Nathan Houchens, MD

Clinician educators face numerous obstacles to their joint mission of facilitating learning while also ensuring high-quality and patient-centered care. Time constraints, including the institution of house officer duty hour limitations,¹ shorter lengths of stay for hospitalized patients,² and competing career responsibilities, combine to create a dynamic learning environment. Additionally, clinician educators must balance the autonomy of their learners with the safety of their patients. They must teach to multiple learning levels and work collaboratively with multiple disciplines to foster an effective team-based approach to patient care. Yet, many clinician educators have no formal training in pedagogical methods.³ Such challenges necessitate increased attention to the work of excellent clinician educators and their respective teaching approaches.

Many studies of clinical teaching rely primarily on survey data of attributes of good clinical teachers.^3-7 While some studies have incorporated direct observations of teaching^8,9 or interviews with clinician educators or learners,^10,11 few have incorporated multiple perspectives from the current team and from former learners in order to provide a comprehensive picture of team-based learning.¹²

The goal of this study was to gain a thorough understanding, through multiple perspectives, of the techniques and behaviors used by exemplary educators within actual clinical environments. We studied attitudes, behaviors, and approaches of 12 such inpatient clinician educators.

Study Design and Sampling

This was a multisite study using an exploratory qualitative approach to inquiry. This approach was used to study the techniques and behaviors of excellent attendings during inpatient general medicine rounds. A modified snowball sampling approach¹³ was used, meaning individuals known to one member of the research team (SS) were initially contacted and asked to identify clinician educators (also referred to as attendings) for potential inclusion in the study. In an effort to identify attendings from a broad range of medical schools, the “2015 U.S. News and World Report Top Medical Schools: Research” rankings¹⁴ were also reviewed, with priority given to the top 25, as these are widely used to represent the best US hospitals. In an attempt to invite attendings from diverse institutions, additional medical schools not in the top 25 as well as historically black medical schools were also included. Division chiefs and chairs of internal medicine and/or directors of internal medicine residency programs at these schools were contacted and asked for recommendations of attendings, both within and outside their institutions, who they considered to be great inpatient teachers. In addition, key experts who have won teaching awards or were known to be specialists in the field of medical education were asked to nominate one or two other outstanding attendings.

By using this sampling method, 59 potential participants were identified. An internet search was conducted to obtain information about the potential participants and their institutions. Organizational characteristics such as geographic location, hospital size and affiliation, and patient population, as well as individual characteristics such as gender, medical education and training, and educational awards received were considered so that a diversity of organizations and backgrounds was represented. The list was narrowed down to 16 attendings who were contacted via e-mail and asked to participate. Interested participants were asked for a list of their current team members and 6 to 10 former learners to contact for interviews and focus groups. Former learners were included in an effort to better understand lasting effects on learners from their exemplary teaching attendings. A total of 12 attending physicians agreed to participate (Table 1). Literature on field methods has shown that 12 interviews are found to be adequate in accomplishing data saturation.¹⁵ Although 2 attendings were located at the same institution, we decided to include them given that both are recognized as master clinician educators and were each recommended by several individuals from various institutions. Hospitals were located throughout the US and included both university-affiliated hospitals and Veterans Affairs medical centers. Despite efforts to include physicians from historically black colleges and universities, only one attending was identified, and they declined the request to participate.

Data Collection

Observations. The one-day site visits were mainly conducted by two research team members, a physician (SS) and a medical anthropologist (MH), both of whom have extensive experience in qualitative methods. Teams were not uniform but were generally comprised of 1 attending, 1 senior medical resident, 1 to 2 interns, and approximately 2 medical students. Occasionally, a pharmacist, clinical assistant, or other health professional accompanied the team on rounds. Not infrequently, the bedside nurse would explicitly be included in the discussion regarding his or her specific patient. Each site visit began with observing attendings (N = 12) and current learners (N = 57) during rounds. Each research team member recorded their own observations via handwritten field notes, paying particular attention to group interactions, teaching approach, conversations occurring within and peripheral to the team, patient-team interactions, and the physical environment. By standing outside of the medical team circle and remaining silent during rounds, research team members remained unobtrusive to the discussion and process of rounds. Materials the attendings used during their teaching rounds were also documented and collected. Rounds generally lasted 2 to 3 hours. After each site visit, the research team met to compare and combine field notes.

Interviews and Focus Groups. The research team then conducted individual, semi-structured interviews with the attendings, focus groups with their current team (N = 46), and interviews or focus groups with their former learners (N = 26; Supplement 1). Eleven of the current team members observed during rounds were unable to participate in the focus groups due to clinical duties. Because the current learners who participated in the focus groups were also observed during rounds, the research team was able to ask them open-ended questions regarding teaching rounds and their roles as learners within this environment. Former learners who were still at the hospital participated in separate focus groups or interviews. Former learners who were no longer present at the hospital were contacted by telephone and individually interviewed by one research team member (MH). All interviews and focus groups were audio-recorded and transcribed.

This study was determined to be exempt by the University of Michigan Institutional Review Board. All participants were informed that their participation was completely voluntary and that they could terminate their involvement at any time.

Data Analysis

Data were analyzed using a thematic analysis approach.¹⁶ Thematic analysis entails reading through the data to identify patterns (and create codes) that relate to behaviors, experiences, meanings, and activities. Once patterns have been identified, they are grouped according to similarity into themes, which help to further explain the findings.¹⁷

After the first site visit was completed, the research team members that participated (SS and MH) met to develop initial ideas about meanings and possible patterns. All transcripts were read by one team member (MH) and, based on review of the data, codes were developed, defined, and documented in a codebook. This process was repeated after every site visit using the codebook to expand or combine codes and refine definitions as necessary. If a new code was added, the previously coded data were reviewed to apply the new code. NVivo® 10 software (QSR International; Melbourne, Australia) was used to manage the data.

Once all field notes and transcripts were coded (MH), the code reports, which list all data described within a specific code, were run to ensure consistency and identify relationships between codes. Once coding was verified, codes were grouped based on similarities and relationships into salient themes by 3 members of the research team (NH, MH, and SM). Themes, along with their supporting codes, were then further defined to understand how these attendings worked to facilitate excellent teaching in clinical settings.

The coded interview data and field notes were categorized into broad, overlapping themes. Three of these major themes include (1) fostering positive relationships, (2) patient-centered teaching, and (3) collaboration and coaching. Table 2 lists each theme, salient behaviors, examples, and selected quotes that further elucidate its meaning.

