality Training Course, which should emerge in the fall of 2011.

As this project proceeds, Paul Grant, MD, chair of the Early Career Hospitalist Committee, and Cheryl O’Malley, MD, chair of the Pipeline Committee, will provide counsel. Both of these groups will continue efforts to improve the process by which residents transition from residency to HM practice, and supporting young physicians with distance mentoring.

The SHM vision of our production capacity is simple: Bring in the best and brightest hospitalists who are interested in teaching quality and patient safety, train them in the fundamentals of medical education, provide them with an “off the net” curriculum for how to teach quality, then return them to their respective training environments to coach residents on the principles of quality.

Training programs that invest in this vision will reap the rewards of fidelity to the new ACGME requirements. Hospitals that support such a vision will receive assurances, should MedPAC’s recommendation come to fruition, that DME and IME funding is secure. Hospitalists investing in this vision will find a fulfilling career in quality education.

And all of us will find assurances that, for as good as things are right now for HM, the future will be even better. TH

Dr. Wiese is president of SHM.

The mark of any great society is balance—balance between the production realized today and the preservation of “production capacity” to ensure the same or greater production in the future. HM is not exempt from this fundamental tenet. What we do now in the way of advancing quality, efficiency, and patient safety will matter little if our contributions are not sustained by the generation that follows us.

It is tempting to think that the issue of how we train residents is germane only to universities, but the reality is that it affects us all. There are 126 “university” medical school programs, but there are 384 residency programs, most of which are within community-based hospitals. The result is that most hospitalists encounter resident physicians in some capacity, and all hospitalists will encounter the results of residency training when they welcome a new recruit to their ranks.

The education and socialization of our residents will define the character of the hospitalists of the future. But the “residency” in which most of us trained does not exist anymore: The duty-hours changes and additional training requirements have dramatically changed the landscape of residency training in the past 10 years, and another series of sea changes is underway. As with all things HM, we again have a choice: Be reactive, wait for the dust to clear, and then lament the results, or be proactive and see this change for what it is—an opportunity to improve healthcare quality now, and in the future.

The ACGME

HM felt the impact of the first wave of duty-hours restrictions beginning in 2003, as many training programs opted to employ hospitalists to provide the coverage that could no longer be maintained by residents working under tighter admission caps and duty-hour restrictions. In doing so, hospitalists have provided a valuable service in preserving the integrity of training environments and fidelity to the Accreditation Council for Graduate Medical Education (ACGME) regulations (more than 85% of training programs have hospitalists working in their systems). But the model of hospitalists working solely as “resident-extenders” is not sustainable.

First, hospitalists who work solely on nonteaching services are at great risk of burning out, especially if the distribution of patients has been manipulated such that the more interesting patients are funneled away from the hospitalist’s service to the teaching service. Second, there is a risk in perception: In models in which the hospitalist is solely the “overflow cap coverage” or the night-float physician (i.e., the resident-extender), residents come to see hospitalists as the “PGY-4, 5, 6 …” physicians—that is, the physician who becomes a resident for life. The result is a serious pipeline issue for us, as the most talented resident physicians are unlikely to forego subspecialty training for a career in HM if hospitalists are perceived as perpetual residents.

The solution is simple: The hospitalist’s role in training environments has to be more than merely solving admission cap or duty-hour issues. It is fine for hospitalists to operate nonteaching services, but the hospitalist also has to be a part of the fulfillment that comes with overseeing teaching services. Further, residents have to see the hospitalist career for what it actually is: Academic or not, HM is much more than merely clinical service. HM is about the value-added services of system interventions to improve quality and patient safety; it is about developing a career as a systems architect. Getting the best and brightest residents to choose HM as a career is contingent upon residents seeing hospitalists in the training environment who are happy and fulfilled in the execution of this career goal.

The hospitalist’s plight was helped substantially on June 23, when ACGME released for comment the revised Common Program Requirements (www.acgme.org). The duty-hours changes are unlikely to substantially alter hospitalists’ lives; the only significant change was a limitation on intern shift durations to fewer than 16 hours in a row (upper-level residents still operate under the 24+6 hour rule, with increased flexibility to stay longer by volition). But the interesting part of the new requirements is an augmented focus on teaching residents transitions-of-care skills, improving direct supervision of residents, and constructing educational systems that minimize handoffs.

There is no specialty that is as suited as HM for fulfilling these unique (and, as of yet, unmet) requirements. Transitions, quality, being present on the hospital wards … this is what we do. And requiring instruction in transitions and quality is an unprecedented leverage point for HM to advance the quality of future physicians. How great it would be to attend HM20 and realize that the attendees had already learned the “Quality 101” lessons (i.e., those we are currently teaching at our annual meeting) as part of their residency? Freed from the need to do basic quality sessions, the content of the annual meeting could escalate to even higher-level principles that would result in substantial and sustainable quality improvement (QI).

MedPAC and GME Funding

Simultaneous with the ACGME changes are changes at the Medicare Payment Advisory Committee (MedPAC), the advisory organization responsible for recommending changes in the distribution of Centers for Medicare and Medicaid Services (CMS) funds to support graduate medical education. CMS is the primary funding agent for residency training. Each hospital receives direct medical expenditures to cover a resident’s salary and benefits. Each hospital has a pre-set per-resident allotment, or PRA. This number varies by hospital, but the average is $100,000 per resident. CMS reimburses the hospital a percentage of this number based upon the percentage of hospital days occupied by Medicare patients (e.g., 35% Medicare days=$35,000 per resident).

The hospital also receives indirect medical expenditures, or IME. IME is not a distinct payment to the hospital, but rather an “inflator” of the clinical-care payments the hospital receives from CMS. IME is paid to the hospital under the presumption that a typical training facility incurs greater cost due to higher patient severity, a higher indigent care percentage, and has higher resource utilization due to residents’ excessive testing, etc. The final presumption is that support is needed for the educational infrastructure (i.e., supervision and teaching).

IME is not inconsequential to a hospital; depending upon the payor mix, a 200-bed hospital might have from $4 million to $8 million in annual IME payments. CMS’ total IME payments to hospitals is more than $6 billion a year. Each hospital’s IME revenue can be found at www.graham-center.org/online/graham/home/tools-resources/data-tables/dt001-gme-2007.html.

The game-changing event occurred in April, when MedPAC announced its intent to reassess the mechanisms of IME funding, with a vision of IME funding eventually being linked to a hospital’s training programs’ ability to demonstrate substantial improvement in quality and patient safety. And here is the leverage point that is a unique opportunity for hospitalists in the training environment. For many hospitalists, especially if employed directly by the hospital, there is little financial incentive to engaging on a teaching service. The ACGME caps limit the service size, and this in turn limits the possible RVUs. Up until now, asking the hospital to compensate for teaching time (i.e., EVUs) was a pipe dream. But the linking of IME funding to quality outcomes (and quality instruction to residents) could change all of that.

If you put the two together: ACGME calling for instruction in quality and transitions, plus MedPAC calling for payments linked to resident outcomes in quality and patient safety, you have one inescapable conclusion—the residency of the future will hinge upon having supervisors with the necessary expertise to ensure that residents participate in, and understand the principles of, patient safety and quality as a part of the residency curriculum. And the people who can ensure that goal are likely to be in a position to warrant compensation for doing so.

Who is better to do this than the hospitalist?

SHM’s Proactive Strategy

This is the opportune time for HM to advance its stature as a profession and to ensure its future via a pipeline of residents adequately training in quality and patient safety. But it is not enough to merely wish for this to happen. There are real barriers that have kept hospitalists from being more intimately involved in physician training, the first of which is age.

HM is a young specialty (the average hospitalist is 37; the average HM leader is 41), and its youth makes it hard to compete with older subspecialists/generalists who have more experience in education. But deficits in experience can be compensated by additional training.

The Academic Hospitalist Academy (AHA)—cosponsored by SHM, the Society of General Internal Medicine (SGIM), and the Association of Chiefs and Leaders of General Internal Medicine (ACLGIM)—is the key to the strategy of catching up quickly. The academy will convene this month outside of Atlanta, and it is very important that each training facility think about sending one of its hospitalists to receive the advanced training in education necessary to compensate for not having years of experience in medical education. Academy details are available at http://academichospitalist.org.

SHM’s initiatives on this front do not stop with the academy. Over the past three months, Kevin O’Leary, MD, and his Quality Improvement Education Committee have been furiously building a “Quality and Patient Safety” curriculum, with a target audience of new hospitalists and resident physicians. The vision is to create a Web-based, interactive curriculum that teaches resident physicians the basics of quality and patient safety, design projects with their colleagues (under the supervision of their hospitalist mentor), and track their data to see real-time results.

Unlike other curricula on the market, the SHM Quality Curriculum for residents will be dynamic, requiring participating institutions commit to SHM’s modus operandi of mentored implementation by sponsoring a hospitalist to receive the training necessary to put the curriculum in motion. To this end, SHM has collaborated with the Alliance for Internal Medicine (AIM) in co-sponsoring the Quality Academy, with a focus on how to teach quality and patient safety. Jen Meyers, MD, FHM, and Jeff Glasheen, MD, SFHM, will be leading the team responsible for the development of this Quality Training Course, which should emerge in the fall of 2011.

As this project proceeds, Paul Grant, MD, chair of the Early Career Hospitalist Committee, and Cheryl O’Malley, MD, chair of the Pipeline Committee, will provide counsel. Both of these groups will continue efforts to improve the process by which residents transition from residency to HM practice, and supporting young physicians with distance mentoring.

The SHM vision of our production capacity is simple: Bring in the best and brightest hospitalists who are interested in teaching quality and patient safety, train them in the fundamentals of medical education, provide them with an “off the net” curriculum for how to teach quality, then return them to their respective training environments to coach residents on the principles of quality.

Training programs that invest in this vision will reap the rewards of fidelity to the new ACGME requirements. Hospitals that support such a vision will receive assurances, should MedPAC’s recommendation come to fruition, that DME and IME funding is secure. Hospitalists investing in this vision will find a fulfilling career in quality education.

And all of us will find assurances that, for as good as things are right now for HM, the future will be even better. TH

Dr. Wiese is president of SHM.

Act I: The Negotiation

(A barren academic office, dimly lit, the pall of difficult negotiations afloat, backlit like dust in the air. Seated, under a strangely intense incandescent bulb, a man, who looks eerily like a good-looking version of me, sits uncomfortably adjusting himself in his seat. His eyes constrict on his counterpart, a miserly sort peering out from behind wire-rim glasses and a shock of hair improbably combed over from ear to ear. The tension crests.)

GOOD-LOOKING ME

(Voice cracking)

I’ve come to ask for a raise for our hospitalist group.

MISER

(Adjusts his clip-on tie)

We just gave you a raise in 2004.

GOOD-LOOKING ME

(Smiles uncomfortably)

That was very gracious, sir, but I think the numbers support another.

MISER

(Incredulous look at his watch)

But your work RVUs are thousands below what I’d like to see.

GOOD-LOOKING ME

(Dabs bead of sweat away from chiseled chin)

That’s because you’ve set your benchmark thousands above a reasonable number.

MISER

(Voice flitting with child-like condescension)

But those are the numbers my finance guy gave me. It’s the benchmark.

(Blackout and end of Act I.)

Mutual Agreement

Tony Award-winning stuff for sure—and based on a true story! In fact, this scene no doubt plays out annually for those of you unfortunate enough to have to negotiate with hospital executives for programmatic support. To be fair, hospital administrators deserve to know that they are getting what they pay for. Thus, the concepts of a benchmark are reasonable. The problem lies in setting mutually-agreed-upon standards.

Act II: Disbelief and Confusion

GOOD-LOOKING ME

(Unsteadily hands document to Miser)

Sir, I’ve highlighted the national benchmarks for you to see. Column four of this 2007-2008 SHM survey clearly shows that the average academic hospitalist should make $168,800 and achieve 2,813 work RVUs. We achieve the latter benchmark but are severely underpaid.

MISER

(Produces a folded cocktail napkin from his shirt pocket)

But look at this: My executive-friends-at-other-medical-centers-who-overwork-and-underpay-their-hospitalists benchmark shows that you should be well over 4,500 work RVUs. And besides, the SHM numbers are skewed; it’s a survey of hospitalists done by a group that represents hospitalists. I don’t believe them.

GOOD-LOOKING ME

(Eyes averted, adopts a tone of trepidation)

But sir, with all due respect, don’t your numbers reflect a survey of hospital administrators who might have a bias toward more expected productivity? Which benchmark should we believe?

(Blackout and end of Act II.)

A New HM Benchmark Arises

It’s all about the benchmark you choose to believe. For years, the best source of data regarding hospitalist compensation and productivity was that published every other year by SHM. It is a fair, but unfounded, concern that these data might tilt toward the benefit of hospitalists. Likewise, the hospital administrator I work most closely with (who, for the record, reads this publication and IS NOT miserly, has a FULL HEAD of hair, and is, for innumerable reasons, a TRULY GREAT man) will produce benchmarks from organizations like the Association of American Medical Colleges (AAMC) or the University HealthSystems Consortium (UHC), all of which show surprisingly disparate numbers dripping with a similar tilt toward the medical center.

Thus, the importance of the 2010 SHM/MGMA report. The Medical Group Management Association (MGMA) consists of administrators and leaders of medical group practices. Since 1926, they’ve been providing accurate, independent data on physician practice metrics. For most hospital administrators, it is the benchmark. The problem is that in the past, MGMA has struggled to identify hospitalists; the MGMA data were always underpowered and, therefore, suspect.

Enter SHM and its large database of HM groups. What has resulted in the new survey rivals those old commercials in which a person walking with a piece of chocolate slams into another with a jar of peanut butter, resulting in the creation of the Reese’s Peanut Butter Cup (apologies to those readers under the age of 35).

Act III: No Raise; Children Go Hungry

GOOD-LOOKING ME

(Unsheathing haloed document from his portfolio)

Perhaps we could agree to use these new SHM/MGMA numbers as our benchmark. It includes data from more than 440 HM groups and 4,200 hospitalists. And it appears to be fair and balanced.

MISER

(Eyes alight, peering through a shroud of compromise)

MGMA, huh? Let’s take a look. Hmmm. Well. But wait—this says the average hospitalist makes $215,000! That’s outrageous.

GOOD-LOOKING ME

(Smugly retorts)

Yes, sir, we are severely underpaid.

MISER

(Reading; a weasel-like countenance overtakes his face)

Let me take a closer look at this. Aha! Here it is. You see, this only included community hospitalist practices. You will be getting no raise!

(Blackout and end of Act III.)

A Cautionary Tale

Alas, the miser is right. It’s not always what the data say but also what they don’t say.

The one snag with the new data is that it only included a handful of academic HM groups (only 1% of respondents). In fact, the survey actively instructed academic HM practices to not complete the survey. Rather, we academic types were instructed to await the MGMA survey of academic practices completed every fall to be reported early next year.

This is emblematic of the need to dig deep when interpreting these data. As tempting as it is to use a sound bite or two of these data to your advantage, the truth lies in the details. It’s easy to say that all hospitalists should make $215,000, see 2,229 encounters, and achieve 4,107 wRVUs annually.

However, just as there is no average hospitalist, there are no average numbers. There are just too many variables (e.g., practice ownership, geography, group size, night coverage, staffing model, compensation structure) to say definitively what an individual hospitalist should look like or achieve. Rather, these numbers should be used as a guide, adapted to each individual situation.

Act IV: See You This Spring

(Standing, Good-Looking Me shakes his foe’s shriveled claw of a hand while looking him intensely in the eye—a look that says, “I’ll see you this spring.” In his rival’s eyes, the Miser sees his future—a future that involves another meeting, more practice-appropriate data, and a dusting off of his checkbook.)

(Blackout and end of Act IV.) TH

Dr. Glasheen is associate professor of medicine at the University of Colorado 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.

Act I: The Negotiation

(A barren academic office, dimly lit, the pall of difficult negotiations afloat, backlit like dust in the air. Seated, under a strangely intense incandescent bulb, a man, who looks eerily like a good-looking version of me, sits uncomfortably adjusting himself in his seat. His eyes constrict on his counterpart, a miserly sort peering out from behind wire-rim glasses and a shock of hair improbably combed over from ear to ear. The tension crests.)

GOOD-LOOKING ME

(Voice cracking)

I’ve come to ask for a raise for our hospitalist group.

MISER

(Adjusts his clip-on tie)

We just gave you a raise in 2004.

GOOD-LOOKING ME

(Smiles uncomfortably)

That was very gracious, sir, but I think the numbers support another.

MISER

(Incredulous look at his watch)

But your work RVUs are thousands below what I’d like to see.

GOOD-LOOKING ME

(Dabs bead of sweat away from chiseled chin)

That’s because you’ve set your benchmark thousands above a reasonable number.

MISER

(Voice flitting with child-like condescension)

But those are the numbers my finance guy gave me. It’s the benchmark.

(Blackout and end of Act I.)

Mutual Agreement

Tony Award-winning stuff for sure—and based on a true story! In fact, this scene no doubt plays out annually for those of you unfortunate enough to have to negotiate with hospital executives for programmatic support. To be fair, hospital administrators deserve to know that they are getting what they pay for. Thus, the concepts of a benchmark are reasonable. The problem lies in setting mutually-agreed-upon standards.