Fostering Positive Relationships

Attending physicians took observable steps to develop positive relationships with their team members, which in turn created a safe learning environment. For instance, attendings used learners’ first names, demonstrated interest in their well-being, deployed humor, and generally displayed informal actions—uncrossed arms, “fist bump” when recognizing learners’ success, standing outside the circle of team members and leaning in to listen—during learner interactions. Attendings also made it a priority to get to know individuals on a personal level. As one current learner put it, “He asks about where we are from. He will try to find some kind of connection that he can establish with not only each of the team members but also with each of the patients.”

Additionally, attendings built positive relationships with their learners by responding thoughtfully to their input, even when learners’ evaluations of patients required modification. In turn, learners reported feeling safe to ask questions, admit uncertainty, and respectfully disagree with their attendings. As one attending reflected, “If I can get them into a place where they feel like the learning environment is someplace where they can make a mistake and know that that mistake does not necessarily mean that it’s going to cost them in their evaluation part, then I feel like that’s why it’s important.”

To build rapport and create a safe learning environment, attendings used a number of strategies to position themselves as learners alongside their team members. For instance, attendings indicated that they wanted their ideas questioned because they saw it as an opportunity to learn. Moreover, in conversations with learners, attendings demonstrated humility, admitting when they did not know something. One former learner noted, “There have been times when he has asked [a] question…nobody knows and then he admits that he doesn’t know either. So everybody goes and looks it up…The whole thing turns out to be a fun learning experience.”

Attendings demonstrated respect for their team members’ time by reading about patients before rounds, identifying learning opportunities during rounds, and integrating teaching points into the daily work of patient care. Teaching was not relegated exclusively to the conference room or confined to the traditional “chalk talk” before or after rounds but rather was assimilated into daily workflow. They appeared to be responsive to the needs of individual patients and the team, which allowed attendings to both directly oversee their patients’ care and overcome the challenges of multiple competing demands for time. The importance of this approach was made clear by one current learner who stated “…she does prepare before, especially you know on call days, she does prepare for the new patients before coming in to staff, which is really appreciated… it saves a lot of time on rounds.”

Attendings also included other health professionals in team discussions. Attendings used many of the same relationship-building techniques with these professionals as they did with learners and patients. They consistently asked these professionals to provide insight and direction in patients’ plans of care. A former learner commented, “He always asks the [nurse] what is her impression of the patient...he truly values the [nurse’s] opinion of the patient.” One attending reiterated this approach, stating “I don’t want them to think that anything I have to say is more valuable than our pharmacist or the [nurse].”

Patient-Centered Teaching

Attending physicians modeled numerous teaching techniques that focused learning around the patient. Attendings knew their patients well through review of the medical records, discussion with the patient, and personal examination. This preparation allowed attendings to focus on key teaching points in the context of the patient. One former learner noted, “He tended to bring up a variety of things that really fit well into the clinical scenario. So whether that is talking about what is the differential for a new symptom that just came up for this patient or kind of here is a new paper talking about this condition or maybe some other pearl of physical exam for a patient that has a certain physical condition.”

Attendings served as effective role models by being directly involved in examining and talking with patients as well as demonstrating excellent physical examination and communication techniques. One current learner articulated the importance of learning these skills by observing them done well: “I think he teaches by example and by doing, again, those little things: being attentive to the patients and being very careful during exams…I think those are things that you teach people by doing them, not by saying you need to do this better during the patient encounter.”

Collaboration and Coaching

Attending physicians used varied collaboration and coaching techniques to facilitate learning across the entire care team. During rounds, attendings utilized visual aids to reinforce key concepts and simplify complex topics. They also collaborated by using discussion rather than lecture to engage with team members. For instance, attendings used Socratic questioning, asking questions that lead learners through critical thinking and allow them to solve problems themselves, to guide learners’ decision-making. One former learner reported, “He never gives you the answer, and he always asks your opinion; ‘So what are your thoughts on this?’”

Coaching for success, rather than directing the various team members, was emphasized. Attendings did not wish to be seen as the “leaders” of the team. During rounds, one attending was noted to explain his role in ensuring that the team was building connections with others: “When we have a bad outcome, if it feels like your soul has been ripped out, then you’ve done something right. You’ve made that connection with the patient. My job, as your coach, was to build communication between all of us so we feel vested in each other and our patients.”

Attendings also fostered clinical reasoning skills in their learners by encouraging them to verbalize their thought processes aloud in order to clarify and check for understanding. Attendings also placed emphasis not simply on memorizing content but rather prioritization of the patient’s problems and thinking step by step through individual medical problems. One current learner applauded an attending who could “come up with schematics of how to approach problems rather than feeding us factual information of this paper or this trial.”

Additionally, attendings facilitated learning across the entire care team by differentiating their teaching to meet the needs of multiple learning levels. While the entire team was explicitly included in the learning process, attendings encouraged learners to play various roles, execute tasks, and answer questions depending on their educational level. Attendings positioned learners as leaders of the team by allowing them to talk without interruption and by encouraging them to take ownership of their patients’ care. One former learner stated, “She set expectations…we would be the ones who would be running the team, that you know it would very much be our team and that she is there to advise us and provide supervision but also safety for the patients as well.”

This study reveals the complex ways effective attendings build rapport, create a safe learning environment, utilize patient-centered teaching strategies, and engage in collaboration and coaching with all members of the team. These findings provide a framework of shared themes and their salient behaviors that may influence the success of inpatient general medicine clinician educators (Table 3).

There is a broad and voluminous literature on the subject of outstanding clinical teaching characteristics, much of which has shaped various faculty development curricula for decades. This study sought not to identify novel approaches of inpatient teaching necessarily but rather to closely examine the techniques and behaviors of clinician educators identified as exemplary. The findings affirm and reinforce the numerous, well-documented lists of personal attributes, techniques, and behaviors that resonate with learners, including creating a positive environment, demonstrating enthusiasm and interest in the learner, reading facial expressions, being student-centered, maintaining a high level of clinical knowledge, and utilizing effective communication skills.^18-24 The strengths of this study lie within the nuanced and rich observations and discussions that move beyond learners’ Likert scale evaluations and responses.^3-7,12 Input was sought from multiple perspectives on the care team, which provided detail from key stakeholders. Out of these comprehensive data arose several conclusions that extend the research literature on medical education.

In their seminal review, Sutkin et al.¹⁸ demonstrate that two thirds of characteristics of outstanding clinical teachers are “noncognitive” and that, “Perhaps what makes a clinical educator truly great depends less on the acquisition of cognitive skills such as medical knowledge and formulating learning objectives, and more on inherent, relationship-based, noncognitive attributes. Whereas cognitive abilities generally involve skills that may be taught and learned, albeit with difficulty, noncognitive abilities represent personal attributes, such as relationship skills, personality types, and emotional states, which are more difficult to develop and teach.”¹⁸ Our study, thus, adds to the literature by (1) highlighting examples of techniques and behaviors that encompass the crucial “noncognitive” arena and (2) informing best practices in teaching clinical medicine, especially those that resonate with learners, for future faculty development.