Act II: Disbelief and Confusion

GOOD-LOOKING ME

(Unsteadily hands document to Miser)

Sir, I’ve highlighted the national benchmarks for you to see. Column four of this 2007-2008 SHM survey clearly shows that the average academic hospitalist should make $168,800 and achieve 2,813 work RVUs. We achieve the latter benchmark but are severely underpaid.

MISER

(Produces a folded cocktail napkin from his shirt pocket)

But look at this: My executive-friends-at-other-medical-centers-who-overwork-and-underpay-their-hospitalists benchmark shows that you should be well over 4,500 work RVUs. And besides, the SHM numbers are skewed; it’s a survey of hospitalists done by a group that represents hospitalists. I don’t believe them.

GOOD-LOOKING ME

(Eyes averted, adopts a tone of trepidation)

But sir, with all due respect, don’t your numbers reflect a survey of hospital administrators who might have a bias toward more expected productivity? Which benchmark should we believe?

(Blackout and end of Act II.)

A New HM Benchmark Arises

It’s all about the benchmark you choose to believe. For years, the best source of data regarding hospitalist compensation and productivity was that published every other year by SHM. It is a fair, but unfounded, concern that these data might tilt toward the benefit of hospitalists. Likewise, the hospital administrator I work most closely with (who, for the record, reads this publication and IS NOT miserly, has a FULL HEAD of hair, and is, for innumerable reasons, a TRULY GREAT man) will produce benchmarks from organizations like the Association of American Medical Colleges (AAMC) or the University HealthSystems Consortium (UHC), all of which show surprisingly disparate numbers dripping with a similar tilt toward the medical center.

Thus, the importance of the 2010 SHM/MGMA report. The Medical Group Management Association (MGMA) consists of administrators and leaders of medical group practices. Since 1926, they’ve been providing accurate, independent data on physician practice metrics. For most hospital administrators, it is the benchmark. The problem is that in the past, MGMA has struggled to identify hospitalists; the MGMA data were always underpowered and, therefore, suspect.

Enter SHM and its large database of HM groups. What has resulted in the new survey rivals those old commercials in which a person walking with a piece of chocolate slams into another with a jar of peanut butter, resulting in the creation of the Reese’s Peanut Butter Cup (apologies to those readers under the age of 35).

Act III: No Raise; Children Go Hungry

GOOD-LOOKING ME

(Unsheathing haloed document from his portfolio)

Perhaps we could agree to use these new SHM/MGMA numbers as our benchmark. It includes data from more than 440 HM groups and 4,200 hospitalists. And it appears to be fair and balanced.

MISER

(Eyes alight, peering through a shroud of compromise)

MGMA, huh? Let’s take a look. Hmmm. Well. But wait—this says the average hospitalist makes $215,000! That’s outrageous.

GOOD-LOOKING ME

(Smugly retorts)

Yes, sir, we are severely underpaid.

MISER

(Reading; a weasel-like countenance overtakes his face)

Let me take a closer look at this. Aha! Here it is. You see, this only included community hospitalist practices. You will be getting no raise!

(Blackout and end of Act III.)

A Cautionary Tale

Alas, the miser is right. It’s not always what the data say but also what they don’t say.

The one snag with the new data is that it only included a handful of academic HM groups (only 1% of respondents). In fact, the survey actively instructed academic HM practices to not complete the survey. Rather, we academic types were instructed to await the MGMA survey of academic practices completed every fall to be reported early next year.

This is emblematic of the need to dig deep when interpreting these data. As tempting as it is to use a sound bite or two of these data to your advantage, the truth lies in the details. It’s easy to say that all hospitalists should make $215,000, see 2,229 encounters, and achieve 4,107 wRVUs annually.

However, just as there is no average hospitalist, there are no average numbers. There are just too many variables (e.g., practice ownership, geography, group size, night coverage, staffing model, compensation structure) to say definitively what an individual hospitalist should look like or achieve. Rather, these numbers should be used as a guide, adapted to each individual situation.

Act IV: See You This Spring

(Standing, Good-Looking Me shakes his foe’s shriveled claw of a hand while looking him intensely in the eye—a look that says, “I’ll see you this spring.” In his rival’s eyes, the Miser sees his future—a future that involves another meeting, more practice-appropriate data, and a dusting off of his checkbook.)

(Blackout and end of Act IV.) TH

Dr. Glasheen is associate professor of medicine at the University of Colorado 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.

Act I: The Negotiation

(A barren academic office, dimly lit, the pall of difficult negotiations afloat, backlit like dust in the air. Seated, under a strangely intense incandescent bulb, a man, who looks eerily like a good-looking version of me, sits uncomfortably adjusting himself in his seat. His eyes constrict on his counterpart, a miserly sort peering out from behind wire-rim glasses and a shock of hair improbably combed over from ear to ear. The tension crests.)

GOOD-LOOKING ME

(Voice cracking)

I’ve come to ask for a raise for our hospitalist group.

MISER

(Adjusts his clip-on tie)

We just gave you a raise in 2004.

GOOD-LOOKING ME

(Smiles uncomfortably)

That was very gracious, sir, but I think the numbers support another.

MISER

(Incredulous look at his watch)

But your work RVUs are thousands below what I’d like to see.

GOOD-LOOKING ME

(Dabs bead of sweat away from chiseled chin)

That’s because you’ve set your benchmark thousands above a reasonable number.

MISER

(Voice flitting with child-like condescension)

But those are the numbers my finance guy gave me. It’s the benchmark.

(Blackout and end of Act I.)

Mutual Agreement

Tony Award-winning stuff for sure—and based on a true story! In fact, this scene no doubt plays out annually for those of you unfortunate enough to have to negotiate with hospital executives for programmatic support. To be fair, hospital administrators deserve to know that they are getting what they pay for. Thus, the concepts of a benchmark are reasonable. The problem lies in setting mutually-agreed-upon standards.

Act II: Disbelief and Confusion

GOOD-LOOKING ME

(Unsteadily hands document to Miser)

Sir, I’ve highlighted the national benchmarks for you to see. Column four of this 2007-2008 SHM survey clearly shows that the average academic hospitalist should make $168,800 and achieve 2,813 work RVUs. We achieve the latter benchmark but are severely underpaid.

MISER

(Produces a folded cocktail napkin from his shirt pocket)

But look at this: My executive-friends-at-other-medical-centers-who-overwork-and-underpay-their-hospitalists benchmark shows that you should be well over 4,500 work RVUs. And besides, the SHM numbers are skewed; it’s a survey of hospitalists done by a group that represents hospitalists. I don’t believe them.

GOOD-LOOKING ME

(Eyes averted, adopts a tone of trepidation)

But sir, with all due respect, don’t your numbers reflect a survey of hospital administrators who might have a bias toward more expected productivity? Which benchmark should we believe?

(Blackout and end of Act II.)

A New HM Benchmark Arises

It’s all about the benchmark you choose to believe. For years, the best source of data regarding hospitalist compensation and productivity was that published every other year by SHM. It is a fair, but unfounded, concern that these data might tilt toward the benefit of hospitalists. Likewise, the hospital administrator I work most closely with (who, for the record, reads this publication and IS NOT miserly, has a FULL HEAD of hair, and is, for innumerable reasons, a TRULY GREAT man) will produce benchmarks from organizations like the Association of American Medical Colleges (AAMC) or the University HealthSystems Consortium (UHC), all of which show surprisingly disparate numbers dripping with a similar tilt toward the medical center.

Thus, the importance of the 2010 SHM/MGMA report. The Medical Group Management Association (MGMA) consists of administrators and leaders of medical group practices. Since 1926, they’ve been providing accurate, independent data on physician practice metrics. For most hospital administrators, it is the benchmark. The problem is that in the past, MGMA has struggled to identify hospitalists; the MGMA data were always underpowered and, therefore, suspect.

Enter SHM and its large database of HM groups. What has resulted in the new survey rivals those old commercials in which a person walking with a piece of chocolate slams into another with a jar of peanut butter, resulting in the creation of the Reese’s Peanut Butter Cup (apologies to those readers under the age of 35).

Act III: No Raise; Children Go Hungry

GOOD-LOOKING ME

(Unsheathing haloed document from his portfolio)

Perhaps we could agree to use these new SHM/MGMA numbers as our benchmark. It includes data from more than 440 HM groups and 4,200 hospitalists. And it appears to be fair and balanced.

MISER

(Eyes alight, peering through a shroud of compromise)

MGMA, huh? Let’s take a look. Hmmm. Well. But wait—this says the average hospitalist makes $215,000! That’s outrageous.

GOOD-LOOKING ME

(Smugly retorts)

Yes, sir, we are severely underpaid.

MISER

(Reading; a weasel-like countenance overtakes his face)

Let me take a closer look at this. Aha! Here it is. You see, this only included community hospitalist practices. You will be getting no raise!

(Blackout and end of Act III.)

A Cautionary Tale

Alas, the miser is right. It’s not always what the data say but also what they don’t say.

The one snag with the new data is that it only included a handful of academic HM groups (only 1% of respondents). In fact, the survey actively instructed academic HM practices to not complete the survey. Rather, we academic types were instructed to await the MGMA survey of academic practices completed every fall to be reported early next year.

This is emblematic of the need to dig deep when interpreting these data. As tempting as it is to use a sound bite or two of these data to your advantage, the truth lies in the details. It’s easy to say that all hospitalists should make $215,000, see 2,229 encounters, and achieve 4,107 wRVUs annually.

However, just as there is no average hospitalist, there are no average numbers. There are just too many variables (e.g., practice ownership, geography, group size, night coverage, staffing model, compensation structure) to say definitively what an individual hospitalist should look like or achieve. Rather, these numbers should be used as a guide, adapted to each individual situation.

Act IV: See You This Spring

(Standing, Good-Looking Me shakes his foe’s shriveled claw of a hand while looking him intensely in the eye—a look that says, “I’ll see you this spring.” In his rival’s eyes, the Miser sees his future—a future that involves another meeting, more practice-appropriate data, and a dusting off of his checkbook.)

(Blackout and end of Act IV.) TH

Dr. Glasheen is associate professor of medicine at the University of Colorado 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.

Unpredictable workloads and frequent interruptions are the things I regard as the most stressful components of work as a hospitalist. Your list might be very different, but I bet unpredictable workloads ranks at least in the top five of every hospitalist’s list.

I’ve discussed interruptions previously (see “Really, It’s Switch-Tasking,” p. 68, November 2008; “Technological Advance or Workplace Setback?” p. 69, December 2008), but this month and next will turn to unpredictable workloads. In other words, what are the strategies available to a hospitalist practice to provide surge capacity in response to such unpredictable increases in patient volume as an uptick in census or daily admissions 50% to 100% above normal? I’ll leave to others the topic of how hospitals respond to such disasters as terrorist attacks, earthquakes, etc.

The Bottom Line

Sadly, there is no magic bullet for the “surge” problem, and no way to protect on-duty hospitalists from the need to work harder when it gets busy. But we needn’t feel too sorry for ourselves; doctors in most other specialties who practice in the hospital face the same problem and tend to rely heavily on simply working harder and longer when it is unusually busy. Sometimes they couple the “work harder” mantra with other strategies, such as calling another doctor in to help.

Hospitalists have a duty to ensure high patient volume doesn’t lead to deterioration in the quality of patient care, but occasionally working longer days than average probably poses a low risk, and might be less risky than the additional handoffs usually associated with having a doctor on “jeopardy” to be called in when it’s busy. Routinely or frequently working unreasonably long days is another story.

The trick for HM programs is to build some surge capacity into the routine daily staffing 1) without exceeding a reasonable budget, while 2) ensuring that the hospitalists don’t simply become accustomed to light workloads as the only reasonable norm, which could lead to them becoming unwilling to accept higher, but still reasonable, workloads when needed. (More on these issues later.) First, I’ll go through what I see as the pros and cons of several approaches to addressing surges in patient volume. All are in use with variable frequency around the country.

“Jeopardy” System

In its most common form, a jeopardy system has an unscheduled doctor each day who must remain available on short notice by pager. When patient volume surges, the unscheduled doctor is paged to come in and help. In most cases, this doctor focuses primarily—or exclusively—on admitting patients for a few hours. So it is most common for this doctor to be called in late in the afternoon or early in the evening. The jeopardy doctor usually turns over all admitted patients to another hospitalist in the group for all subsequent care. In addition to providing surge capacity, the jeopardy doctor almost always is used to cover unexpected absences of scheduled doctors, including illness-related absences.

Sometimes this doctor is paid extra for each day or week spent being “available” on jeopardy duty (not to be confused with jury duty, though it can be equally difficult to get exempted from). Then again, it is not uncommon to have jeopardy duty included in base compensation. However, once a jeopardy doctor is actually called in to work, most practices pay additional compensation, often based on an hourly rate that usually is higher than the average compensation generated per hour for nonjeopardy work.

There are a number of reasonable ways to compensate the jeopardy doctor. You probably can get some good ideas by talking with others in your hospital who function in a similar capacity, such as cath-lab technicians who get called in on nights and weekends.

No definitive data are available to show how common the jeopardy system is, but my experience is that 30% to 50% of HM groups use some form of it. Its popularity is proof that it is a reasonable system, but I’m not convinced. I think it is in use by a lot of groups not because it is an optimal way to ensure surge capacity, but because it is easy to conceptualize and put in place, and because many hospitalists came from residency programs in which the system was standard.

The gaps between theoretical and realized benefits become evident once a practice implements a jeopardy system. For example, it might be really busy today, but Dr. Stravinsky doesn’t call in Dr. Copeland, who is on jeopardy, because next week their roles will be reversed and Dr. Stravinsky sure hopes he won’t be called in. No one wants to be the weak doctor who calls in the jeopardy doctor and spoils what was otherwise a day off.

I’ve worked with a lot of practices who say they have a jeopardy system in place, but when I ask for the last time the jeopardy doctor was called in, they say it has been more than a year, or in some cases never. So even if the policy manual says they have a jeopardy system, the doctors never activate it, so it provides no benefit.

Practices that do utilize the jeopardy doctor have their own problems, such as assigning that doctor’s admissions the next day. The jeopardy doctor might provide some relief today, but they essentially just delay the work of having to get to know all of those new patients until the morning, when everyone is very busy with rounds. So while there might be significant benefit in activating the jeopardy system today, it could just delay the problem of high workload until the next morning, which isn’t much of a net benefit for the practice.

A small number of practices call in the jeopardy doctor frequently, and sometimes have that doctor continue to round on admitted patients for the next few days. This usage might get the most value out of the system, but the practice should consider if it is more cost-effective, and less stressful for the doctors, if the system were reversed. For example, instead of having the doctor on jeopardy and called in as necessary, the doctor would report to work and be given the day off or let go early when it isn’t busy.

Despite my reservations, if you are convinced the jeopardy system is valuable and cost-effective, keep it in place. However, if your group is thinking about options to handle surge capacity, don’t be too quick to adopt a jeopardy system. It usually falls far short of a perfect solution.

Patient Volume Cap

Another way to address the problem of unpredictable increases in patient volume is to establish a patient volume (e.g., total census) cap for the whole hospitalist practice. Like the jeopardy system, this is an appealingly uncomplicated idea, and hospitalists who have finished residency within the last few years all worked with a cap.

Except for the rarest of exceptions, this is a poor idea and should be avoided if at all possible. I’ll leave for another time a discussion of all the political and financial costs of a cap system, but trust me on this one. It is best to avoid a cap.

Stay Tuned …

Next month, I’ll examine other strategies to provide surge capacity. I think they’re more valuable than the two I’ve mentioned here, but I need to warn you that they aren’t perfect and are more complicated to operationalize. TH

Dr. Nelson has been a practicing hospitalist since 1988 and is co-founder and past president of SHM. He is a principal in Nelson Flores Hospital Medicine Consultants, a national hospitalist practice management consulting firm (www.nelson flores.com). He is course co-director and faculty for SHM’s “Best Practices in Managing a Hospital Medicine Program” course. This column represents his views and is not intended to reflect an official position of SHM.

Unpredictable workloads and frequent interruptions are the things I regard as the most stressful components of work as a hospitalist. Your list might be very different, but I bet unpredictable workloads ranks at least in the top five of every hospitalist’s list.

I’ve discussed interruptions previously (see “Really, It’s Switch-Tasking,” p. 68, November 2008; “Technological Advance or Workplace Setback?” p. 69, December 2008), but this month and next will turn to unpredictable workloads. In other words, what are the strategies available to a hospitalist practice to provide surge capacity in response to such unpredictable increases in patient volume as an uptick in census or daily admissions 50% to 100% above normal? I’ll leave to others the topic of how hospitals respond to such disasters as terrorist attacks, earthquakes, etc.

The Bottom Line

Sadly, there is no magic bullet for the “surge” problem, and no way to protect on-duty hospitalists from the need to work harder when it gets busy. But we needn’t feel too sorry for ourselves; doctors in most other specialties who practice in the hospital face the same problem and tend to rely heavily on simply working harder and longer when it is unusually busy. Sometimes they couple the “work harder” mantra with other strategies, such as calling another doctor in to help.