The findings highlight the role that relationships play in the teaching and learning of team-based medicine. Building rapport and sustaining successful relationships are cornerstones of effective teaching.¹⁸ For the attendings in this study, this manifested in observable, tangible behaviors such as greeting others by name, joking, using physical touch, and actively involving all team members, regardless of role or level of education. Previous literature has highlighted the importance of showing interest in learners.^7,19,25-27 This study provides multiple and varied examples of ways in which interest might be displayed.

For patients, the critical role of relationships was evidenced through rapport building and attention to patients as people outside their acute hospitalization. For instance, attendings regularly put patients’ medical issues into context and anticipated future outpatient challenges. To the authors’ knowledge, previous scholarship has not significantly emphasized this form of contextualized medicine, which involves the mindful consideration of the ongoing needs patients may experience upon transitions of care.

Several participants highlighted humility as an important characteristic of effective clinician educators. Attendings recognized that the field produces more new knowledge than can possibly be assimilated and that uncertainty is a mainstay of modern medical care. Attendings frequently utilized self-deprecation to acknowledge doubt, a technique that created a collaborative environment in which learners also felt safe to ask questions. These findings support the viewpoints by Reilly and Beckman that humility and an appreciation for questions and push-back from learners encourage lifelong learning through role modeling.^19,23 In responding to the interviewer’s question “And what happens when [the attending] is wrong?” one learner simply stated, “He makes fun of himself.”

This study has several limitations. First, it was conducted in a limited number of US based healthcare systems. The majority of institutions represented were larger, research intensive hospitals. While these hospitals were purposefully selected to provide a range in geography, size, type, and access to resources, the findings may differ in other settings. Second, it was conducted with a limited number of attendings and learners, which may limit the study’s generalizability. However, enough interviews were conducted to reach data saturation.¹⁵ Because evidence for a causal relationship between quality teaching and student and patient outcomes is lacking,¹⁸ we must rely on imperfect proxies for teaching excellence, including awards and recognition. This study attempted to identify exemplary educators through various means, but it is recognized that bias is likely. Third, because attendings provided lists of former learners, selection and recall biases may have been introduced, as attendings may have more readily identified former learners with whom they formed strong relationships. Fourth, focus was placed exclusively on teaching and learning within general medicine rounds. This was because there would be ample opportunity for teaching on this service, the structure of the teams and the types of patients would be comparable across sites, and the principal investigator was also a general medicine attending and would have a frame of reference for these types of rounds. Due to this narrow focus, the findings may not be generalizable to other subspecialties. Fifth, attendings were selected through a nonexhaustive method. However, the multisite design, the modified snowball sampling, and the inclusion of several types of institutions in the final participant pool introduced diversity to the final list. Finally, although we cannot discount the potential role of a Hawthorne effect on our data collection, the research team did attempt to mitigate this by standing apart from the care teams and remaining unobtrusive during observations.

Using a combination of interviews, focus group discussions, and direct observation, we identified consistent techniques and behaviors of excellent teaching attendings during inpatient general medicine rounds. We hope that all levels of clinician educators may use them to elevate their own teaching.

Disclosure

Dr. Saint is on a medical advisory board of Doximity, a new social networking site for physicians, and receives an honorarium. He is also on the scientific advisory board of Jvion, a healthcare technology company. Drs. Houchens, Harrod, Moody, and Ms. Fowler have no conflicts of interest.

Clinician educators face numerous obstacles to their joint mission of facilitating learning while also ensuring high-quality and patient-centered care. Time constraints, including the institution of house officer duty hour limitations,¹ shorter lengths of stay for hospitalized patients,² and competing career responsibilities, combine to create a dynamic learning environment. Additionally, clinician educators must balance the autonomy of their learners with the safety of their patients. They must teach to multiple learning levels and work collaboratively with multiple disciplines to foster an effective team-based approach to patient care. Yet, many clinician educators have no formal training in pedagogical methods.³ Such challenges necessitate increased attention to the work of excellent clinician educators and their respective teaching approaches.

Many studies of clinical teaching rely primarily on survey data of attributes of good clinical teachers.^3-7 While some studies have incorporated direct observations of teaching^8,9 or interviews with clinician educators or learners,^10,11 few have incorporated multiple perspectives from the current team and from former learners in order to provide a comprehensive picture of team-based learning.¹²

The goal of this study was to gain a thorough understanding, through multiple perspectives, of the techniques and behaviors used by exemplary educators within actual clinical environments. We studied attitudes, behaviors, and approaches of 12 such inpatient clinician educators.

Study Design and Sampling

This was a multisite study using an exploratory qualitative approach to inquiry. This approach was used to study the techniques and behaviors of excellent attendings during inpatient general medicine rounds. A modified snowball sampling approach¹³ was used, meaning individuals known to one member of the research team (SS) were initially contacted and asked to identify clinician educators (also referred to as attendings) for potential inclusion in the study. In an effort to identify attendings from a broad range of medical schools, the “2015 U.S. News and World Report Top Medical Schools: Research” rankings¹⁴ were also reviewed, with priority given to the top 25, as these are widely used to represent the best US hospitals. In an attempt to invite attendings from diverse institutions, additional medical schools not in the top 25 as well as historically black medical schools were also included. Division chiefs and chairs of internal medicine and/or directors of internal medicine residency programs at these schools were contacted and asked for recommendations of attendings, both within and outside their institutions, who they considered to be great inpatient teachers. In addition, key experts who have won teaching awards or were known to be specialists in the field of medical education were asked to nominate one or two other outstanding attendings.

By using this sampling method, 59 potential participants were identified. An internet search was conducted to obtain information about the potential participants and their institutions. Organizational characteristics such as geographic location, hospital size and affiliation, and patient population, as well as individual characteristics such as gender, medical education and training, and educational awards received were considered so that a diversity of organizations and backgrounds was represented. The list was narrowed down to 16 attendings who were contacted via e-mail and asked to participate. Interested participants were asked for a list of their current team members and 6 to 10 former learners to contact for interviews and focus groups. Former learners were included in an effort to better understand lasting effects on learners from their exemplary teaching attendings. A total of 12 attending physicians agreed to participate (Table 1). Literature on field methods has shown that 12 interviews are found to be adequate in accomplishing data saturation.¹⁵ Although 2 attendings were located at the same institution, we decided to include them given that both are recognized as master clinician educators and were each recommended by several individuals from various institutions. Hospitals were located throughout the US and included both university-affiliated hospitals and Veterans Affairs medical centers. Despite efforts to include physicians from historically black colleges and universities, only one attending was identified, and they declined the request to participate.

Data Collection

Observations. The one-day site visits were mainly conducted by two research team members, a physician (SS) and a medical anthropologist (MH), both of whom have extensive experience in qualitative methods. Teams were not uniform but were generally comprised of 1 attending, 1 senior medical resident, 1 to 2 interns, and approximately 2 medical students. Occasionally, a pharmacist, clinical assistant, or other health professional accompanied the team on rounds. Not infrequently, the bedside nurse would explicitly be included in the discussion regarding his or her specific patient. Each site visit began with observing attendings (N = 12) and current learners (N = 57) during rounds. Each research team member recorded their own observations via handwritten field notes, paying particular attention to group interactions, teaching approach, conversations occurring within and peripheral to the team, patient-team interactions, and the physical environment. By standing outside of the medical team circle and remaining silent during rounds, research team members remained unobtrusive to the discussion and process of rounds. Materials the attendings used during their teaching rounds were also documented and collected. Rounds generally lasted 2 to 3 hours. After each site visit, the research team met to compare and combine field notes.

Interviews and Focus Groups. The research team then conducted individual, semi-structured interviews with the attendings, focus groups with their current team (N = 46), and interviews or focus groups with their former learners (N = 26; Supplement 1). Eleven of the current team members observed during rounds were unable to participate in the focus groups due to clinical duties. Because the current learners who participated in the focus groups were also observed during rounds, the research team was able to ask them open-ended questions regarding teaching rounds and their roles as learners within this environment. Former learners who were still at the hospital participated in separate focus groups or interviews. Former learners who were no longer present at the hospital were contacted by telephone and individually interviewed by one research team member (MH). All interviews and focus groups were audio-recorded and transcribed.

This study was determined to be exempt by the University of Michigan Institutional Review Board. All participants were informed that their participation was completely voluntary and that they could terminate their involvement at any time.

Data Analysis

Data were analyzed using a thematic analysis approach.¹⁶ Thematic analysis entails reading through the data to identify patterns (and create codes) that relate to behaviors, experiences, meanings, and activities. Once patterns have been identified, they are grouped according to similarity into themes, which help to further explain the findings.¹⁷

After the first site visit was completed, the research team members that participated (SS and MH) met to develop initial ideas about meanings and possible patterns. All transcripts were read by one team member (MH) and, based on review of the data, codes were developed, defined, and documented in a codebook. This process was repeated after every site visit using the codebook to expand or combine codes and refine definitions as necessary. If a new code was added, the previously coded data were reviewed to apply the new code. NVivo® 10 software (QSR International; Melbourne, Australia) was used to manage the data.

Once all field notes and transcripts were coded (MH), the code reports, which list all data described within a specific code, were run to ensure consistency and identify relationships between codes. Once coding was verified, codes were grouped based on similarities and relationships into salient themes by 3 members of the research team (NH, MH, and SM). Themes, along with their supporting codes, were then further defined to understand how these attendings worked to facilitate excellent teaching in clinical settings.

The coded interview data and field notes were categorized into broad, overlapping themes. Three of these major themes include (1) fostering positive relationships, (2) patient-centered teaching, and (3) collaboration and coaching. Table 2 lists each theme, salient behaviors, examples, and selected quotes that further elucidate its meaning.

Fostering Positive Relationships

Attending physicians took observable steps to develop positive relationships with their team members, which in turn created a safe learning environment. For instance, attendings used learners’ first names, demonstrated interest in their well-being, deployed humor, and generally displayed informal actions—uncrossed arms, “fist bump” when recognizing learners’ success, standing outside the circle of team members and leaning in to listen—during learner interactions. Attendings also made it a priority to get to know individuals on a personal level. As one current learner put it, “He asks about where we are from. He will try to find some kind of connection that he can establish with not only each of the team members but also with each of the patients.”

Additionally, attendings built positive relationships with their learners by responding thoughtfully to their input, even when learners’ evaluations of patients required modification. In turn, learners reported feeling safe to ask questions, admit uncertainty, and respectfully disagree with their attendings. As one attending reflected, “If I can get them into a place where they feel like the learning environment is someplace where they can make a mistake and know that that mistake does not necessarily mean that it’s going to cost them in their evaluation part, then I feel like that’s why it’s important.”

To build rapport and create a safe learning environment, attendings used a number of strategies to position themselves as learners alongside their team members. For instance, attendings indicated that they wanted their ideas questioned because they saw it as an opportunity to learn. Moreover, in conversations with learners, attendings demonstrated humility, admitting when they did not know something. One former learner noted, “There have been times when he has asked [a] question…nobody knows and then he admits that he doesn’t know either. So everybody goes and looks it up…The whole thing turns out to be a fun learning experience.”

Attendings demonstrated respect for their team members’ time by reading about patients before rounds, identifying learning opportunities during rounds, and integrating teaching points into the daily work of patient care. Teaching was not relegated exclusively to the conference room or confined to the traditional “chalk talk” before or after rounds but rather was assimilated into daily workflow. They appeared to be responsive to the needs of individual patients and the team, which allowed attendings to both directly oversee their patients’ care and overcome the challenges of multiple competing demands for time. The importance of this approach was made clear by one current learner who stated “…she does prepare before, especially you know on call days, she does prepare for the new patients before coming in to staff, which is really appreciated… it saves a lot of time on rounds.”

Attendings also included other health professionals in team discussions. Attendings used many of the same relationship-building techniques with these professionals as they did with learners and patients. They consistently asked these professionals to provide insight and direction in patients’ plans of care. A former learner commented, “He always asks the [nurse] what is her impression of the patient...he truly values the [nurse’s] opinion of the patient.” One attending reiterated this approach, stating “I don’t want them to think that anything I have to say is more valuable than our pharmacist or the [nurse].”

Patient-Centered Teaching

Attending physicians modeled numerous teaching techniques that focused learning around the patient. Attendings knew their patients well through review of the medical records, discussion with the patient, and personal examination. This preparation allowed attendings to focus on key teaching points in the context of the patient. One former learner noted, “He tended to bring up a variety of things that really fit well into the clinical scenario. So whether that is talking about what is the differential for a new symptom that just came up for this patient or kind of here is a new paper talking about this condition or maybe some other pearl of physical exam for a patient that has a certain physical condition.”