Hospitalists have a duty to ensure high patient volume doesn’t lead to deterioration in the quality of patient care, but occasionally working longer days than average probably poses a low risk, and might be less risky than the additional handoffs usually associated with having a doctor on “jeopardy” to be called in when it’s busy. Routinely or frequently working unreasonably long days is another story.

The trick for HM programs is to build some surge capacity into the routine daily staffing 1) without exceeding a reasonable budget, while 2) ensuring that the hospitalists don’t simply become accustomed to light workloads as the only reasonable norm, which could lead to them becoming unwilling to accept higher, but still reasonable, workloads when needed. (More on these issues later.) First, I’ll go through what I see as the pros and cons of several approaches to addressing surges in patient volume. All are in use with variable frequency around the country.

“Jeopardy” System

In its most common form, a jeopardy system has an unscheduled doctor each day who must remain available on short notice by pager. When patient volume surges, the unscheduled doctor is paged to come in and help. In most cases, this doctor focuses primarily—or exclusively—on admitting patients for a few hours. So it is most common for this doctor to be called in late in the afternoon or early in the evening. The jeopardy doctor usually turns over all admitted patients to another hospitalist in the group for all subsequent care. In addition to providing surge capacity, the jeopardy doctor almost always is used to cover unexpected absences of scheduled doctors, including illness-related absences.

Sometimes this doctor is paid extra for each day or week spent being “available” on jeopardy duty (not to be confused with jury duty, though it can be equally difficult to get exempted from). Then again, it is not uncommon to have jeopardy duty included in base compensation. However, once a jeopardy doctor is actually called in to work, most practices pay additional compensation, often based on an hourly rate that usually is higher than the average compensation generated per hour for nonjeopardy work.

There are a number of reasonable ways to compensate the jeopardy doctor. You probably can get some good ideas by talking with others in your hospital who function in a similar capacity, such as cath-lab technicians who get called in on nights and weekends.

No definitive data are available to show how common the jeopardy system is, but my experience is that 30% to 50% of HM groups use some form of it. Its popularity is proof that it is a reasonable system, but I’m not convinced. I think it is in use by a lot of groups not because it is an optimal way to ensure surge capacity, but because it is easy to conceptualize and put in place, and because many hospitalists came from residency programs in which the system was standard.

The gaps between theoretical and realized benefits become evident once a practice implements a jeopardy system. For example, it might be really busy today, but Dr. Stravinsky doesn’t call in Dr. Copeland, who is on jeopardy, because next week their roles will be reversed and Dr. Stravinsky sure hopes he won’t be called in. No one wants to be the weak doctor who calls in the jeopardy doctor and spoils what was otherwise a day off.

I’ve worked with a lot of practices who say they have a jeopardy system in place, but when I ask for the last time the jeopardy doctor was called in, they say it has been more than a year, or in some cases never. So even if the policy manual says they have a jeopardy system, the doctors never activate it, so it provides no benefit.

Practices that do utilize the jeopardy doctor have their own problems, such as assigning that doctor’s admissions the next day. The jeopardy doctor might provide some relief today, but they essentially just delay the work of having to get to know all of those new patients until the morning, when everyone is very busy with rounds. So while there might be significant benefit in activating the jeopardy system today, it could just delay the problem of high workload until the next morning, which isn’t much of a net benefit for the practice.

A small number of practices call in the jeopardy doctor frequently, and sometimes have that doctor continue to round on admitted patients for the next few days. This usage might get the most value out of the system, but the practice should consider if it is more cost-effective, and less stressful for the doctors, if the system were reversed. For example, instead of having the doctor on jeopardy and called in as necessary, the doctor would report to work and be given the day off or let go early when it isn’t busy.

Despite my reservations, if you are convinced the jeopardy system is valuable and cost-effective, keep it in place. However, if your group is thinking about options to handle surge capacity, don’t be too quick to adopt a jeopardy system. It usually falls far short of a perfect solution.

Patient Volume Cap

Another way to address the problem of unpredictable increases in patient volume is to establish a patient volume (e.g., total census) cap for the whole hospitalist practice. Like the jeopardy system, this is an appealingly uncomplicated idea, and hospitalists who have finished residency within the last few years all worked with a cap.

Except for the rarest of exceptions, this is a poor idea and should be avoided if at all possible. I’ll leave for another time a discussion of all the political and financial costs of a cap system, but trust me on this one. It is best to avoid a cap.

Stay Tuned …

Next month, I’ll examine other strategies to provide surge capacity. I think they’re more valuable than the two I’ve mentioned here, but I need to warn you that they aren’t perfect and are more complicated to operationalize. TH

Dr. Nelson has been a practicing hospitalist since 1988 and is co-founder and past president of SHM. He is a principal in Nelson Flores Hospital Medicine Consultants, a national hospitalist practice management consulting firm (www.nelson flores.com). He is course co-director and faculty for SHM’s “Best Practices in Managing a Hospital Medicine Program” course. This column represents his views and is not intended to reflect an official position of SHM.

Unpredictable workloads and frequent interruptions are the things I regard as the most stressful components of work as a hospitalist. Your list might be very different, but I bet unpredictable workloads ranks at least in the top five of every hospitalist’s list.

I’ve discussed interruptions previously (see “Really, It’s Switch-Tasking,” p. 68, November 2008; “Technological Advance or Workplace Setback?” p. 69, December 2008), but this month and next will turn to unpredictable workloads. In other words, what are the strategies available to a hospitalist practice to provide surge capacity in response to such unpredictable increases in patient volume as an uptick in census or daily admissions 50% to 100% above normal? I’ll leave to others the topic of how hospitals respond to such disasters as terrorist attacks, earthquakes, etc.

The Bottom Line

Sadly, there is no magic bullet for the “surge” problem, and no way to protect on-duty hospitalists from the need to work harder when it gets busy. But we needn’t feel too sorry for ourselves; doctors in most other specialties who practice in the hospital face the same problem and tend to rely heavily on simply working harder and longer when it is unusually busy. Sometimes they couple the “work harder” mantra with other strategies, such as calling another doctor in to help.

Hospitalists have a duty to ensure high patient volume doesn’t lead to deterioration in the quality of patient care, but occasionally working longer days than average probably poses a low risk, and might be less risky than the additional handoffs usually associated with having a doctor on “jeopardy” to be called in when it’s busy. Routinely or frequently working unreasonably long days is another story.

The trick for HM programs is to build some surge capacity into the routine daily staffing 1) without exceeding a reasonable budget, while 2) ensuring that the hospitalists don’t simply become accustomed to light workloads as the only reasonable norm, which could lead to them becoming unwilling to accept higher, but still reasonable, workloads when needed. (More on these issues later.) First, I’ll go through what I see as the pros and cons of several approaches to addressing surges in patient volume. All are in use with variable frequency around the country.

“Jeopardy” System

In its most common form, a jeopardy system has an unscheduled doctor each day who must remain available on short notice by pager. When patient volume surges, the unscheduled doctor is paged to come in and help. In most cases, this doctor focuses primarily—or exclusively—on admitting patients for a few hours. So it is most common for this doctor to be called in late in the afternoon or early in the evening. The jeopardy doctor usually turns over all admitted patients to another hospitalist in the group for all subsequent care. In addition to providing surge capacity, the jeopardy doctor almost always is used to cover unexpected absences of scheduled doctors, including illness-related absences.

Sometimes this doctor is paid extra for each day or week spent being “available” on jeopardy duty (not to be confused with jury duty, though it can be equally difficult to get exempted from). Then again, it is not uncommon to have jeopardy duty included in base compensation. However, once a jeopardy doctor is actually called in to work, most practices pay additional compensation, often based on an hourly rate that usually is higher than the average compensation generated per hour for nonjeopardy work.

There are a number of reasonable ways to compensate the jeopardy doctor. You probably can get some good ideas by talking with others in your hospital who function in a similar capacity, such as cath-lab technicians who get called in on nights and weekends.

No definitive data are available to show how common the jeopardy system is, but my experience is that 30% to 50% of HM groups use some form of it. Its popularity is proof that it is a reasonable system, but I’m not convinced. I think it is in use by a lot of groups not because it is an optimal way to ensure surge capacity, but because it is easy to conceptualize and put in place, and because many hospitalists came from residency programs in which the system was standard.

The gaps between theoretical and realized benefits become evident once a practice implements a jeopardy system. For example, it might be really busy today, but Dr. Stravinsky doesn’t call in Dr. Copeland, who is on jeopardy, because next week their roles will be reversed and Dr. Stravinsky sure hopes he won’t be called in. No one wants to be the weak doctor who calls in the jeopardy doctor and spoils what was otherwise a day off.

I’ve worked with a lot of practices who say they have a jeopardy system in place, but when I ask for the last time the jeopardy doctor was called in, they say it has been more than a year, or in some cases never. So even if the policy manual says they have a jeopardy system, the doctors never activate it, so it provides no benefit.

Practices that do utilize the jeopardy doctor have their own problems, such as assigning that doctor’s admissions the next day. The jeopardy doctor might provide some relief today, but they essentially just delay the work of having to get to know all of those new patients until the morning, when everyone is very busy with rounds. So while there might be significant benefit in activating the jeopardy system today, it could just delay the problem of high workload until the next morning, which isn’t much of a net benefit for the practice.

A small number of practices call in the jeopardy doctor frequently, and sometimes have that doctor continue to round on admitted patients for the next few days. This usage might get the most value out of the system, but the practice should consider if it is more cost-effective, and less stressful for the doctors, if the system were reversed. For example, instead of having the doctor on jeopardy and called in as necessary, the doctor would report to work and be given the day off or let go early when it isn’t busy.

Despite my reservations, if you are convinced the jeopardy system is valuable and cost-effective, keep it in place. However, if your group is thinking about options to handle surge capacity, don’t be too quick to adopt a jeopardy system. It usually falls far short of a perfect solution.

Patient Volume Cap

Another way to address the problem of unpredictable increases in patient volume is to establish a patient volume (e.g., total census) cap for the whole hospitalist practice. Like the jeopardy system, this is an appealingly uncomplicated idea, and hospitalists who have finished residency within the last few years all worked with a cap.

Except for the rarest of exceptions, this is a poor idea and should be avoided if at all possible. I’ll leave for another time a discussion of all the political and financial costs of a cap system, but trust me on this one. It is best to avoid a cap.

Stay Tuned …

Next month, I’ll examine other strategies to provide surge capacity. I think they’re more valuable than the two I’ve mentioned here, but I need to warn you that they aren’t perfect and are more complicated to operationalize. TH

Dr. Nelson has been a practicing hospitalist since 1988 and is co-founder and past president of SHM. He is a principal in Nelson Flores Hospital Medicine Consultants, a national hospitalist practice management consulting firm (www.nelson flores.com). He is course co-director and faculty for SHM’s “Best Practices in Managing a Hospital Medicine Program” course. This column represents his views and is not intended to reflect an official position of SHM.

How much has Medicare saved by not paying hospitals when patients get infections?

Hugh Black, DO

Charlotte, N.C.

Dr. Hospitalist responds: Since 2007, the Centers for Medicare and Medicaid Services (CMS) has tried to reduce the number of high-cost, hospital-acquired conditions (HACs), including infections, by encouraging providers to adhere to evidence-based guidelines. Some examples of these hospital-acquired conditions include:

• Catheter-associated urinary tract infections;
• Foreign objects retained after surgery; and
• Stage III and IV pressure ulcers.

CMS requires that acute-care hospitals, “effective with discharges occurring on or after Oct. 1, 2007, submit information on Medicare claims specifying whether diagnoses were present on admission.” Effective Oct. 1, 2008, Medicare no longer pays for charges associated with these HACs. So, if a Medicare beneficiary developed a Stage III pressure ulcer during his stay at an acute-care hospital, CMS would not pay for the incremental cost of the care associated with the “HAC.”

click for large version

The U.S. government, in the May 4, 2010, edition of the Federal Register, reviewed the impact of this program. The data are based on Medicare claims data from October 2008 to June 2009. During this period of time, there were approximately 7.17 million acute-care hospital Medicare discharges.

The total net savings during this nine-month period for all HACs was $16.4 million. Three HACs (Stage III and IV pressure ulcers, DVT/PE after orthopedic procedure, and falls and trauma) accounted for more than $15.1 million in savings. Pro-rated for a 12-month period, the total net savings for all HACs would exceed $20 million.

Falls and trauma accounted for 34% of all HACs reported (11,253), followed by vascular catheter-associated infection (16%) and catheter-associated UTIs (16%). Air embolism and mediastinitis after CABG were the least recorded HACs; both were less than .01% if the total.

The goal is that, over time, with improvement in care, there would be a decrease in the number of hospital discharges where these conditions would be present. Therefore, the net savings would be expected to decline.

Medicare has considered a number of other HACs for this program, and reviewed the numbers of these conditions over the same nine-month period (see “Medicare’s Potential New Hospital-Acquired Conditions,” above). Despite some large numbers, CMS has stated it’s not proposing to add or remove HAC categories at this time. If you are interested in reviewing the entire report, visit http://edocket.access.gpo.gov/2010/pdf/ 2010-9163.pdf. TH

How much has Medicare saved by not paying hospitals when patients get infections?

Hugh Black, DO

Charlotte, N.C.

Dr. Hospitalist responds: Since 2007, the Centers for Medicare and Medicaid Services (CMS) has tried to reduce the number of high-cost, hospital-acquired conditions (HACs), including infections, by encouraging providers to adhere to evidence-based guidelines. Some examples of these hospital-acquired conditions include:

• Catheter-associated urinary tract infections;
• Foreign objects retained after surgery; and
• Stage III and IV pressure ulcers.

CMS requires that acute-care hospitals, “effective with discharges occurring on or after Oct. 1, 2007, submit information on Medicare claims specifying whether diagnoses were present on admission.” Effective Oct. 1, 2008, Medicare no longer pays for charges associated with these HACs. So, if a Medicare beneficiary developed a Stage III pressure ulcer during his stay at an acute-care hospital, CMS would not pay for the incremental cost of the care associated with the “HAC.”

click for large version

The U.S. government, in the May 4, 2010, edition of the Federal Register, reviewed the impact of this program. The data are based on Medicare claims data from October 2008 to June 2009. During this period of time, there were approximately 7.17 million acute-care hospital Medicare discharges.

The total net savings during this nine-month period for all HACs was $16.4 million. Three HACs (Stage III and IV pressure ulcers, DVT/PE after orthopedic procedure, and falls and trauma) accounted for more than $15.1 million in savings. Pro-rated for a 12-month period, the total net savings for all HACs would exceed $20 million.

Falls and trauma accounted for 34% of all HACs reported (11,253), followed by vascular catheter-associated infection (16%) and catheter-associated UTIs (16%). Air embolism and mediastinitis after CABG were the least recorded HACs; both were less than .01% if the total.

The goal is that, over time, with improvement in care, there would be a decrease in the number of hospital discharges where these conditions would be present. Therefore, the net savings would be expected to decline.

Medicare has considered a number of other HACs for this program, and reviewed the numbers of these conditions over the same nine-month period (see “Medicare’s Potential New Hospital-Acquired Conditions,” above). Despite some large numbers, CMS has stated it’s not proposing to add or remove HAC categories at this time. If you are interested in reviewing the entire report, visit http://edocket.access.gpo.gov/2010/pdf/ 2010-9163.pdf. TH

How much has Medicare saved by not paying hospitals when patients get infections?

Hugh Black, DO

Charlotte, N.C.

Dr. Hospitalist responds: Since 2007, the Centers for Medicare and Medicaid Services (CMS) has tried to reduce the number of high-cost, hospital-acquired conditions (HACs), including infections, by encouraging providers to adhere to evidence-based guidelines. Some examples of these hospital-acquired conditions include:

• Catheter-associated urinary tract infections;
• Foreign objects retained after surgery; and
• Stage III and IV pressure ulcers.

CMS requires that acute-care hospitals, “effective with discharges occurring on or after Oct. 1, 2007, submit information on Medicare claims specifying whether diagnoses were present on admission.” Effective Oct. 1, 2008, Medicare no longer pays for charges associated with these HACs. So, if a Medicare beneficiary developed a Stage III pressure ulcer during his stay at an acute-care hospital, CMS would not pay for the incremental cost of the care associated with the “HAC.”

click for large version

The U.S. government, in the May 4, 2010, edition of the Federal Register, reviewed the impact of this program. The data are based on Medicare claims data from October 2008 to June 2009. During this period of time, there were approximately 7.17 million acute-care hospital Medicare discharges.

The total net savings during this nine-month period for all HACs was $16.4 million. Three HACs (Stage III and IV pressure ulcers, DVT/PE after orthopedic procedure, and falls and trauma) accounted for more than $15.1 million in savings. Pro-rated for a 12-month period, the total net savings for all HACs would exceed $20 million.

Falls and trauma accounted for 34% of all HACs reported (11,253), followed by vascular catheter-associated infection (16%) and catheter-associated UTIs (16%). Air embolism and mediastinitis after CABG were the least recorded HACs; both were less than .01% if the total.

The goal is that, over time, with improvement in care, there would be a decrease in the number of hospital discharges where these conditions would be present. Therefore, the net savings would be expected to decline.