Attendings served as effective role models by being directly involved in examining and talking with patients as well as demonstrating excellent physical examination and communication techniques. One current learner articulated the importance of learning these skills by observing them done well: “I think he teaches by example and by doing, again, those little things: being attentive to the patients and being very careful during exams…I think those are things that you teach people by doing them, not by saying you need to do this better during the patient encounter.”

Collaboration and Coaching

Attending physicians used varied collaboration and coaching techniques to facilitate learning across the entire care team. During rounds, attendings utilized visual aids to reinforce key concepts and simplify complex topics. They also collaborated by using discussion rather than lecture to engage with team members. For instance, attendings used Socratic questioning, asking questions that lead learners through critical thinking and allow them to solve problems themselves, to guide learners’ decision-making. One former learner reported, “He never gives you the answer, and he always asks your opinion; ‘So what are your thoughts on this?’”

Coaching for success, rather than directing the various team members, was emphasized. Attendings did not wish to be seen as the “leaders” of the team. During rounds, one attending was noted to explain his role in ensuring that the team was building connections with others: “When we have a bad outcome, if it feels like your soul has been ripped out, then you’ve done something right. You’ve made that connection with the patient. My job, as your coach, was to build communication between all of us so we feel vested in each other and our patients.”

Attendings also fostered clinical reasoning skills in their learners by encouraging them to verbalize their thought processes aloud in order to clarify and check for understanding. Attendings also placed emphasis not simply on memorizing content but rather prioritization of the patient’s problems and thinking step by step through individual medical problems. One current learner applauded an attending who could “come up with schematics of how to approach problems rather than feeding us factual information of this paper or this trial.”

Additionally, attendings facilitated learning across the entire care team by differentiating their teaching to meet the needs of multiple learning levels. While the entire team was explicitly included in the learning process, attendings encouraged learners to play various roles, execute tasks, and answer questions depending on their educational level. Attendings positioned learners as leaders of the team by allowing them to talk without interruption and by encouraging them to take ownership of their patients’ care. One former learner stated, “She set expectations…we would be the ones who would be running the team, that you know it would very much be our team and that she is there to advise us and provide supervision but also safety for the patients as well.”

This study reveals the complex ways effective attendings build rapport, create a safe learning environment, utilize patient-centered teaching strategies, and engage in collaboration and coaching with all members of the team. These findings provide a framework of shared themes and their salient behaviors that may influence the success of inpatient general medicine clinician educators (Table 3).

There is a broad and voluminous literature on the subject of outstanding clinical teaching characteristics, much of which has shaped various faculty development curricula for decades. This study sought not to identify novel approaches of inpatient teaching necessarily but rather to closely examine the techniques and behaviors of clinician educators identified as exemplary. The findings affirm and reinforce the numerous, well-documented lists of personal attributes, techniques, and behaviors that resonate with learners, including creating a positive environment, demonstrating enthusiasm and interest in the learner, reading facial expressions, being student-centered, maintaining a high level of clinical knowledge, and utilizing effective communication skills.^18-24 The strengths of this study lie within the nuanced and rich observations and discussions that move beyond learners’ Likert scale evaluations and responses.^3-7,12 Input was sought from multiple perspectives on the care team, which provided detail from key stakeholders. Out of these comprehensive data arose several conclusions that extend the research literature on medical education.

In their seminal review, Sutkin et al.¹⁸ demonstrate that two thirds of characteristics of outstanding clinical teachers are “noncognitive” and that, “Perhaps what makes a clinical educator truly great depends less on the acquisition of cognitive skills such as medical knowledge and formulating learning objectives, and more on inherent, relationship-based, noncognitive attributes. Whereas cognitive abilities generally involve skills that may be taught and learned, albeit with difficulty, noncognitive abilities represent personal attributes, such as relationship skills, personality types, and emotional states, which are more difficult to develop and teach.”¹⁸ Our study, thus, adds to the literature by (1) highlighting examples of techniques and behaviors that encompass the crucial “noncognitive” arena and (2) informing best practices in teaching clinical medicine, especially those that resonate with learners, for future faculty development.

The findings highlight the role that relationships play in the teaching and learning of team-based medicine. Building rapport and sustaining successful relationships are cornerstones of effective teaching.¹⁸ For the attendings in this study, this manifested in observable, tangible behaviors such as greeting others by name, joking, using physical touch, and actively involving all team members, regardless of role or level of education. Previous literature has highlighted the importance of showing interest in learners.^7,19,25-27 This study provides multiple and varied examples of ways in which interest might be displayed.

For patients, the critical role of relationships was evidenced through rapport building and attention to patients as people outside their acute hospitalization. For instance, attendings regularly put patients’ medical issues into context and anticipated future outpatient challenges. To the authors’ knowledge, previous scholarship has not significantly emphasized this form of contextualized medicine, which involves the mindful consideration of the ongoing needs patients may experience upon transitions of care.

Several participants highlighted humility as an important characteristic of effective clinician educators. Attendings recognized that the field produces more new knowledge than can possibly be assimilated and that uncertainty is a mainstay of modern medical care. Attendings frequently utilized self-deprecation to acknowledge doubt, a technique that created a collaborative environment in which learners also felt safe to ask questions. These findings support the viewpoints by Reilly and Beckman that humility and an appreciation for questions and push-back from learners encourage lifelong learning through role modeling.^19,23 In responding to the interviewer’s question “And what happens when [the attending] is wrong?” one learner simply stated, “He makes fun of himself.”

This study has several limitations. First, it was conducted in a limited number of US based healthcare systems. The majority of institutions represented were larger, research intensive hospitals. While these hospitals were purposefully selected to provide a range in geography, size, type, and access to resources, the findings may differ in other settings. Second, it was conducted with a limited number of attendings and learners, which may limit the study’s generalizability. However, enough interviews were conducted to reach data saturation.¹⁵ Because evidence for a causal relationship between quality teaching and student and patient outcomes is lacking,¹⁸ we must rely on imperfect proxies for teaching excellence, including awards and recognition. This study attempted to identify exemplary educators through various means, but it is recognized that bias is likely. Third, because attendings provided lists of former learners, selection and recall biases may have been introduced, as attendings may have more readily identified former learners with whom they formed strong relationships. Fourth, focus was placed exclusively on teaching and learning within general medicine rounds. This was because there would be ample opportunity for teaching on this service, the structure of the teams and the types of patients would be comparable across sites, and the principal investigator was also a general medicine attending and would have a frame of reference for these types of rounds. Due to this narrow focus, the findings may not be generalizable to other subspecialties. Fifth, attendings were selected through a nonexhaustive method. However, the multisite design, the modified snowball sampling, and the inclusion of several types of institutions in the final participant pool introduced diversity to the final list. Finally, although we cannot discount the potential role of a Hawthorne effect on our data collection, the research team did attempt to mitigate this by standing apart from the care teams and remaining unobtrusive during observations.

Using a combination of interviews, focus group discussions, and direct observation, we identified consistent techniques and behaviors of excellent teaching attendings during inpatient general medicine rounds. We hope that all levels of clinician educators may use them to elevate their own teaching.

Disclosure

Dr. Saint is on a medical advisory board of Doximity, a new social networking site for physicians, and receives an honorarium. He is also on the scientific advisory board of Jvion, a healthcare technology company. Drs. Houchens, Harrod, Moody, and Ms. Fowler have no conflicts of interest.

Clinician educators face numerous obstacles to their joint mission of facilitating learning while also ensuring high-quality and patient-centered care. Time constraints, including the institution of house officer duty hour limitations,¹ shorter lengths of stay for hospitalized patients,² and competing career responsibilities, combine to create a dynamic learning environment. Additionally, clinician educators must balance the autonomy of their learners with the safety of their patients. They must teach to multiple learning levels and work collaboratively with multiple disciplines to foster an effective team-based approach to patient care. Yet, many clinician educators have no formal training in pedagogical methods.³ Such challenges necessitate increased attention to the work of excellent clinician educators and their respective teaching approaches.

Many studies of clinical teaching rely primarily on survey data of attributes of good clinical teachers.^3-7 While some studies have incorporated direct observations of teaching^8,9 or interviews with clinician educators or learners,^10,11 few have incorporated multiple perspectives from the current team and from former learners in order to provide a comprehensive picture of team-based learning.¹²

The goal of this study was to gain a thorough understanding, through multiple perspectives, of the techniques and behaviors used by exemplary educators within actual clinical environments. We studied attitudes, behaviors, and approaches of 12 such inpatient clinician educators.

Study Design and Sampling

This was a multisite study using an exploratory qualitative approach to inquiry. This approach was used to study the techniques and behaviors of excellent attendings during inpatient general medicine rounds. A modified snowball sampling approach¹³ was used, meaning individuals known to one member of the research team (SS) were initially contacted and asked to identify clinician educators (also referred to as attendings) for potential inclusion in the study. In an effort to identify attendings from a broad range of medical schools, the “2015 U.S. News and World Report Top Medical Schools: Research” rankings¹⁴ were also reviewed, with priority given to the top 25, as these are widely used to represent the best US hospitals. In an attempt to invite attendings from diverse institutions, additional medical schools not in the top 25 as well as historically black medical schools were also included. Division chiefs and chairs of internal medicine and/or directors of internal medicine residency programs at these schools were contacted and asked for recommendations of attendings, both within and outside their institutions, who they considered to be great inpatient teachers. In addition, key experts who have won teaching awards or were known to be specialists in the field of medical education were asked to nominate one or two other outstanding attendings.

By using this sampling method, 59 potential participants were identified. An internet search was conducted to obtain information about the potential participants and their institutions. Organizational characteristics such as geographic location, hospital size and affiliation, and patient population, as well as individual characteristics such as gender, medical education and training, and educational awards received were considered so that a diversity of organizations and backgrounds was represented. The list was narrowed down to 16 attendings who were contacted via e-mail and asked to participate. Interested participants were asked for a list of their current team members and 6 to 10 former learners to contact for interviews and focus groups. Former learners were included in an effort to better understand lasting effects on learners from their exemplary teaching attendings. A total of 12 attending physicians agreed to participate (Table 1). Literature on field methods has shown that 12 interviews are found to be adequate in accomplishing data saturation.¹⁵ Although 2 attendings were located at the same institution, we decided to include them given that both are recognized as master clinician educators and were each recommended by several individuals from various institutions. Hospitals were located throughout the US and included both university-affiliated hospitals and Veterans Affairs medical centers. Despite efforts to include physicians from historically black colleges and universities, only one attending was identified, and they declined the request to participate.

Data Collection

Observations. The one-day site visits were mainly conducted by two research team members, a physician (SS) and a medical anthropologist (MH), both of whom have extensive experience in qualitative methods. Teams were not uniform but were generally comprised of 1 attending, 1 senior medical resident, 1 to 2 interns, and approximately 2 medical students. Occasionally, a pharmacist, clinical assistant, or other health professional accompanied the team on rounds. Not infrequently, the bedside nurse would explicitly be included in the discussion regarding his or her specific patient. Each site visit began with observing attendings (N = 12) and current learners (N = 57) during rounds. Each research team member recorded their own observations via handwritten field notes, paying particular attention to group interactions, teaching approach, conversations occurring within and peripheral to the team, patient-team interactions, and the physical environment. By standing outside of the medical team circle and remaining silent during rounds, research team members remained unobtrusive to the discussion and process of rounds. Materials the attendings used during their teaching rounds were also documented and collected. Rounds generally lasted 2 to 3 hours. After each site visit, the research team met to compare and combine field notes.

Interviews and Focus Groups. The research team then conducted individual, semi-structured interviews with the attendings, focus groups with their current team (N = 46), and interviews or focus groups with their former learners (N = 26; Supplement 1). Eleven of the current team members observed during rounds were unable to participate in the focus groups due to clinical duties. Because the current learners who participated in the focus groups were also observed during rounds, the research team was able to ask them open-ended questions regarding teaching rounds and their roles as learners within this environment. Former learners who were still at the hospital participated in separate focus groups or interviews. Former learners who were no longer present at the hospital were contacted by telephone and individually interviewed by one research team member (MH). All interviews and focus groups were audio-recorded and transcribed.

This study was determined to be exempt by the University of Michigan Institutional Review Board. All participants were informed that their participation was completely voluntary and that they could terminate their involvement at any time.

Data Analysis

Data were analyzed using a thematic analysis approach.¹⁶ Thematic analysis entails reading through the data to identify patterns (and create codes) that relate to behaviors, experiences, meanings, and activities. Once patterns have been identified, they are grouped according to similarity into themes, which help to further explain the findings.¹⁷

After the first site visit was completed, the research team members that participated (SS and MH) met to develop initial ideas about meanings and possible patterns. All transcripts were read by one team member (MH) and, based on review of the data, codes were developed, defined, and documented in a codebook. This process was repeated after every site visit using the codebook to expand or combine codes and refine definitions as necessary. If a new code was added, the previously coded data were reviewed to apply the new code. NVivo® 10 software (QSR International; Melbourne, Australia) was used to manage the data.