Medicare has considered a number of other HACs for this program, and reviewed the numbers of these conditions over the same nine-month period (see “Medicare’s Potential New Hospital-Acquired Conditions,” above). Despite some large numbers, CMS has stated it’s not proposing to add or remove HAC categories at this time. If you are interested in reviewing the entire report, visit http://edocket.access.gpo.gov/2010/pdf/ 2010-9163.pdf. TH

SHM Practice Analysis Committee member Troy Ahlstrom, MD, FHM, discusses the new compensation and productivity report, and gives advice on how best to use benchmarking data in your practice.

Click here to listen.

SHM Practice Analysis Committee member Troy Ahlstrom, MD, FHM, discusses the new compensation and productivity report, and gives advice on how best to use benchmarking data in your practice.

Click here to listen.

SHM Practice Analysis Committee member Troy Ahlstrom, MD, FHM, discusses the new compensation and productivity report, and gives advice on how best to use benchmarking data in your practice.

Click here to listen.

Primary care clinicians play a crucial role in the assessment and management of chronic pain. As many as one-third of primary care patients report having chronic pain. As a result, primary care clinicians are expected to have skills in a broad array of analgesic strategies, including analgesic pharmacotherapy. Ideally, drug treatments for pain are combined with nonpharmacologic strategies, including specific psychological and rehabilitative approaches that also may enhance comfort and promote functional restoration.

Primary care clinicians play a crucial role in the assessment and management of chronic pain. As many as one-third of primary care patients report having chronic pain. As a result, primary care clinicians are expected to have skills in a broad array of analgesic strategies, including analgesic pharmacotherapy. Ideally, drug treatments for pain are combined with nonpharmacologic strategies, including specific psychological and rehabilitative approaches that also may enhance comfort and promote functional restoration.

Primary care clinicians play a crucial role in the assessment and management of chronic pain. As many as one-third of primary care patients report having chronic pain. As a result, primary care clinicians are expected to have skills in a broad array of analgesic strategies, including analgesic pharmacotherapy. Ideally, drug treatments for pain are combined with nonpharmacologic strategies, including specific psychological and rehabilitative approaches that also may enhance comfort and promote functional restoration.

Several weeks before coming to our orthopedic surgery clinic, a 53-year-old man presented to an emergency department because of pain, swelling, and redness in his right foot, which began 3 days before. He recalled no overt trauma, but he was jogging when he first noticed the pain, which he described as a constant aching and rated as high as 8 on a scale of 10.

At that time, he had no fever, chills, or night sweats, no cough, and no shortness of breath. About 10 years ago he was diagnosed with diabetes mellitus, for which he currently takes rosiglitazone (Avandia) 2 mg/day and metformin (Glucophage XR) 500 mg four tablets daily. He also takes ramipril (Altace) 10 mg/day for hypertension, as well as a daily multivitamin. He has a history of hyperlipidemia and a family history of diabetes mellitus and Parkinson disease. He has never been hospitalized and has never undergone surgery.

His blood glucose level was 239 mg/dL (normal 70–110), hemoglobin A1c 9.7% (normal 4%–6%), and white blood cell count 13.41 × 109/L (normal 4.5–11.0).

Based on that evaluation, the patient was admitted to the hospital with a diagnosis of cellulitis. He received intravenous antibiotics for 3 days and then was discharged with a prescription for oral antibiotics. He visited his primary care physician several times over the next 2 to 4 weeks and then was referred to our orthopedic surgery clinic for further evaluation. A neurologic evaluation in our clinic revealed a loss of protective sensation, contraction of the toes, and dryness, consistent with peripheral neuropathy. Given what we know so far, which is the most likely diagnosis?

DIFFERENTIAL DIAGNOSIS

While cellulitis may seem to be the likely diagnosis, if a patient with long-standing diabetes, a history of poor glycemic control, and peripheral neuropathy presents with a red, hot, swollen foot with no history of open ulceration, then Charcot neuroarthropathy should be at the top of the list in the differential diagnosis. Other possibilities include osteomyelitis, acute gout, cellulitis, abscess, neuropathic fracture, and deep venous thrombosis. However, if the patient has no open ulceration or history of an open wound, infection is probably not the culprit. Most diabetic foot infections begin with a direct inoculation through an opening in the skin, such as a diabetic neuropathic foot ulcer.

Further, in the case of cellulitis or deep venous thrombosis, the predominating feature would be asymmetric edema of the leg. Also, the location of the edema and ecchymosis in our patient—namely, the midfoot—leads to suspicion of an acute musculoskeletal injury, particularly Charcot neuroarthropathy of the midfoot and neuropathic fractures in the region of the ecchymotic second and third digits. Acute gout could be discounted because gout pain is severe, with rapid onset, and slowly improves even without treatment.

A COMPLICATION OF DIABETES

Charcot neuroarthropathy presents as a warm, swollen, erythematous foot and ankle, a picture that may be indistinguishable from that of infection. Most patients are in their 50s or 60s, and most present on an emergency basis; they often present late in the process, ie, 2 to 3 months after the initial symptoms, because the symptoms often are not painful.

This condition has been reported to occur with leprosy, syringomyelia, toxic exposure, poliomyelitis, rheumatoid arthritis, multiple sclerosis, congenital neuropathy, traumatic injury, and tertiary syphilis.^1–4 Other conditions that reportedly trigger it include cellulitis, osteomyelitis, synovitis, surgery of the foot, and renal transplant surgery.^5–7 However, today, the most common cause is diabetes mellitus.^4,8

Other names for this condition are diabetic neuropathic osteoarthropathy and neuropathic arthropathy.

Current estimates of its prevalence range from 0.08% in the general diabetic population to 13% in high-risk diabetic patients.⁹

CHARCOT NEUROARTHROPATHY BEGINS WITH PERIPHERAL NEUROPATHY

The pathophysiologic mechanism of Charcot neuroarthropathy is not completely known, but it is thought to begin with peripheral neuropathy. Being insensitive to pain, patients may subject the joints of the foot (most commonly in the midfoot) to stress injuries that lead to the active Charcot process.^10–12 About half of Charcot patients present with pain, as did our patient.

Although our patient remembered no trauma, he was physically active at the time he first noticed the symptoms.

Four stages of Charcot neuroarthropathy are recognized^11–15:

Stage 0 (inflammation), also called Charcot in situ or pre-stage 1, is characterized by erythema, edema, and heat but no structural changes.^11,12,14,15

Stage 1 (development) is characterized by bone resorption, bone fragmentation, and joint dislocation. The swelling, warmth, and redness persist, but there are also radiographic changes such as evidence of debris formation at the articular margins, osseous fragmentation, and joint disruption.

Stage 2 (coalescence) involves bony consolidation, osteosclerosis, and fusion after bony destruction. Absorption of small bone fragments, fusion of joints, and sclerosis of the bone are noticeable.

Stage 3 (reconstruction) is characterized by osteogenesis, decreased osteosclerosis, and progressive fusion.¹³ Healing and new bone formation occur. Decreased sclerosis and bony remodeling signify that the deformity (for example, subluxation, incongruity, and dislocation) is permanent.⁴

MISDIAGNOSIS IS COMMON

Charcot neuroarthropathy is an often overlooked complication in diabetic patients with peripheral neuropathy. A group of experts reported that 25% of patients referred to their facility who had Charcot neuroarthropathy had not received a correct diagnosis at the referring institution.¹⁶ The incorrect diagnoses included infection, gout, arthritis, fracture, venous insufficiency, and tumor.

Laboratory tests can narrow the differential diagnosis

There are no laboratory criteria for the diagnosis of Charcot neuroarthropathy and no hematologic markers, but laboratory testing can help narrow the differential diagnosis. Leukocytosis, an elevated C-reactive protein and erythrocyte sedimentation rate, and recent unexplained hyperglycemia suggest infection.¹⁷ However, unremarkable results on clinical tests in this population may not comprehensively exclude infection.

Our patient’s elevated white blood cell count confused the diagnosis. Further, when he was treated with antibiotics, he reported having less pain, although the edema and erythema continued.

Imaging studies

Although advanced imaging may help confirm the diagnosis of Charcot neuropathy in some patients, it is not always necessary.

Radiography. Radiographic findings are important in diagnosing Charcot neuroarthropathy, although they are less helpful in patients with stage 0 disease, such as our patient, in whom the condition has not yet progressed to fracture or dislocation. All foot and ankle radiographs should be taken in the weight-bearing position. Subtle changes may be missed if non-weight-bearing images are taken.

Magnetic resonance imaging (MRI) can show changes in stage 0, thus enabling treatment to be started sooner,¹⁸ and it is increasingly being recommended for diagnosing Charcot neuroarthropathy, especially in the early stages.³ Although bone scintigraphy and white blood cell scans have been traditionally advocated, MRI offers the highest diagnostic accuracy.¹⁹ Signs on MRI consistent with Charcot neuroarthropathy include ligamentous disruption, concomitant joint deformity, and the center of signal enhancement within joints and subchondral bone.²⁰

MRI can also differentiate Charcot neuroarthropathy from transient regional osteoporosis. The latter has a different anatomic location and does not cause fractures and dislocations, and patients do not have a clinical history of pain.

Another condition MRI can identify is complex regional pain syndrome. In this condition, patients have no radiographic abnormalities except for periarticular osteopenia, but they may have severe pain out of proportion with the clinical appearance, and they may develop soft-tissue deformity in the late stages, which is not seen in Charcot neuroarthropathy.

Positron emission tomography (PET) with fluorine-18 fluorodeoxyglucose is also gaining support,²¹ especially when combined with computed tomography (CT). This PET-CT hybrid has better anatomic localization than PET alone.

PET-CT is very reliable for differentiating Charcot neuroarthropathy from osteomyelitis, a distinction that can be difficult to make when Charcot neuroarthropathy is complicated by adjacent loss of skin integrity. The sensitivity of PET-CT in this situation has been reported as 100%, and its sensitivity 93.8%.²²

Patients with Charcot neuroarthropathy demonstrate a low-intensity diffuse uptake that is easily distinguishable from normal joints on visual examination of the images. In addition, the maximum standardized uptake value, a quantitative measurement, is low to intermediate in Charcot neuroarthropathy but significantly higher in osteomyelitis. In one study,²² the mean standardized uptake values were 0.42 in normal feet, 1.3 in Charcot neuroarthropathy, and 4.38 in osteomyelitis.

TREATMENT: IMMOBILIZATION, BISPHOSPHONATES, SURGERY

The goals of treatment for acute or quiescent Charcot neuroarthropathy should be to maintain or achieve structural stability of the foot and ankle, to prevent skin ulceration, and to preserve the plantigrade shape of the foot so that prescription footwear can be used.

Patient and family education is important for compliance with the regimen, particularly because patients with diabetic neuropathy lack the protective pain response.

Immobilization. A total-contact cast is worn until the redness, swelling, and heat subside, generally 8 to 12 weeks, after which the patient should use removable braces or a Charcot restraint orthotic walker for a total of 4 to 6 months of treatment.²³ The cast is typically changed every 1 to 2 weeks as the swelling subsides to minimize irritation to the insensate limb.

Many physicians also recommend elastic stockings (eg, Stockinette) or an elastic tubular bandage (eg, Tubigrip) to reduce edema under the cast.

Bisphosphonates. Some clinicians also prescribe bisphosphonates in the early stages of treatment, as the bone mineral density of the affected foot is low.²⁴ Unfortunately, while these drugs can significantly reduce the levels of bone turnover markers, temperature, and pain, evidence of clinical benefit such as an earlier return to ambulation or radiographic improvement is weak at best.

Surgery is reserved for severe ankle and midfoot deformities that are susceptible to skin ulcerations and that make braces and orthotic devices difficult to use.

TREATMENT OUTCOME

The patient’s condition resolved, with eventual multiplanar deformity and with widening of the midfoot and increased pressure points, particularly to the first ray. He is able to wear an extra-depth shoe, with a custom totalcontact inlay. He continues his profession as an attorney and goes about his normal daily activities; however, he is no longer able to golf and must limit his walking. He subsequently developed ulcerations to both feet, but they resolved with conservative wound care and surgical care. He is seen in the diabetic foot clinic every 6 to 8 weeks.

Several weeks before coming to our orthopedic surgery clinic, a 53-year-old man presented to an emergency department because of pain, swelling, and redness in his right foot, which began 3 days before. He recalled no overt trauma, but he was jogging when he first noticed the pain, which he described as a constant aching and rated as high as 8 on a scale of 10.

At that time, he had no fever, chills, or night sweats, no cough, and no shortness of breath. About 10 years ago he was diagnosed with diabetes mellitus, for which he currently takes rosiglitazone (Avandia) 2 mg/day and metformin (Glucophage XR) 500 mg four tablets daily. He also takes ramipril (Altace) 10 mg/day for hypertension, as well as a daily multivitamin. He has a history of hyperlipidemia and a family history of diabetes mellitus and Parkinson disease. He has never been hospitalized and has never undergone surgery.

His blood glucose level was 239 mg/dL (normal 70–110), hemoglobin A1c 9.7% (normal 4%–6%), and white blood cell count 13.41 × 109/L (normal 4.5–11.0).

Based on that evaluation, the patient was admitted to the hospital with a diagnosis of cellulitis. He received intravenous antibiotics for 3 days and then was discharged with a prescription for oral antibiotics. He visited his primary care physician several times over the next 2 to 4 weeks and then was referred to our orthopedic surgery clinic for further evaluation. A neurologic evaluation in our clinic revealed a loss of protective sensation, contraction of the toes, and dryness, consistent with peripheral neuropathy. Given what we know so far, which is the most likely diagnosis?

DIFFERENTIAL DIAGNOSIS

While cellulitis may seem to be the likely diagnosis, if a patient with long-standing diabetes, a history of poor glycemic control, and peripheral neuropathy presents with a red, hot, swollen foot with no history of open ulceration, then Charcot neuroarthropathy should be at the top of the list in the differential diagnosis. Other possibilities include osteomyelitis, acute gout, cellulitis, abscess, neuropathic fracture, and deep venous thrombosis. However, if the patient has no open ulceration or history of an open wound, infection is probably not the culprit. Most diabetic foot infections begin with a direct inoculation through an opening in the skin, such as a diabetic neuropathic foot ulcer.

Further, in the case of cellulitis or deep venous thrombosis, the predominating feature would be asymmetric edema of the leg. Also, the location of the edema and ecchymosis in our patient—namely, the midfoot—leads to suspicion of an acute musculoskeletal injury, particularly Charcot neuroarthropathy of the midfoot and neuropathic fractures in the region of the ecchymotic second and third digits. Acute gout could be discounted because gout pain is severe, with rapid onset, and slowly improves even without treatment.

A COMPLICATION OF DIABETES

Charcot neuroarthropathy presents as a warm, swollen, erythematous foot and ankle, a picture that may be indistinguishable from that of infection. Most patients are in their 50s or 60s, and most present on an emergency basis; they often present late in the process, ie, 2 to 3 months after the initial symptoms, because the symptoms often are not painful.

This condition has been reported to occur with leprosy, syringomyelia, toxic exposure, poliomyelitis, rheumatoid arthritis, multiple sclerosis, congenital neuropathy, traumatic injury, and tertiary syphilis.^1–4 Other conditions that reportedly trigger it include cellulitis, osteomyelitis, synovitis, surgery of the foot, and renal transplant surgery.^5–7 However, today, the most common cause is diabetes mellitus.^4,8

Other names for this condition are diabetic neuropathic osteoarthropathy and neuropathic arthropathy.

Current estimates of its prevalence range from 0.08% in the general diabetic population to 13% in high-risk diabetic patients.⁹

CHARCOT NEUROARTHROPATHY BEGINS WITH PERIPHERAL NEUROPATHY

The pathophysiologic mechanism of Charcot neuroarthropathy is not completely known, but it is thought to begin with peripheral neuropathy. Being insensitive to pain, patients may subject the joints of the foot (most commonly in the midfoot) to stress injuries that lead to the active Charcot process.^10–12 About half of Charcot patients present with pain, as did our patient.

Although our patient remembered no trauma, he was physically active at the time he first noticed the symptoms.

Four stages of Charcot neuroarthropathy are recognized^11–15:

Stage 0 (inflammation), also called Charcot in situ or pre-stage 1, is characterized by erythema, edema, and heat but no structural changes.^11,12,14,15

Stage 1 (development) is characterized by bone resorption, bone fragmentation, and joint dislocation. The swelling, warmth, and redness persist, but there are also radiographic changes such as evidence of debris formation at the articular margins, osseous fragmentation, and joint disruption.

Stage 2 (coalescence) involves bony consolidation, osteosclerosis, and fusion after bony destruction. Absorption of small bone fragments, fusion of joints, and sclerosis of the bone are noticeable.

Stage 3 (reconstruction) is characterized by osteogenesis, decreased osteosclerosis, and progressive fusion.¹³ Healing and new bone formation occur. Decreased sclerosis and bony remodeling signify that the deformity (for example, subluxation, incongruity, and dislocation) is permanent.⁴

MISDIAGNOSIS IS COMMON

Charcot neuroarthropathy is an often overlooked complication in diabetic patients with peripheral neuropathy. A group of experts reported that 25% of patients referred to their facility who had Charcot neuroarthropathy had not received a correct diagnosis at the referring institution.¹⁶ The incorrect diagnoses included infection, gout, arthritis, fracture, venous insufficiency, and tumor.

Laboratory tests can narrow the differential diagnosis

There are no laboratory criteria for the diagnosis of Charcot neuroarthropathy and no hematologic markers, but laboratory testing can help narrow the differential diagnosis. Leukocytosis, an elevated C-reactive protein and erythrocyte sedimentation rate, and recent unexplained hyperglycemia suggest infection.¹⁷ However, unremarkable results on clinical tests in this population may not comprehensively exclude infection.

Our patient’s elevated white blood cell count confused the diagnosis. Further, when he was treated with antibiotics, he reported having less pain, although the edema and erythema continued.

Imaging studies

Although advanced imaging may help confirm the diagnosis of Charcot neuropathy in some patients, it is not always necessary.

Radiography. Radiographic findings are important in diagnosing Charcot neuroarthropathy, although they are less helpful in patients with stage 0 disease, such as our patient, in whom the condition has not yet progressed to fracture or dislocation. All foot and ankle radiographs should be taken in the weight-bearing position. Subtle changes may be missed if non-weight-bearing images are taken.

Magnetic resonance imaging (MRI) can show changes in stage 0, thus enabling treatment to be started sooner,¹⁸ and it is increasingly being recommended for diagnosing Charcot neuroarthropathy, especially in the early stages.³ Although bone scintigraphy and white blood cell scans have been traditionally advocated, MRI offers the highest diagnostic accuracy.¹⁹ Signs on MRI consistent with Charcot neuroarthropathy include ligamentous disruption, concomitant joint deformity, and the center of signal enhancement within joints and subchondral bone.²⁰

MRI can also differentiate Charcot neuroarthropathy from transient regional osteoporosis. The latter has a different anatomic location and does not cause fractures and dislocations, and patients do not have a clinical history of pain.

Another condition MRI can identify is complex regional pain syndrome. In this condition, patients have no radiographic abnormalities except for periarticular osteopenia, but they may have severe pain out of proportion with the clinical appearance, and they may develop soft-tissue deformity in the late stages, which is not seen in Charcot neuroarthropathy.

Positron emission tomography (PET) with fluorine-18 fluorodeoxyglucose is also gaining support,²¹ especially when combined with computed tomography (CT). This PET-CT hybrid has better anatomic localization than PET alone.

PET-CT is very reliable for differentiating Charcot neuroarthropathy from osteomyelitis, a distinction that can be difficult to make when Charcot neuroarthropathy is complicated by adjacent loss of skin integrity. The sensitivity of PET-CT in this situation has been reported as 100%, and its sensitivity 93.8%.²²

Patients with Charcot neuroarthropathy demonstrate a low-intensity diffuse uptake that is easily distinguishable from normal joints on visual examination of the images. In addition, the maximum standardized uptake value, a quantitative measurement, is low to intermediate in Charcot neuroarthropathy but significantly higher in osteomyelitis. In one study,²² the mean standardized uptake values were 0.42 in normal feet, 1.3 in Charcot neuroarthropathy, and 4.38 in osteomyelitis.

TREATMENT: IMMOBILIZATION, BISPHOSPHONATES, SURGERY

The goals of treatment for acute or quiescent Charcot neuroarthropathy should be to maintain or achieve structural stability of the foot and ankle, to prevent skin ulceration, and to preserve the plantigrade shape of the foot so that prescription footwear can be used.

Patient and family education is important for compliance with the regimen, particularly because patients with diabetic neuropathy lack the protective pain response.

Immobilization. A total-contact cast is worn until the redness, swelling, and heat subside, generally 8 to 12 weeks, after which the patient should use removable braces or a Charcot restraint orthotic walker for a total of 4 to 6 months of treatment.²³ The cast is typically changed every 1 to 2 weeks as the swelling subsides to minimize irritation to the insensate limb.

Many physicians also recommend elastic stockings (eg, Stockinette) or an elastic tubular bandage (eg, Tubigrip) to reduce edema under the cast.

Bisphosphonates. Some clinicians also prescribe bisphosphonates in the early stages of treatment, as the bone mineral density of the affected foot is low.²⁴ Unfortunately, while these drugs can significantly reduce the levels of bone turnover markers, temperature, and pain, evidence of clinical benefit such as an earlier return to ambulation or radiographic improvement is weak at best.

Surgery is reserved for severe ankle and midfoot deformities that are susceptible to skin ulcerations and that make braces and orthotic devices difficult to use.

TREATMENT OUTCOME

The patient’s condition resolved, with eventual multiplanar deformity and with widening of the midfoot and increased pressure points, particularly to the first ray. He is able to wear an extra-depth shoe, with a custom totalcontact inlay. He continues his profession as an attorney and goes about his normal daily activities; however, he is no longer able to golf and must limit his walking. He subsequently developed ulcerations to both feet, but they resolved with conservative wound care and surgical care. He is seen in the diabetic foot clinic every 6 to 8 weeks.

Several weeks before coming to our orthopedic surgery clinic, a 53-year-old man presented to an emergency department because of pain, swelling, and redness in his right foot, which began 3 days before. He recalled no overt trauma, but he was jogging when he first noticed the pain, which he described as a constant aching and rated as high as 8 on a scale of 10.

At that time, he had no fever, chills, or night sweats, no cough, and no shortness of breath. About 10 years ago he was diagnosed with diabetes mellitus, for which he currently takes rosiglitazone (Avandia) 2 mg/day and metformin (Glucophage XR) 500 mg four tablets daily. He also takes ramipril (Altace) 10 mg/day for hypertension, as well as a daily multivitamin. He has a history of hyperlipidemia and a family history of diabetes mellitus and Parkinson disease. He has never been hospitalized and has never undergone surgery.

His blood glucose level was 239 mg/dL (normal 70–110), hemoglobin A1c 9.7% (normal 4%–6%), and white blood cell count 13.41 × 109/L (normal 4.5–11.0).

Based on that evaluation, the patient was admitted to the hospital with a diagnosis of cellulitis. He received intravenous antibiotics for 3 days and then was discharged with a prescription for oral antibiotics. He visited his primary care physician several times over the next 2 to 4 weeks and then was referred to our orthopedic surgery clinic for further evaluation. A neurologic evaluation in our clinic revealed a loss of protective sensation, contraction of the toes, and dryness, consistent with peripheral neuropathy. Given what we know so far, which is the most likely diagnosis?

DIFFERENTIAL DIAGNOSIS

While cellulitis may seem to be the likely diagnosis, if a patient with long-standing diabetes, a history of poor glycemic control, and peripheral neuropathy presents with a red, hot, swollen foot with no history of open ulceration, then Charcot neuroarthropathy should be at the top of the list in the differential diagnosis. Other possibilities include osteomyelitis, acute gout, cellulitis, abscess, neuropathic fracture, and deep venous thrombosis. However, if the patient has no open ulceration or history of an open wound, infection is probably not the culprit. Most diabetic foot infections begin with a direct inoculation through an opening in the skin, such as a diabetic neuropathic foot ulcer.

Further, in the case of cellulitis or deep venous thrombosis, the predominating feature would be asymmetric edema of the leg. Also, the location of the edema and ecchymosis in our patient—namely, the midfoot—leads to suspicion of an acute musculoskeletal injury, particularly Charcot neuroarthropathy of the midfoot and neuropathic fractures in the region of the ecchymotic second and third digits. Acute gout could be discounted because gout pain is severe, with rapid onset, and slowly improves even without treatment.

A COMPLICATION OF DIABETES

Charcot neuroarthropathy presents as a warm, swollen, erythematous foot and ankle, a picture that may be indistinguishable from that of infection. Most patients are in their 50s or 60s, and most present on an emergency basis; they often present late in the process, ie, 2 to 3 months after the initial symptoms, because the symptoms often are not painful.

This condition has been reported to occur with leprosy, syringomyelia, toxic exposure, poliomyelitis, rheumatoid arthritis, multiple sclerosis, congenital neuropathy, traumatic injury, and tertiary syphilis.^1–4 Other conditions that reportedly trigger it include cellulitis, osteomyelitis, synovitis, surgery of the foot, and renal transplant surgery.^5–7 However, today, the most common cause is diabetes mellitus.^4,8

Other names for this condition are diabetic neuropathic osteoarthropathy and neuropathic arthropathy.

Current estimates of its prevalence range from 0.08% in the general diabetic population to 13% in high-risk diabetic patients.⁹

CHARCOT NEUROARTHROPATHY BEGINS WITH PERIPHERAL NEUROPATHY

The pathophysiologic mechanism of Charcot neuroarthropathy is not completely known, but it is thought to begin with peripheral neuropathy. Being insensitive to pain, patients may subject the joints of the foot (most commonly in the midfoot) to stress injuries that lead to the active Charcot process.^10–12 About half of Charcot patients present with pain, as did our patient.

Although our patient remembered no trauma, he was physically active at the time he first noticed the symptoms.

Four stages of Charcot neuroarthropathy are recognized^11–15:

Stage 0 (inflammation), also called Charcot in situ or pre-stage 1, is characterized by erythema, edema, and heat but no structural changes.^11,12,14,15

Stage 1 (development) is characterized by bone resorption, bone fragmentation, and joint dislocation. The swelling, warmth, and redness persist, but there are also radiographic changes such as evidence of debris formation at the articular margins, osseous fragmentation, and joint disruption.

Stage 2 (coalescence) involves bony consolidation, osteosclerosis, and fusion after bony destruction. Absorption of small bone fragments, fusion of joints, and sclerosis of the bone are noticeable.

Stage 3 (reconstruction) is characterized by osteogenesis, decreased osteosclerosis, and progressive fusion.¹³ Healing and new bone formation occur. Decreased sclerosis and bony remodeling signify that the deformity (for example, subluxation, incongruity, and dislocation) is permanent.⁴

MISDIAGNOSIS IS COMMON

Charcot neuroarthropathy is an often overlooked complication in diabetic patients with peripheral neuropathy. A group of experts reported that 25% of patients referred to their facility who had Charcot neuroarthropathy had not received a correct diagnosis at the referring institution.¹⁶ The incorrect diagnoses included infection, gout, arthritis, fracture, venous insufficiency, and tumor.

Laboratory tests can narrow the differential diagnosis

There are no laboratory criteria for the diagnosis of Charcot neuroarthropathy and no hematologic markers, but laboratory testing can help narrow the differential diagnosis. Leukocytosis, an elevated C-reactive protein and erythrocyte sedimentation rate, and recent unexplained hyperglycemia suggest infection.¹⁷ However, unremarkable results on clinical tests in this population may not comprehensively exclude infection.

Our patient’s elevated white blood cell count confused the diagnosis. Further, when he was treated with antibiotics, he reported having less pain, although the edema and erythema continued.

Imaging studies

Although advanced imaging may help confirm the diagnosis of Charcot neuropathy in some patients, it is not always necessary.

Radiography. Radiographic findings are important in diagnosing Charcot neuroarthropathy, although they are less helpful in patients with stage 0 disease, such as our patient, in whom the condition has not yet progressed to fracture or dislocation. All foot and ankle radiographs should be taken in the weight-bearing position. Subtle changes may be missed if non-weight-bearing images are taken.

Magnetic resonance imaging (MRI) can show changes in stage 0, thus enabling treatment to be started sooner,¹⁸ and it is increasingly being recommended for diagnosing Charcot neuroarthropathy, especially in the early stages.³ Although bone scintigraphy and white blood cell scans have been traditionally advocated, MRI offers the highest diagnostic accuracy.¹⁹ Signs on MRI consistent with Charcot neuroarthropathy include ligamentous disruption, concomitant joint deformity, and the center of signal enhancement within joints and subchondral bone.²⁰

MRI can also differentiate Charcot neuroarthropathy from transient regional osteoporosis. The latter has a different anatomic location and does not cause fractures and dislocations, and patients do not have a clinical history of pain.

Another condition MRI can identify is complex regional pain syndrome. In this condition, patients have no radiographic abnormalities except for periarticular osteopenia, but they may have severe pain out of proportion with the clinical appearance, and they may develop soft-tissue deformity in the late stages, which is not seen in Charcot neuroarthropathy.

Positron emission tomography (PET) with fluorine-18 fluorodeoxyglucose is also gaining support,²¹ especially when combined with computed tomography (CT). This PET-CT hybrid has better anatomic localization than PET alone.

PET-CT is very reliable for differentiating Charcot neuroarthropathy from osteomyelitis, a distinction that can be difficult to make when Charcot neuroarthropathy is complicated by adjacent loss of skin integrity. The sensitivity of PET-CT in this situation has been reported as 100%, and its sensitivity 93.8%.²²

Patients with Charcot neuroarthropathy demonstrate a low-intensity diffuse uptake that is easily distinguishable from normal joints on visual examination of the images. In addition, the maximum standardized uptake value, a quantitative measurement, is low to intermediate in Charcot neuroarthropathy but significantly higher in osteomyelitis. In one study,²² the mean standardized uptake values were 0.42 in normal feet, 1.3 in Charcot neuroarthropathy, and 4.38 in osteomyelitis.

TREATMENT: IMMOBILIZATION, BISPHOSPHONATES, SURGERY

The goals of treatment for acute or quiescent Charcot neuroarthropathy should be to maintain or achieve structural stability of the foot and ankle, to prevent skin ulceration, and to preserve the plantigrade shape of the foot so that prescription footwear can be used.

Patient and family education is important for compliance with the regimen, particularly because patients with diabetic neuropathy lack the protective pain response.

Immobilization. A total-contact cast is worn until the redness, swelling, and heat subside, generally 8 to 12 weeks, after which the patient should use removable braces or a Charcot restraint orthotic walker for a total of 4 to 6 months of treatment.²³ The cast is typically changed every 1 to 2 weeks as the swelling subsides to minimize irritation to the insensate limb.

Many physicians also recommend elastic stockings (eg, Stockinette) or an elastic tubular bandage (eg, Tubigrip) to reduce edema under the cast.

Bisphosphonates. Some clinicians also prescribe bisphosphonates in the early stages of treatment, as the bone mineral density of the affected foot is low.²⁴ Unfortunately, while these drugs can significantly reduce the levels of bone turnover markers, temperature, and pain, evidence of clinical benefit such as an earlier return to ambulation or radiographic improvement is weak at best.

Surgery is reserved for severe ankle and midfoot deformities that are susceptible to skin ulcerations and that make braces and orthotic devices difficult to use.

TREATMENT OUTCOME

The patient’s condition resolved, with eventual multiplanar deformity and with widening of the midfoot and increased pressure points, particularly to the first ray. He is able to wear an extra-depth shoe, with a custom totalcontact inlay. He continues his profession as an attorney and goes about his normal daily activities; however, he is no longer able to golf and must limit his walking. He subsequently developed ulcerations to both feet, but they resolved with conservative wound care and surgical care. He is seen in the diabetic foot clinic every 6 to 8 weeks.

A 45-year-old woman who has been undergoing hemodialysis for 20 years presents with diffuse bone pain, pruritic skin, muscle weakness, and disability. About 8 years ago, she was diagnosed with uremic hyperparathyroidism and underwent two parathyroidectomy procedures and eight sessions of percutaneous alcohol ablation of the parathyroid gland.

A technetium-99m sestamibi radionuclide scan shows bilateral parathyroid hyperplasia with no ectopic parathyroid adenomas. Although a surgeon she consulted 1 month ago declined to perform another parathyroidectomy for technical reasons, another surgeon agreed to do it at this time. Parathyroidectomy was successfully performed, after which the parathyroid hormone level decreased drastically, to 85 pg/mL.

To our surprise, her total body length increased by 6 to 7 cm after surgery, with partial straightening of the back and legs noted about 4 to 5 weeks after surgery. Furthermore, she was able to walk a short distance just a few weeks after surgery. Unfortunately, she died from sepsis the next year.

SEVERE UREMIC HYPERPARATHYROIDISM: A CLINICAL DILEMMA

The clinical appearance of reduced body length and diffuse bony deformity leading to “shrinking” as a consequence of prolonged severe uremic hyperparathyroidism has only rarely been reported.¹ However, it is not uncommon for surgeons to decide against parathyroidectomy because of concerns of extensive subcutaneous fibrosis and recurrent laryngeal nerve damage associated with previous operations. The result is that the patient’s uremic hyperparathyroidism goes untreated, increasing the risk of long-term complications, as in this patient.

TYPICAL RADIOGRAPHIC FEATURES

Why the body length increases after parathyroidectomy is not yet known, but plausible mechanisms are the effects of a recovery of muscle strength and the vigorous bony remineralization that strengthens weight-bearing bones after resolution of uremic hyperparathyroidism.

THE DANGERS OF DELAYED TREATMENT

Delaying parathyroidectomy may induce prolonged and severe uremic hyperparathyroidism, as in this patient. Nevertheless, despite the delay, surgery was able to partially ameliorate the symptoms of hyperparathyroidism and improve the extreme bone deformity. However, the patient’s informed consent, a detailed preoperative evaluation, and exclusion of ectopic parathyroid adenomas are imperative before surgical treatment.