Once all field notes and transcripts were coded (MH), the code reports, which list all data described within a specific code, were run to ensure consistency and identify relationships between codes. Once coding was verified, codes were grouped based on similarities and relationships into salient themes by 3 members of the research team (NH, MH, and SM). Themes, along with their supporting codes, were then further defined to understand how these attendings worked to facilitate excellent teaching in clinical settings.

The coded interview data and field notes were categorized into broad, overlapping themes. Three of these major themes include (1) fostering positive relationships, (2) patient-centered teaching, and (3) collaboration and coaching. Table 2 lists each theme, salient behaviors, examples, and selected quotes that further elucidate its meaning.

Fostering Positive Relationships

Attending physicians took observable steps to develop positive relationships with their team members, which in turn created a safe learning environment. For instance, attendings used learners’ first names, demonstrated interest in their well-being, deployed humor, and generally displayed informal actions—uncrossed arms, “fist bump” when recognizing learners’ success, standing outside the circle of team members and leaning in to listen—during learner interactions. Attendings also made it a priority to get to know individuals on a personal level. As one current learner put it, “He asks about where we are from. He will try to find some kind of connection that he can establish with not only each of the team members but also with each of the patients.”

Additionally, attendings built positive relationships with their learners by responding thoughtfully to their input, even when learners’ evaluations of patients required modification. In turn, learners reported feeling safe to ask questions, admit uncertainty, and respectfully disagree with their attendings. As one attending reflected, “If I can get them into a place where they feel like the learning environment is someplace where they can make a mistake and know that that mistake does not necessarily mean that it’s going to cost them in their evaluation part, then I feel like that’s why it’s important.”

To build rapport and create a safe learning environment, attendings used a number of strategies to position themselves as learners alongside their team members. For instance, attendings indicated that they wanted their ideas questioned because they saw it as an opportunity to learn. Moreover, in conversations with learners, attendings demonstrated humility, admitting when they did not know something. One former learner noted, “There have been times when he has asked [a] question…nobody knows and then he admits that he doesn’t know either. So everybody goes and looks it up…The whole thing turns out to be a fun learning experience.”

Attendings demonstrated respect for their team members’ time by reading about patients before rounds, identifying learning opportunities during rounds, and integrating teaching points into the daily work of patient care. Teaching was not relegated exclusively to the conference room or confined to the traditional “chalk talk” before or after rounds but rather was assimilated into daily workflow. They appeared to be responsive to the needs of individual patients and the team, which allowed attendings to both directly oversee their patients’ care and overcome the challenges of multiple competing demands for time. The importance of this approach was made clear by one current learner who stated “…she does prepare before, especially you know on call days, she does prepare for the new patients before coming in to staff, which is really appreciated… it saves a lot of time on rounds.”

Attendings also included other health professionals in team discussions. Attendings used many of the same relationship-building techniques with these professionals as they did with learners and patients. They consistently asked these professionals to provide insight and direction in patients’ plans of care. A former learner commented, “He always asks the [nurse] what is her impression of the patient...he truly values the [nurse’s] opinion of the patient.” One attending reiterated this approach, stating “I don’t want them to think that anything I have to say is more valuable than our pharmacist or the [nurse].”

Patient-Centered Teaching

Attending physicians modeled numerous teaching techniques that focused learning around the patient. Attendings knew their patients well through review of the medical records, discussion with the patient, and personal examination. This preparation allowed attendings to focus on key teaching points in the context of the patient. One former learner noted, “He tended to bring up a variety of things that really fit well into the clinical scenario. So whether that is talking about what is the differential for a new symptom that just came up for this patient or kind of here is a new paper talking about this condition or maybe some other pearl of physical exam for a patient that has a certain physical condition.”

Attendings served as effective role models by being directly involved in examining and talking with patients as well as demonstrating excellent physical examination and communication techniques. One current learner articulated the importance of learning these skills by observing them done well: “I think he teaches by example and by doing, again, those little things: being attentive to the patients and being very careful during exams…I think those are things that you teach people by doing them, not by saying you need to do this better during the patient encounter.”

Collaboration and Coaching

Attending physicians used varied collaboration and coaching techniques to facilitate learning across the entire care team. During rounds, attendings utilized visual aids to reinforce key concepts and simplify complex topics. They also collaborated by using discussion rather than lecture to engage with team members. For instance, attendings used Socratic questioning, asking questions that lead learners through critical thinking and allow them to solve problems themselves, to guide learners’ decision-making. One former learner reported, “He never gives you the answer, and he always asks your opinion; ‘So what are your thoughts on this?’”

Coaching for success, rather than directing the various team members, was emphasized. Attendings did not wish to be seen as the “leaders” of the team. During rounds, one attending was noted to explain his role in ensuring that the team was building connections with others: “When we have a bad outcome, if it feels like your soul has been ripped out, then you’ve done something right. You’ve made that connection with the patient. My job, as your coach, was to build communication between all of us so we feel vested in each other and our patients.”

Attendings also fostered clinical reasoning skills in their learners by encouraging them to verbalize their thought processes aloud in order to clarify and check for understanding. Attendings also placed emphasis not simply on memorizing content but rather prioritization of the patient’s problems and thinking step by step through individual medical problems. One current learner applauded an attending who could “come up with schematics of how to approach problems rather than feeding us factual information of this paper or this trial.”

Additionally, attendings facilitated learning across the entire care team by differentiating their teaching to meet the needs of multiple learning levels. While the entire team was explicitly included in the learning process, attendings encouraged learners to play various roles, execute tasks, and answer questions depending on their educational level. Attendings positioned learners as leaders of the team by allowing them to talk without interruption and by encouraging them to take ownership of their patients’ care. One former learner stated, “She set expectations…we would be the ones who would be running the team, that you know it would very much be our team and that she is there to advise us and provide supervision but also safety for the patients as well.”

This study reveals the complex ways effective attendings build rapport, create a safe learning environment, utilize patient-centered teaching strategies, and engage in collaboration and coaching with all members of the team. These findings provide a framework of shared themes and their salient behaviors that may influence the success of inpatient general medicine clinician educators (Table 3).

There is a broad and voluminous literature on the subject of outstanding clinical teaching characteristics, much of which has shaped various faculty development curricula for decades. This study sought not to identify novel approaches of inpatient teaching necessarily but rather to closely examine the techniques and behaviors of clinician educators identified as exemplary. The findings affirm and reinforce the numerous, well-documented lists of personal attributes, techniques, and behaviors that resonate with learners, including creating a positive environment, demonstrating enthusiasm and interest in the learner, reading facial expressions, being student-centered, maintaining a high level of clinical knowledge, and utilizing effective communication skills.^18-24 The strengths of this study lie within the nuanced and rich observations and discussions that move beyond learners’ Likert scale evaluations and responses.^3-7,12 Input was sought from multiple perspectives on the care team, which provided detail from key stakeholders. Out of these comprehensive data arose several conclusions that extend the research literature on medical education.