A 45-year-old woman who has been undergoing hemodialysis for 20 years presents with diffuse bone pain, pruritic skin, muscle weakness, and disability. About 8 years ago, she was diagnosed with uremic hyperparathyroidism and underwent two parathyroidectomy procedures and eight sessions of percutaneous alcohol ablation of the parathyroid gland.

A technetium-99m sestamibi radionuclide scan shows bilateral parathyroid hyperplasia with no ectopic parathyroid adenomas. Although a surgeon she consulted 1 month ago declined to perform another parathyroidectomy for technical reasons, another surgeon agreed to do it at this time. Parathyroidectomy was successfully performed, after which the parathyroid hormone level decreased drastically, to 85 pg/mL.

To our surprise, her total body length increased by 6 to 7 cm after surgery, with partial straightening of the back and legs noted about 4 to 5 weeks after surgery. Furthermore, she was able to walk a short distance just a few weeks after surgery. Unfortunately, she died from sepsis the next year.

SEVERE UREMIC HYPERPARATHYROIDISM: A CLINICAL DILEMMA

The clinical appearance of reduced body length and diffuse bony deformity leading to “shrinking” as a consequence of prolonged severe uremic hyperparathyroidism has only rarely been reported.¹ However, it is not uncommon for surgeons to decide against parathyroidectomy because of concerns of extensive subcutaneous fibrosis and recurrent laryngeal nerve damage associated with previous operations. The result is that the patient’s uremic hyperparathyroidism goes untreated, increasing the risk of long-term complications, as in this patient.

TYPICAL RADIOGRAPHIC FEATURES

Why the body length increases after parathyroidectomy is not yet known, but plausible mechanisms are the effects of a recovery of muscle strength and the vigorous bony remineralization that strengthens weight-bearing bones after resolution of uremic hyperparathyroidism.

THE DANGERS OF DELAYED TREATMENT

Delaying parathyroidectomy may induce prolonged and severe uremic hyperparathyroidism, as in this patient. Nevertheless, despite the delay, surgery was able to partially ameliorate the symptoms of hyperparathyroidism and improve the extreme bone deformity. However, the patient’s informed consent, a detailed preoperative evaluation, and exclusion of ectopic parathyroid adenomas are imperative before surgical treatment.

A 45-year-old woman who has been undergoing hemodialysis for 20 years presents with diffuse bone pain, pruritic skin, muscle weakness, and disability. About 8 years ago, she was diagnosed with uremic hyperparathyroidism and underwent two parathyroidectomy procedures and eight sessions of percutaneous alcohol ablation of the parathyroid gland.

A technetium-99m sestamibi radionuclide scan shows bilateral parathyroid hyperplasia with no ectopic parathyroid adenomas. Although a surgeon she consulted 1 month ago declined to perform another parathyroidectomy for technical reasons, another surgeon agreed to do it at this time. Parathyroidectomy was successfully performed, after which the parathyroid hormone level decreased drastically, to 85 pg/mL.

To our surprise, her total body length increased by 6 to 7 cm after surgery, with partial straightening of the back and legs noted about 4 to 5 weeks after surgery. Furthermore, she was able to walk a short distance just a few weeks after surgery. Unfortunately, she died from sepsis the next year.

SEVERE UREMIC HYPERPARATHYROIDISM: A CLINICAL DILEMMA

The clinical appearance of reduced body length and diffuse bony deformity leading to “shrinking” as a consequence of prolonged severe uremic hyperparathyroidism has only rarely been reported.¹ However, it is not uncommon for surgeons to decide against parathyroidectomy because of concerns of extensive subcutaneous fibrosis and recurrent laryngeal nerve damage associated with previous operations. The result is that the patient’s uremic hyperparathyroidism goes untreated, increasing the risk of long-term complications, as in this patient.

TYPICAL RADIOGRAPHIC FEATURES

Why the body length increases after parathyroidectomy is not yet known, but plausible mechanisms are the effects of a recovery of muscle strength and the vigorous bony remineralization that strengthens weight-bearing bones after resolution of uremic hyperparathyroidism.

THE DANGERS OF DELAYED TREATMENT

Delaying parathyroidectomy may induce prolonged and severe uremic hyperparathyroidism, as in this patient. Nevertheless, despite the delay, surgery was able to partially ameliorate the symptoms of hyperparathyroidism and improve the extreme bone deformity. However, the patient’s informed consent, a detailed preoperative evaluation, and exclusion of ectopic parathyroid adenomas are imperative before surgical treatment.

Antibiotics are indicated for primary bacterial aspiration pneumonia and secondary bacterial infection of aspiration (chemical) pneumonitis, but not for uncomplicated chemical pneumonitis.

THREE TYPES OF ‘ASPIRATION PNEUMONIA’

Aspiration pneumonia is a broad and vague term mainly used to refer to the pulmonary consequences of abnormal entry of exogenous or endogenous substances into the lower airways. It can be classified as:

• Aspiration (chemical) pneumonitis
• Primary bacterial aspiration pneumonia
• Secondary bacterial infection of chemical pneumonitis.

These three are sometimes difficult to differentiate, as their signs and symptoms can overlap.

CHEMICAL PNEUMONITIS

Aspiration of stomach contents is relatively common, even in healthy people, and usually has no clinical consequences.¹ However, it has also been closely related to community-acquired and nosocomial pneumonia in some studies.^2,3

Chemical pneumonitis is usually a consequence of the aspiration of a large volume (≥ 4 mL/kg) of sterile acidic (pH < 2.5) gastric contents into the lower airways (Mendelson syndrome).^4,5 The clinical picture varies from asymptomatic to signs of severe dyspnea, hypoxia, cough, and low-grade fever; these signs and symptoms may develop rapidly, within minutes to hours after a witnessed or suspected episode of aspiration.^2,6,7 However, they represent an inflammatory reaction to the gastric acid rather than a reaction to bacterial infection.^8–10

Chemical pneumonitis affects the most dependent regions of the lungs

Chest radiography shows infiltrates in the most dependent regions of the lung. If aspiration occurs while the patient is supine, the posterior segments of the upper lobes and the apical segments of the lower lobes are most affected. The basal segments of the lower lobes are usually affected if aspiration occurs while the patient is standing or upright.^1,2,11,12

Clinical course varies

The clinical course varies. In almost 60% of cases, the patient’s condition improves and the lung infiltrates resolve rapidly, within 2 to 4 days. On the other hand, in about 15% of cases, the patient’s condition deteriorates quickly, within 24 to 36 hours, and progresses to hypoxic respiratory failure and acute respiratory distress syndrome.

In the other 25% of cases, the patient’s condition may improve initially but then worsen as a secondary bacterial infection sets in. The death rate in these patients is almost three times higher than the rate in patients with uncomplicated chemical pneumonitis.^11,13

Treatment of uncomplicated cases is mainly supportive

The treatment of uncomplicated chemical pneumonitis involves supportive measures such as airway clearance, oxygen supplementation, and positive pressure ventilation if needed. An obstructing foreign body may need to be removed.^12,14 Corticosteroids have been tried, without success.^11–13,15

Empiric antibiotic treatment is controversial

Chemical pneumonitis can be difficult to differentiate from bacterial aspiration pneumonia, and whether to give antibiotics is controversial. ¹⁶ A survey of current practices among intensivists showed that antimicrobial therapy was often given empirically for noninfectious chemical pneumonitis.¹⁷ This practice raises concerns of higher treatment costs and antibiotic resistance.^16,18,19 Additionally, antibiotics do not seem to alter the clinical outcome, including radiographic resolution, duration of hospitalization, or death rate, nor do they influence the subsequent development of infection.^1,11,13,20

In cases of witnessed or strongly suspected aspiration of gastric contents, antibiotics are not warranted since bacterial infection is not likely to be the cause of any signs or symptoms. ^2,7,16 However, to detect secondary infection early, the patient’s respiratory status should be monitored carefully and chest radiography should be repeated.

In less clear-cut cases, ie, if it is not clear whether the patient actually has chemical pneumonitis or primary bacterial aspiration pneumonia, it is prudent to start antibiotics empirically after obtaining lower-respiratory-tract secretions for stains and cultures, and then to reassess within 48 to 72 hours. The antibiotics can be discontinued if the patient has rapid clinical and radiographic improvement and negative cultures. Those whose condition does not improve or who have positive cultures should receive a full course of antibiotics.^21,22

PRIMARY BACTERIAL ASPIRATION PNEUMONIA

Primary bacterial aspiration pneumonia—ie, caused by bacteria residing in the upper airways and stomach gaining access to lower airways through aspiration in small or large amounts—is the most common form of aspiration pneumonia, although the actual episode of aspiration is seldom observed.

Signs of bacterial pneumonia

Primary bacterial aspiration pneumonia bears the hallmarks of bacterial pneumonia.¹² The clinical picture is more indolent than chemical pneumonitis and includes cough, fever, and putrid sputum, mainly in patients who have clinical conditions predisposing to aspiration (eg, coma, stroke, alcoholism, poor dentition, tube feedings).^1,12,20

The characteristic signs on chest radiography are infiltrates involving mainly the lung bases (the right more then the left). If untreated or inadequately treated, complications such as lung abscess, empyema, bronchiectasis, and broncopleural fistula are common.²³

Are aerobic organisms replacing anaerobic ones in the community?

The causative organisms in community-acquired aspiration pneumonia are still debated despite abundant research. Older studies^1,24,25 found mostly anaerobic organisms (pepto-streptococci, peptococci, Fusobacterium, Prevotela, Bacteroides) as the underlying pathogens, whereas more recent studies^16,26,27 found mostly aerobic organisms (Streptococcus pneumoniae, Haemophilus influenzae, Staphylococcus aureus, Enterobacteriaceae) and failed to recover anaerobic organisms. These discrepancies may be the result of different techniques used to isolate organisms: older studies used transtracheal sampling, and transtracheal aspirates may be easily contaminated or colonized by oropharyngeal flora; more recent studies used protected specimen brushes to collect lower-airway specimens.²

In addition, the pathogenic organisms that predominate in community-acquired aspiration pneumonia, as listed above, are different from those most often found in nosocomial cases; gram-negative bacilli (Pseudomonas aeruginosa, Klebsiella pneumoniae, Escherichia coli) are most often isolated in patients with aspiration pneumonia acquired in hospitals and nursing homes.^16,27,28S aureus also is an important causative organism in nosocomial cases.^16,28

Knowing the causative organisms in bacterial aspiration pneumonia is important for guiding antimicrobial therapy.

Antibiotics are required for bacterial aspiration pneumonia

A course of antibiotics is required for bacterial aspiration pneumonia. While there are no definitive recommendations for the duration of treatment, 7 to 8 days is probably appropriate in uncomplicated cases (ie, no lung abscess, empyema, bronchopleural fistula).^22,29 Patients who have complications may need drainage of abscesses or empyema along with a longer duration of antibiotic therapy until clinical and radiographic signs improve.

For community-acquired cases of aspiration pneumonia, a number of antibiotics have proven effective:

• Clindamycin (Cleocin) is still the agent most commonly used, although it lacks gram-negative bacterial coverage.
• Beta-lactam penicillins and newer quinolones have been used successfully.^2,29–31 In addition to covering the previously mentioned bacteria, these antibiotics have the added benefit of covering anaerobic bacteria.
• Metronidazole (Flagyl) should not be used alone because it has a higher clinical failure rate.^32,33

For nosocomial aspiration pneumonia, giving a broad-spectrum antibiotic empirically is warranted. Beta-lactam penicillins with extended gram-negative coverage, carbapenems, or monobactams in combination with an anti-staphylococcal drug have been advocated for nosocomial aspiration.^2,22 A strategy of broad-spectrum coverage followed by narrowing or de-escalating coverage according to lower respiratory tract cultures is encouraged.^21,22,34

SECONDARY BACTERIAL INFECTION OF CHEMICAL PNEUMONITIS

Nearly 25% of patients with chemical pneumonitis improve initially, then show clinical deterioration secondary to superimposed bacterial infection.¹³ Chest radiographs show worsening of initial infiltrates or the development of new ones. The causative organisms and treatment depend on whether the superimposed infection is community-acquired or nosocomial, as is the case in primary bacterial aspiration pneumonia.

PREVENTING ASPIRATION

Measures should be taken to prevent aspiration pneumonia and chemical pneumonitis, especially in institutionalized patients at high risk.¹²

Elevation of the head of the bed while feeding, dental prophylaxis, and good oral hygiene are known to reduce the incidence of these problems.^35–37

A swallowing evaluation for patients with dysphagia can identify those at higher risk of aspiration. These patients may be candidates for postural adjustments, diet modification, strengthening, and other measures offered by the speech and language pathology teams to improve swallowing physiology, biomechanics, safety, and endurance.^2,35

Although percutaneous endoscopic gastrostomy tubes are often placed in patients who have aspirated or who are at high risk of aspiration, they do not protect against aspiration, nor do orogastric or nasogastric tubes.³⁸

To date, we have no evidence that prophylactic antibiotic therapy prevents bacterial aspiration pneumonia. In addition, this practice encourages the development of resistant organisms.^19,39,40

Antibiotics are indicated for primary bacterial aspiration pneumonia and secondary bacterial infection of aspiration (chemical) pneumonitis, but not for uncomplicated chemical pneumonitis.

THREE TYPES OF ‘ASPIRATION PNEUMONIA’

Aspiration pneumonia is a broad and vague term mainly used to refer to the pulmonary consequences of abnormal entry of exogenous or endogenous substances into the lower airways. It can be classified as:

• Aspiration (chemical) pneumonitis
• Primary bacterial aspiration pneumonia
• Secondary bacterial infection of chemical pneumonitis.

These three are sometimes difficult to differentiate, as their signs and symptoms can overlap.

CHEMICAL PNEUMONITIS

Aspiration of stomach contents is relatively common, even in healthy people, and usually has no clinical consequences.¹ However, it has also been closely related to community-acquired and nosocomial pneumonia in some studies.^2,3

Chemical pneumonitis is usually a consequence of the aspiration of a large volume (≥ 4 mL/kg) of sterile acidic (pH < 2.5) gastric contents into the lower airways (Mendelson syndrome).^4,5 The clinical picture varies from asymptomatic to signs of severe dyspnea, hypoxia, cough, and low-grade fever; these signs and symptoms may develop rapidly, within minutes to hours after a witnessed or suspected episode of aspiration.^2,6,7 However, they represent an inflammatory reaction to the gastric acid rather than a reaction to bacterial infection.^8–10

Chemical pneumonitis affects the most dependent regions of the lungs

Chest radiography shows infiltrates in the most dependent regions of the lung. If aspiration occurs while the patient is supine, the posterior segments of the upper lobes and the apical segments of the lower lobes are most affected. The basal segments of the lower lobes are usually affected if aspiration occurs while the patient is standing or upright.^1,2,11,12

Clinical course varies

The clinical course varies. In almost 60% of cases, the patient’s condition improves and the lung infiltrates resolve rapidly, within 2 to 4 days. On the other hand, in about 15% of cases, the patient’s condition deteriorates quickly, within 24 to 36 hours, and progresses to hypoxic respiratory failure and acute respiratory distress syndrome.

In the other 25% of cases, the patient’s condition may improve initially but then worsen as a secondary bacterial infection sets in. The death rate in these patients is almost three times higher than the rate in patients with uncomplicated chemical pneumonitis.^11,13

Treatment of uncomplicated cases is mainly supportive

The treatment of uncomplicated chemical pneumonitis involves supportive measures such as airway clearance, oxygen supplementation, and positive pressure ventilation if needed. An obstructing foreign body may need to be removed.^12,14 Corticosteroids have been tried, without success.^11–13,15

Empiric antibiotic treatment is controversial

Chemical pneumonitis can be difficult to differentiate from bacterial aspiration pneumonia, and whether to give antibiotics is controversial. ¹⁶ A survey of current practices among intensivists showed that antimicrobial therapy was often given empirically for noninfectious chemical pneumonitis.¹⁷ This practice raises concerns of higher treatment costs and antibiotic resistance.^16,18,19 Additionally, antibiotics do not seem to alter the clinical outcome, including radiographic resolution, duration of hospitalization, or death rate, nor do they influence the subsequent development of infection.^1,11,13,20

In cases of witnessed or strongly suspected aspiration of gastric contents, antibiotics are not warranted since bacterial infection is not likely to be the cause of any signs or symptoms. ^2,7,16 However, to detect secondary infection early, the patient’s respiratory status should be monitored carefully and chest radiography should be repeated.

In less clear-cut cases, ie, if it is not clear whether the patient actually has chemical pneumonitis or primary bacterial aspiration pneumonia, it is prudent to start antibiotics empirically after obtaining lower-respiratory-tract secretions for stains and cultures, and then to reassess within 48 to 72 hours. The antibiotics can be discontinued if the patient has rapid clinical and radiographic improvement and negative cultures. Those whose condition does not improve or who have positive cultures should receive a full course of antibiotics.^21,22

PRIMARY BACTERIAL ASPIRATION PNEUMONIA

Primary bacterial aspiration pneumonia—ie, caused by bacteria residing in the upper airways and stomach gaining access to lower airways through aspiration in small or large amounts—is the most common form of aspiration pneumonia, although the actual episode of aspiration is seldom observed.

Signs of bacterial pneumonia

Primary bacterial aspiration pneumonia bears the hallmarks of bacterial pneumonia.¹² The clinical picture is more indolent than chemical pneumonitis and includes cough, fever, and putrid sputum, mainly in patients who have clinical conditions predisposing to aspiration (eg, coma, stroke, alcoholism, poor dentition, tube feedings).^1,12,20

The characteristic signs on chest radiography are infiltrates involving mainly the lung bases (the right more then the left). If untreated or inadequately treated, complications such as lung abscess, empyema, bronchiectasis, and broncopleural fistula are common.²³

Are aerobic organisms replacing anaerobic ones in the community?

The causative organisms in community-acquired aspiration pneumonia are still debated despite abundant research. Older studies^1,24,25 found mostly anaerobic organisms (pepto-streptococci, peptococci, Fusobacterium, Prevotela, Bacteroides) as the underlying pathogens, whereas more recent studies^16,26,27 found mostly aerobic organisms (Streptococcus pneumoniae, Haemophilus influenzae, Staphylococcus aureus, Enterobacteriaceae) and failed to recover anaerobic organisms. These discrepancies may be the result of different techniques used to isolate organisms: older studies used transtracheal sampling, and transtracheal aspirates may be easily contaminated or colonized by oropharyngeal flora; more recent studies used protected specimen brushes to collect lower-airway specimens.²

In addition, the pathogenic organisms that predominate in community-acquired aspiration pneumonia, as listed above, are different from those most often found in nosocomial cases; gram-negative bacilli (Pseudomonas aeruginosa, Klebsiella pneumoniae, Escherichia coli) are most often isolated in patients with aspiration pneumonia acquired in hospitals and nursing homes.^16,27,28S aureus also is an important causative organism in nosocomial cases.^16,28

Knowing the causative organisms in bacterial aspiration pneumonia is important for guiding antimicrobial therapy.

Antibiotics are required for bacterial aspiration pneumonia

A course of antibiotics is required for bacterial aspiration pneumonia. While there are no definitive recommendations for the duration of treatment, 7 to 8 days is probably appropriate in uncomplicated cases (ie, no lung abscess, empyema, bronchopleural fistula).^22,29 Patients who have complications may need drainage of abscesses or empyema along with a longer duration of antibiotic therapy until clinical and radiographic signs improve.

For community-acquired cases of aspiration pneumonia, a number of antibiotics have proven effective:

• Clindamycin (Cleocin) is still the agent most commonly used, although it lacks gram-negative bacterial coverage.
• Beta-lactam penicillins and newer quinolones have been used successfully.^2,29–31 In addition to covering the previously mentioned bacteria, these antibiotics have the added benefit of covering anaerobic bacteria.
• Metronidazole (Flagyl) should not be used alone because it has a higher clinical failure rate.^32,33

For nosocomial aspiration pneumonia, giving a broad-spectrum antibiotic empirically is warranted. Beta-lactam penicillins with extended gram-negative coverage, carbapenems, or monobactams in combination with an anti-staphylococcal drug have been advocated for nosocomial aspiration.^2,22 A strategy of broad-spectrum coverage followed by narrowing or de-escalating coverage according to lower respiratory tract cultures is encouraged.^21,22,34

SECONDARY BACTERIAL INFECTION OF CHEMICAL PNEUMONITIS

Nearly 25% of patients with chemical pneumonitis improve initially, then show clinical deterioration secondary to superimposed bacterial infection.¹³ Chest radiographs show worsening of initial infiltrates or the development of new ones. The causative organisms and treatment depend on whether the superimposed infection is community-acquired or nosocomial, as is the case in primary bacterial aspiration pneumonia.

PREVENTING ASPIRATION

Measures should be taken to prevent aspiration pneumonia and chemical pneumonitis, especially in institutionalized patients at high risk.¹²

Elevation of the head of the bed while feeding, dental prophylaxis, and good oral hygiene are known to reduce the incidence of these problems.^35–37

A swallowing evaluation for patients with dysphagia can identify those at higher risk of aspiration. These patients may be candidates for postural adjustments, diet modification, strengthening, and other measures offered by the speech and language pathology teams to improve swallowing physiology, biomechanics, safety, and endurance.^2,35

Although percutaneous endoscopic gastrostomy tubes are often placed in patients who have aspirated or who are at high risk of aspiration, they do not protect against aspiration, nor do orogastric or nasogastric tubes.³⁸

To date, we have no evidence that prophylactic antibiotic therapy prevents bacterial aspiration pneumonia. In addition, this practice encourages the development of resistant organisms.^19,39,40

Antibiotics are indicated for primary bacterial aspiration pneumonia and secondary bacterial infection of aspiration (chemical) pneumonitis, but not for uncomplicated chemical pneumonitis.

THREE TYPES OF ‘ASPIRATION PNEUMONIA’

Aspiration pneumonia is a broad and vague term mainly used to refer to the pulmonary consequences of abnormal entry of exogenous or endogenous substances into the lower airways. It can be classified as:

• Aspiration (chemical) pneumonitis
• Primary bacterial aspiration pneumonia
• Secondary bacterial infection of chemical pneumonitis.

These three are sometimes difficult to differentiate, as their signs and symptoms can overlap.

CHEMICAL PNEUMONITIS

Aspiration of stomach contents is relatively common, even in healthy people, and usually has no clinical consequences.¹ However, it has also been closely related to community-acquired and nosocomial pneumonia in some studies.^2,3

Chemical pneumonitis is usually a consequence of the aspiration of a large volume (≥ 4 mL/kg) of sterile acidic (pH < 2.5) gastric contents into the lower airways (Mendelson syndrome).^4,5 The clinical picture varies from asymptomatic to signs of severe dyspnea, hypoxia, cough, and low-grade fever; these signs and symptoms may develop rapidly, within minutes to hours after a witnessed or suspected episode of aspiration.^2,6,7 However, they represent an inflammatory reaction to the gastric acid rather than a reaction to bacterial infection.^8–10

Chemical pneumonitis affects the most dependent regions of the lungs

Chest radiography shows infiltrates in the most dependent regions of the lung. If aspiration occurs while the patient is supine, the posterior segments of the upper lobes and the apical segments of the lower lobes are most affected. The basal segments of the lower lobes are usually affected if aspiration occurs while the patient is standing or upright.^1,2,11,12

Clinical course varies

The clinical course varies. In almost 60% of cases, the patient’s condition improves and the lung infiltrates resolve rapidly, within 2 to 4 days. On the other hand, in about 15% of cases, the patient’s condition deteriorates quickly, within 24 to 36 hours, and progresses to hypoxic respiratory failure and acute respiratory distress syndrome.

In the other 25% of cases, the patient’s condition may improve initially but then worsen as a secondary bacterial infection sets in. The death rate in these patients is almost three times higher than the rate in patients with uncomplicated chemical pneumonitis.^11,13

Treatment of uncomplicated cases is mainly supportive

The treatment of uncomplicated chemical pneumonitis involves supportive measures such as airway clearance, oxygen supplementation, and positive pressure ventilation if needed. An obstructing foreign body may need to be removed.^12,14 Corticosteroids have been tried, without success.^11–13,15

Empiric antibiotic treatment is controversial

Chemical pneumonitis can be difficult to differentiate from bacterial aspiration pneumonia, and whether to give antibiotics is controversial. ¹⁶ A survey of current practices among intensivists showed that antimicrobial therapy was often given empirically for noninfectious chemical pneumonitis.¹⁷ This practice raises concerns of higher treatment costs and antibiotic resistance.^16,18,19 Additionally, antibiotics do not seem to alter the clinical outcome, including radiographic resolution, duration of hospitalization, or death rate, nor do they influence the subsequent development of infection.^1,11,13,20

In cases of witnessed or strongly suspected aspiration of gastric contents, antibiotics are not warranted since bacterial infection is not likely to be the cause of any signs or symptoms. ^2,7,16 However, to detect secondary infection early, the patient’s respiratory status should be monitored carefully and chest radiography should be repeated.

In less clear-cut cases, ie, if it is not clear whether the patient actually has chemical pneumonitis or primary bacterial aspiration pneumonia, it is prudent to start antibiotics empirically after obtaining lower-respiratory-tract secretions for stains and cultures, and then to reassess within 48 to 72 hours. The antibiotics can be discontinued if the patient has rapid clinical and radiographic improvement and negative cultures. Those whose condition does not improve or who have positive cultures should receive a full course of antibiotics.^21,22

PRIMARY BACTERIAL ASPIRATION PNEUMONIA

Primary bacterial aspiration pneumonia—ie, caused by bacteria residing in the upper airways and stomach gaining access to lower airways through aspiration in small or large amounts—is the most common form of aspiration pneumonia, although the actual episode of aspiration is seldom observed.

Signs of bacterial pneumonia

Primary bacterial aspiration pneumonia bears the hallmarks of bacterial pneumonia.¹² The clinical picture is more indolent than chemical pneumonitis and includes cough, fever, and putrid sputum, mainly in patients who have clinical conditions predisposing to aspiration (eg, coma, stroke, alcoholism, poor dentition, tube feedings).^1,12,20

The characteristic signs on chest radiography are infiltrates involving mainly the lung bases (the right more then the left). If untreated or inadequately treated, complications such as lung abscess, empyema, bronchiectasis, and broncopleural fistula are common.²³

Are aerobic organisms replacing anaerobic ones in the community?

The causative organisms in community-acquired aspiration pneumonia are still debated despite abundant research. Older studies^1,24,25 found mostly anaerobic organisms (pepto-streptococci, peptococci, Fusobacterium, Prevotela, Bacteroides) as the underlying pathogens, whereas more recent studies^16,26,27 found mostly aerobic organisms (Streptococcus pneumoniae, Haemophilus influenzae, Staphylococcus aureus, Enterobacteriaceae) and failed to recover anaerobic organisms. These discrepancies may be the result of different techniques used to isolate organisms: older studies used transtracheal sampling, and transtracheal aspirates may be easily contaminated or colonized by oropharyngeal flora; more recent studies used protected specimen brushes to collect lower-airway specimens.²

In addition, the pathogenic organisms that predominate in community-acquired aspiration pneumonia, as listed above, are different from those most often found in nosocomial cases; gram-negative bacilli (Pseudomonas aeruginosa, Klebsiella pneumoniae, Escherichia coli) are most often isolated in patients with aspiration pneumonia acquired in hospitals and nursing homes.^16,27,28S aureus also is an important causative organism in nosocomial cases.^16,28

Knowing the causative organisms in bacterial aspiration pneumonia is important for guiding antimicrobial therapy.

Antibiotics are required for bacterial aspiration pneumonia

A course of antibiotics is required for bacterial aspiration pneumonia. While there are no definitive recommendations for the duration of treatment, 7 to 8 days is probably appropriate in uncomplicated cases (ie, no lung abscess, empyema, bronchopleural fistula).^22,29 Patients who have complications may need drainage of abscesses or empyema along with a longer duration of antibiotic therapy until clinical and radiographic signs improve.

For community-acquired cases of aspiration pneumonia, a number of antibiotics have proven effective:

• Clindamycin (Cleocin) is still the agent most commonly used, although it lacks gram-negative bacterial coverage.
• Beta-lactam penicillins and newer quinolones have been used successfully.^2,29–31 In addition to covering the previously mentioned bacteria, these antibiotics have the added benefit of covering anaerobic bacteria.
• Metronidazole (Flagyl) should not be used alone because it has a higher clinical failure rate.^32,33

For nosocomial aspiration pneumonia, giving a broad-spectrum antibiotic empirically is warranted. Beta-lactam penicillins with extended gram-negative coverage, carbapenems, or monobactams in combination with an anti-staphylococcal drug have been advocated for nosocomial aspiration.^2,22 A strategy of broad-spectrum coverage followed by narrowing or de-escalating coverage according to lower respiratory tract cultures is encouraged.^21,22,34

SECONDARY BACTERIAL INFECTION OF CHEMICAL PNEUMONITIS

Nearly 25% of patients with chemical pneumonitis improve initially, then show clinical deterioration secondary to superimposed bacterial infection.¹³ Chest radiographs show worsening of initial infiltrates or the development of new ones. The causative organisms and treatment depend on whether the superimposed infection is community-acquired or nosocomial, as is the case in primary bacterial aspiration pneumonia.

PREVENTING ASPIRATION

Measures should be taken to prevent aspiration pneumonia and chemical pneumonitis, especially in institutionalized patients at high risk.¹²

Elevation of the head of the bed while feeding, dental prophylaxis, and good oral hygiene are known to reduce the incidence of these problems.^35–37

A swallowing evaluation for patients with dysphagia can identify those at higher risk of aspiration. These patients may be candidates for postural adjustments, diet modification, strengthening, and other measures offered by the speech and language pathology teams to improve swallowing physiology, biomechanics, safety, and endurance.^2,35

Although percutaneous endoscopic gastrostomy tubes are often placed in patients who have aspirated or who are at high risk of aspiration, they do not protect against aspiration, nor do orogastric or nasogastric tubes.³⁸

To date, we have no evidence that prophylactic antibiotic therapy prevents bacterial aspiration pneumonia. In addition, this practice encourages the development of resistant organisms.^19,39,40

Reperfusion therapy decreases morbidity and mortality rates in patients with ST-segment elevation myocardial infarction (MI). Primary percutaneous coronary intervention (PCI) is preferred over fibrinolytic therapy as a reperfusion strategy when the delay in the time to treatment is short and the patient presents to a high-volume, well-equipped center with expert interventional cardiologists.

See related article

Compared with fibrinolytic therapy in randomized clinical trials, primary PCI produces higher rates of infarct artery patency, complete reperfusion (grade 3 by the criteria of the Thrombolysis in Myocardial Infarction [TIMI] study), and access-site bleeding. It also produces lower rates of recurrent ischemia, reinfarction, emergency repeat revascularization procedures, intracranial hemorrhage, and death.¹ If performed early and successfully, primary PCI also greatly decreases the rates of complications of ST-elevation MI that result from longer ischemic times or unsuccessful fibrinolytic therapy, allowing earlier hospital discharge and resumption of daily activities. Primary PCI is also the best reperfusion option in patients who present late after the onset of symptoms and in patients with cardiogenic shock, and it is the only option in patients who have contraindications to fibrinolytic therapy because of bleeding risk.

However, most hospitals do not have PCI capability. Two options at these hospitals are to transfer the patient to a PCI center quickly for primary PCI or to keep the patient on site and give fibrinolytic therapy, with its limitations. Earlier trials suggested that the transfer strategy was superior, but they had limitations: most patients received streptokinase, an inferior fibrinolytic agent, and door-to-door-to-balloon times were rapid, averaging only 95 minutes because of excellent logistical protocols and careful patient selection.² Most importantly, rescue PCI and routine PCI were seldom performed in patients receiving fibrinolytics, so fibrinolytic therapy was being tested as monotherapy.

In the real world, however, treatment delays are much longer, and fibrinolytic therapy has evolved into a strategy that includes crossover to rescue PCI or routine PCI. Therefore, the initial trials of transfer for primary PCI do not reflect current practice. In fact, recent registry data suggest that prehospital fibrinolytic therapy followed by early angiography is superior to primary PCI because most patients can be treated within 2 hours of symptom onset; they also suggest that on-site fibrinolytic therapy followed by early angiography is equal in efficacy to primary PCI as long as rescue PCI and routine PCI can be performed.^3,4

The most important modifiable predictor of outcome in ST-elevation MI is the time to treatment, a biological truth that continues to be supported by clinical evidence despite ideologic arguments by some interventional cardiology enthusiasts who claim that primary PCI is always superior to the fibrinolytic strategy, regardless of delays.

SURPRISINGLY, OUTCOMES WERE WORSE WITH FACILITATED PCI

It made sense, then, to conclude that the perfect strategy for hospitals without PCI capability would be a combined strategy of immediate fibrinolytic therapy to decrease the time delay associated with organizing PCI, and rapid transfer for immediate PCI to improve the limited reperfusion rates associated with fibrinolytic therapy.

Surprisingly, though, randomized trials found worse outcomes with this “facilitated PCI” strategy.⁵

Again, limitations in trial design might explain the lack of benefit in the trials. Inadequate anticoagulant and antiplatelet therapy were given to the fibrinolytic patients, and primary PCI patients had relatively short treatment delays, with many patients enrolled at hospitals with PCI capability.

PROGRESS IN REPERFUSION THERAPY

Great strides have been made in reperfusion therapy in recent years. Adjunctive therapy with clopidogrel (Plavix) and enoxaparin (Lovenox) has been shown to improve outcomes with fibrinolytic therapy. Bivalirudin (Angiomax) and stents have improved primary PCI’s performance. Reducing bleeding complications has become a clinical priority, with increasing emphasis on adjusting some drug doses according to renal function and using the radial artery for cardiac catheterization.

The American College of Cardiology initiative, “Door-to-Balloon (D2B): An Alliance for Quality,” focused much attention on organizing in-hospital systems of care for primary PCI, thus increasing the national rate of achieving a door-to-balloon time within 90 minutes from 50% to over 75% in patients who presented to hospitals with PCI capability.⁶

The American Heart Association has launched “Mission: Lifeline,” a national campaign to organize prehospital systems of care with their program,⁷ working within communities to address their unique needs, resources, and barriers to implementing systems of care for ST-elevation MI. The key aspect of this effort is to help geographic regions develop local solutions, an explicit recognition that there is no one-size-fits-all solution. Early triage by emergency medical services, rapid diagnosis with prehospital electrocardiography, destination and interhospital transfer protocols, and prehospital activation of the cardiac catheterization laboratory can greatly streamline emergency care and decrease treatment delays for primary PCI.

FOR OUTLYING HOSPITALS, A PHARMACOINVASIVE STRATEGY

So what about hospitals without PCI capability that cannot routinely transfer patients to a hospital with PCI capability within 90 minutes?

Lessons learned from the experiences with immediate PCI, rescue PCI, and facilitated PCI have evolved into the “pharmacoinvasive strategy.” Patients with ST-elevation MI are treated as rapidly as possible with a bolus of a fibrinolytic drug, eg, tenecteplase (TNKase) or reteplase (Retavase), and are also given aspirin, clopidogrel, and enoxaparin. Then, they are rapidly transferred to a PCI-capable hospital so that emergency PCI can be performed if reperfusion is not clinically apparent or if the patient develops pulmonary edema or cardiogenic shock. If the clinical signs suggest that reperfusion has been achieved (relief of chest pain, rapid resolution of ST-segment elevation, bursts of accelerated idioventricular rhythm), coronary angiography (and PCI, if indicated) can be performed within 3 to 24 hours of fibrinolytic therapy. This time frame allows the initial fibrinolytic effect to dissipate, while the antiplatelet and anticoagulant drugs achieve therapeutic levels.

Today, the goal is to treat every patient with the best reperfusion strategy available, given the limitations in resources and the geographic location of some centers, and to maximize the possibility of sustained patency of the infarct-related artery by implanting a stent, even if it takes several hours and transfer to another hospital to perform PCI.⁸ The pharmacoinvasive strategy of rapid administration of fibrinolytic therapy followed by PCI within 24 hours would be practical in most hospitals without PCI capability where treatment delays prohibit performance of primary PCI within 90 minutes of first medical contact.⁹

THE ‘STREAM’ TRIAL IS UNDER WAY

As proof of concept, the Strategic Reperfusion Early After Myocardial Infarction (STREAM) trial is enrolling 2,000 patients with ST-elevation MI presenting within 3 hours of symptom onset if primary PCI is not feasible within 60 minutes of first medical contact.¹⁰ Patients will be randomized to either of the following:

• Receive prehospital therapy with tenecteplase, aspirin, clopidogrel, and enoxaparin and undergo cardiac catheterization in 6 to 24 hours (or rescue PCI if reperfusion fails within 90 minutes of fibrinolysis)
• Undergo primary PCI performed according to local guidelines.

The primary measure of efficacy will be the composite rate of death, cardiogenic shock, heart failure, and reinfarction at 30 days. Measures of safety include the rates of ischemic stroke, intracranial hemorrhage, and major nonintracranial bleeding.

Reperfusion therapy decreases morbidity and mortality rates in patients with ST-segment elevation myocardial infarction (MI). Primary percutaneous coronary intervention (PCI) is preferred over fibrinolytic therapy as a reperfusion strategy when the delay in the time to treatment is short and the patient presents to a high-volume, well-equipped center with expert interventional cardiologists.

See related article

Compared with fibrinolytic therapy in randomized clinical trials, primary PCI produces higher rates of infarct artery patency, complete reperfusion (grade 3 by the criteria of the Thrombolysis in Myocardial Infarction [TIMI] study), and access-site bleeding. It also produces lower rates of recurrent ischemia, reinfarction, emergency repeat revascularization procedures, intracranial hemorrhage, and death.¹ If performed early and successfully, primary PCI also greatly decreases the rates of complications of ST-elevation MI that result from longer ischemic times or unsuccessful fibrinolytic therapy, allowing earlier hospital discharge and resumption of daily activities. Primary PCI is also the best reperfusion option in patients who present late after the onset of symptoms and in patients with cardiogenic shock, and it is the only option in patients who have contraindications to fibrinolytic therapy because of bleeding risk.

However, most hospitals do not have PCI capability. Two options at these hospitals are to transfer the patient to a PCI center quickly for primary PCI or to keep the patient on site and give fibrinolytic therapy, with its limitations. Earlier trials suggested that the transfer strategy was superior, but they had limitations: most patients received streptokinase, an inferior fibrinolytic agent, and door-to-door-to-balloon times were rapid, averaging only 95 minutes because of excellent logistical protocols and careful patient selection.² Most importantly, rescue PCI and routine PCI were seldom performed in patients receiving fibrinolytics, so fibrinolytic therapy was being tested as monotherapy.

In the real world, however, treatment delays are much longer, and fibrinolytic therapy has evolved into a strategy that includes crossover to rescue PCI or routine PCI. Therefore, the initial trials of transfer for primary PCI do not reflect current practice. In fact, recent registry data suggest that prehospital fibrinolytic therapy followed by early angiography is superior to primary PCI because most patients can be treated within 2 hours of symptom onset; they also suggest that on-site fibrinolytic therapy followed by early angiography is equal in efficacy to primary PCI as long as rescue PCI and routine PCI can be performed.^3,4

The most important modifiable predictor of outcome in ST-elevation MI is the time to treatment, a biological truth that continues to be supported by clinical evidence despite ideologic arguments by some interventional cardiology enthusiasts who claim that primary PCI is always superior to the fibrinolytic strategy, regardless of delays.

SURPRISINGLY, OUTCOMES WERE WORSE WITH FACILITATED PCI

It made sense, then, to conclude that the perfect strategy for hospitals without PCI capability would be a combined strategy of immediate fibrinolytic therapy to decrease the time delay associated with organizing PCI, and rapid transfer for immediate PCI to improve the limited reperfusion rates associated with fibrinolytic therapy.

Surprisingly, though, randomized trials found worse outcomes with this “facilitated PCI” strategy.⁵

Again, limitations in trial design might explain the lack of benefit in the trials. Inadequate anticoagulant and antiplatelet therapy were given to the fibrinolytic patients, and primary PCI patients had relatively short treatment delays, with many patients enrolled at hospitals with PCI capability.

PROGRESS IN REPERFUSION THERAPY

Great strides have been made in reperfusion therapy in recent years. Adjunctive therapy with clopidogrel (Plavix) and enoxaparin (Lovenox) has been shown to improve outcomes with fibrinolytic therapy. Bivalirudin (Angiomax) and stents have improved primary PCI’s performance. Reducing bleeding complications has become a clinical priority, with increasing emphasis on adjusting some drug doses according to renal function and using the radial artery for cardiac catheterization.

The American College of Cardiology initiative, “Door-to-Balloon (D2B): An Alliance for Quality,” focused much attention on organizing in-hospital systems of care for primary PCI, thus increasing the national rate of achieving a door-to-balloon time within 90 minutes from 50% to over 75% in patients who presented to hospitals with PCI capability.⁶

The American Heart Association has launched “Mission: Lifeline,” a national campaign to organize prehospital systems of care with their program,⁷ working within communities to address their unique needs, resources, and barriers to implementing systems of care for ST-elevation MI. The key aspect of this effort is to help geographic regions develop local solutions, an explicit recognition that there is no one-size-fits-all solution. Early triage by emergency medical services, rapid diagnosis with prehospital electrocardiography, destination and interhospital transfer protocols, and prehospital activation of the cardiac catheterization laboratory can greatly streamline emergency care and decrease treatment delays for primary PCI.

FOR OUTLYING HOSPITALS, A PHARMACOINVASIVE STRATEGY

So what about hospitals without PCI capability that cannot routinely transfer patients to a hospital with PCI capability within 90 minutes?

Lessons learned from the experiences with immediate PCI, rescue PCI, and facilitated PCI have evolved into the “pharmacoinvasive strategy.” Patients with ST-elevation MI are treated as rapidly as possible with a bolus of a fibrinolytic drug, eg, tenecteplase (TNKase) or reteplase (Retavase), and are also given aspirin, clopidogrel, and enoxaparin. Then, they are rapidly transferred to a PCI-capable hospital so that emergency PCI can be performed if reperfusion is not clinically apparent or if the patient develops pulmonary edema or cardiogenic shock. If the clinical signs suggest that reperfusion has been achieved (relief of chest pain, rapid resolution of ST-segment elevation, bursts of accelerated idioventricular rhythm), coronary angiography (and PCI, if indicated) can be performed within 3 to 24 hours of fibrinolytic therapy. This time frame allows the initial fibrinolytic effect to dissipate, while the antiplatelet and anticoagulant drugs achieve therapeutic levels.

Today, the goal is to treat every patient with the best reperfusion strategy available, given the limitations in resources and the geographic location of some centers, and to maximize the possibility of sustained patency of the infarct-related artery by implanting a stent, even if it takes several hours and transfer to another hospital to perform PCI.⁸ The pharmacoinvasive strategy of rapid administration of fibrinolytic therapy followed by PCI within 24 hours would be practical in most hospitals without PCI capability where treatment delays prohibit performance of primary PCI within 90 minutes of first medical contact.⁹

THE ‘STREAM’ TRIAL IS UNDER WAY

As proof of concept, the Strategic Reperfusion Early After Myocardial Infarction (STREAM) trial is enrolling 2,000 patients with ST-elevation MI presenting within 3 hours of symptom onset if primary PCI is not feasible within 60 minutes of first medical contact.¹⁰ Patients will be randomized to either of the following:

• Receive prehospital therapy with tenecteplase, aspirin, clopidogrel, and enoxaparin and undergo cardiac catheterization in 6 to 24 hours (or rescue PCI if reperfusion fails within 90 minutes of fibrinolysis)
• Undergo primary PCI performed according to local guidelines.

The primary measure of efficacy will be the composite rate of death, cardiogenic shock, heart failure, and reinfarction at 30 days. Measures of safety include the rates of ischemic stroke, intracranial hemorrhage, and major nonintracranial bleeding.

Reperfusion therapy decreases morbidity and mortality rates in patients with ST-segment elevation myocardial infarction (MI). Primary percutaneous coronary intervention (PCI) is preferred over fibrinolytic therapy as a reperfusion strategy when the delay in the time to treatment is short and the patient presents to a high-volume, well-equipped center with expert interventional cardiologists.

See related article

Compared with fibrinolytic therapy in randomized clinical trials, primary PCI produces higher rates of infarct artery patency, complete reperfusion (grade 3 by the criteria of the Thrombolysis in Myocardial Infarction [TIMI] study), and access-site bleeding. It also produces lower rates of recurrent ischemia, reinfarction, emergency repeat revascularization procedures, intracranial hemorrhage, and death.¹ If performed early and successfully, primary PCI also greatly decreases the rates of complications of ST-elevation MI that result from longer ischemic times or unsuccessful fibrinolytic therapy, allowing earlier hospital discharge and resumption of daily activities. Primary PCI is also the best reperfusion option in patients who present late after the onset of symptoms and in patients with cardiogenic shock, and it is the only option in patients who have contraindications to fibrinolytic therapy because of bleeding risk.

However, most hospitals do not have PCI capability. Two options at these hospitals are to transfer the patient to a PCI center quickly for primary PCI or to keep the patient on site and give fibrinolytic therapy, with its limitations. Earlier trials suggested that the transfer strategy was superior, but they had limitations: most patients received streptokinase, an inferior fibrinolytic agent, and door-to-door-to-balloon times were rapid, averaging only 95 minutes because of excellent logistical protocols and careful patient selection.² Most importantly, rescue PCI and routine PCI were seldom performed in patients receiving fibrinolytics, so fibrinolytic therapy was being tested as monotherapy.

In the real world, however, treatment delays are much longer, and fibrinolytic therapy has evolved into a strategy that includes crossover to rescue PCI or routine PCI. Therefore, the initial trials of transfer for primary PCI do not reflect current practice. In fact, recent registry data suggest that prehospital fibrinolytic therapy followed by early angiography is superior to primary PCI because most patients can be treated within 2 hours of symptom onset; they also suggest that on-site fibrinolytic therapy followed by early angiography is equal in efficacy to primary PCI as long as rescue PCI and routine PCI can be performed.^3,4

The most important modifiable predictor of outcome in ST-elevation MI is the time to treatment, a biological truth that continues to be supported by clinical evidence despite ideologic arguments by some interventional cardiology enthusiasts who claim that primary PCI is always superior to the fibrinolytic strategy, regardless of delays.

SURPRISINGLY, OUTCOMES WERE WORSE WITH FACILITATED PCI

It made sense, then, to conclude that the perfect strategy for hospitals without PCI capability would be a combined strategy of immediate fibrinolytic therapy to decrease the time delay associated with organizing PCI, and rapid transfer for immediate PCI to improve the limited reperfusion rates associated with fibrinolytic therapy.

Surprisingly, though, randomized trials found worse outcomes with this “facilitated PCI” strategy.⁵

Again, limitations in trial design might explain the lack of benefit in the trials. Inadequate anticoagulant and antiplatelet therapy were given to the fibrinolytic patients, and primary PCI patients had relatively short treatment delays, with many patients enrolled at hospitals with PCI capability.

PROGRESS IN REPERFUSION THERAPY

Great strides have been made in reperfusion therapy in recent years. Adjunctive therapy with clopidogrel (Plavix) and enoxaparin (Lovenox) has been shown to improve outcomes with fibrinolytic therapy. Bivalirudin (Angiomax) and stents have improved primary PCI’s performance. Reducing bleeding complications has become a clinical priority, with increasing emphasis on adjusting some drug doses according to renal function and using the radial artery for cardiac catheterization.

The American College of Cardiology initiative, “Door-to-Balloon (D2B): An Alliance for Quality,” focused much attention on organizing in-hospital systems of care for primary PCI, thus increasing the national rate of achieving a door-to-balloon time within 90 minutes from 50% to over 75% in patients who presented to hospitals with PCI capability.⁶

The American Heart Association has launched “Mission: Lifeline,” a national campaign to organize prehospital systems of care with their program,⁷ working within communities to address their unique needs, resources, and barriers to implementing systems of care for ST-elevation MI. The key aspect of this effort is to help geographic regions develop local solutions, an explicit recognition that there is no one-size-fits-all solution. Early triage by emergency medical services, rapid diagnosis with prehospital electrocardiography, destination and interhospital transfer protocols, and prehospital activation of the cardiac catheterization laboratory can greatly streamline emergency care and decrease treatment delays for primary PCI.

FOR OUTLYING HOSPITALS, A PHARMACOINVASIVE STRATEGY

So what about hospitals without PCI capability that cannot routinely transfer patients to a hospital with PCI capability within 90 minutes?

Lessons learned from the experiences with immediate PCI, rescue PCI, and facilitated PCI have evolved into the “pharmacoinvasive strategy.” Patients with ST-elevation MI are treated as rapidly as possible with a bolus of a fibrinolytic drug, eg, tenecteplase (TNKase) or reteplase (Retavase), and are also given aspirin, clopidogrel, and enoxaparin. Then, they are rapidly transferred to a PCI-capable hospital so that emergency PCI can be performed if reperfusion is not clinically apparent or if the patient develops pulmonary edema or cardiogenic shock. If the clinical signs suggest that reperfusion has been achieved (relief of chest pain, rapid resolution of ST-segment elevation, bursts of accelerated idioventricular rhythm), coronary angiography (and PCI, if indicated) can be performed within 3 to 24 hours of fibrinolytic therapy. This time frame allows the initial fibrinolytic effect to dissipate, while the antiplatelet and anticoagulant drugs achieve therapeutic levels.

Today, the goal is to treat every patient with the best reperfusion strategy available, given the limitations in resources and the geographic location of some centers, and to maximize the possibility of sustained patency of the infarct-related artery by implanting a stent, even if it takes several hours and transfer to another hospital to perform PCI.⁸ The pharmacoinvasive strategy of rapid administration of fibrinolytic therapy followed by PCI within 24 hours would be practical in most hospitals without PCI capability where treatment delays prohibit performance of primary PCI within 90 minutes of first medical contact.⁹

THE ‘STREAM’ TRIAL IS UNDER WAY

As proof of concept, the Strategic Reperfusion Early After Myocardial Infarction (STREAM) trial is enrolling 2,000 patients with ST-elevation MI presenting within 3 hours of symptom onset if primary PCI is not feasible within 60 minutes of first medical contact.¹⁰ Patients will be randomized to either of the following:

• Receive prehospital therapy with tenecteplase, aspirin, clopidogrel, and enoxaparin and undergo cardiac catheterization in 6 to 24 hours (or rescue PCI if reperfusion fails within 90 minutes of fibrinolysis)
• Undergo primary PCI performed according to local guidelines.

The primary measure of efficacy will be the composite rate of death, cardiogenic shock, heart failure, and reinfarction at 30 days. Measures of safety include the rates of ischemic stroke, intracranial hemorrhage, and major nonintracranial bleeding.