In their seminal review, Sutkin et al.¹⁸ demonstrate that two thirds of characteristics of outstanding clinical teachers are “noncognitive” and that, “Perhaps what makes a clinical educator truly great depends less on the acquisition of cognitive skills such as medical knowledge and formulating learning objectives, and more on inherent, relationship-based, noncognitive attributes. Whereas cognitive abilities generally involve skills that may be taught and learned, albeit with difficulty, noncognitive abilities represent personal attributes, such as relationship skills, personality types, and emotional states, which are more difficult to develop and teach.”¹⁸ Our study, thus, adds to the literature by (1) highlighting examples of techniques and behaviors that encompass the crucial “noncognitive” arena and (2) informing best practices in teaching clinical medicine, especially those that resonate with learners, for future faculty development.

The findings highlight the role that relationships play in the teaching and learning of team-based medicine. Building rapport and sustaining successful relationships are cornerstones of effective teaching.¹⁸ For the attendings in this study, this manifested in observable, tangible behaviors such as greeting others by name, joking, using physical touch, and actively involving all team members, regardless of role or level of education. Previous literature has highlighted the importance of showing interest in learners.^7,19,25-27 This study provides multiple and varied examples of ways in which interest might be displayed.

For patients, the critical role of relationships was evidenced through rapport building and attention to patients as people outside their acute hospitalization. For instance, attendings regularly put patients’ medical issues into context and anticipated future outpatient challenges. To the authors’ knowledge, previous scholarship has not significantly emphasized this form of contextualized medicine, which involves the mindful consideration of the ongoing needs patients may experience upon transitions of care.

Several participants highlighted humility as an important characteristic of effective clinician educators. Attendings recognized that the field produces more new knowledge than can possibly be assimilated and that uncertainty is a mainstay of modern medical care. Attendings frequently utilized self-deprecation to acknowledge doubt, a technique that created a collaborative environment in which learners also felt safe to ask questions. These findings support the viewpoints by Reilly and Beckman that humility and an appreciation for questions and push-back from learners encourage lifelong learning through role modeling.^19,23 In responding to the interviewer’s question “And what happens when [the attending] is wrong?” one learner simply stated, “He makes fun of himself.”

This study has several limitations. First, it was conducted in a limited number of US based healthcare systems. The majority of institutions represented were larger, research intensive hospitals. While these hospitals were purposefully selected to provide a range in geography, size, type, and access to resources, the findings may differ in other settings. Second, it was conducted with a limited number of attendings and learners, which may limit the study’s generalizability. However, enough interviews were conducted to reach data saturation.¹⁵ Because evidence for a causal relationship between quality teaching and student and patient outcomes is lacking,¹⁸ we must rely on imperfect proxies for teaching excellence, including awards and recognition. This study attempted to identify exemplary educators through various means, but it is recognized that bias is likely. Third, because attendings provided lists of former learners, selection and recall biases may have been introduced, as attendings may have more readily identified former learners with whom they formed strong relationships. Fourth, focus was placed exclusively on teaching and learning within general medicine rounds. This was because there would be ample opportunity for teaching on this service, the structure of the teams and the types of patients would be comparable across sites, and the principal investigator was also a general medicine attending and would have a frame of reference for these types of rounds. Due to this narrow focus, the findings may not be generalizable to other subspecialties. Fifth, attendings were selected through a nonexhaustive method. However, the multisite design, the modified snowball sampling, and the inclusion of several types of institutions in the final participant pool introduced diversity to the final list. Finally, although we cannot discount the potential role of a Hawthorne effect on our data collection, the research team did attempt to mitigate this by standing apart from the care teams and remaining unobtrusive during observations.

Using a combination of interviews, focus group discussions, and direct observation, we identified consistent techniques and behaviors of excellent teaching attendings during inpatient general medicine rounds. We hope that all levels of clinician educators may use them to elevate their own teaching.

Disclosure

Dr. Saint is on a medical advisory board of Doximity, a new social networking site for physicians, and receives an honorarium. He is also on the scientific advisory board of Jvion, a healthcare technology company. Drs. Houchens, Harrod, Moody, and Ms. Fowler have no conflicts of interest.

click for large version

Case

A 38-year-old man with a history of tobacco use presents to the emergency department complaining of constant substernal chest pain for three hours. His temperature is 37.7°C, his heart rate is 110 beats per minute, and his blood pressure is 155/95 mmHg. He appears anxious and diaphoretic but examination is otherwise unremarkable. He admits to cocaine use one hour before the onset of symptoms. What are the appropriate treatments for his condition?

Overview

Cocaine is the second-most-commonly used illicit drug in the U.S. and represents 31% of all ED visits related to substance abuse.^1,2 According to recent survey results, 2.1 million people report recent cocaine use, and 1.6 million engage in cocaine abuse or dependence.² Acute cardiopulmonary complaints are common in individuals who present to the ED after cocaine use, with chest pain being the most frequently reported symptom in 40%.³

Numerous etiologies for cocaine-associated chest pain (CACP) have been discovered, including musculoskeletal pain, pulmonary hypertension, cardiomyopathy, arrhythmias, and endocarditis.⁴ Only 0.5% of patients with aortic dissection over a four-year period had a recent history of cocaine use, making cocaine a rare cause of a rare condition.⁵ Cardiac chest pain remains the most frequent underlying etiology, resulting in the most common complication of myocardial infarction (MI) in up to 6% of patients.^6,7

The ways in which cocaine use can cause myocardial ischemia and MI are multifactorial. A vigorous central sympathomimetic effect, coronary artery vasoconstriction, stimulation of platelets, and enhanced atherosclerosis all lead to a myocardial oxygen supply-demand imbalance.⁸ Other key interactions in the cardiovascular system are displayed in Figure 1. Understanding the role of these mechanisms in CACP is crucial to patient care.

Clinician goals in the management of CACP are to rapidly and accurately exclude life-threatening etiologies; assess the need for urgent acute coronary syndrome (ACS) evaluation; risk-stratify patients and ensure appropriate disposition; normalize the toxic effects of cocaine; treat resultant organ damage; and prevent long-term complications. An algorithm detailing this approach is provided in Figure 2.

click for large version

click for large version

Review of the Data

Diagnostic evaluation. Given potential differences in treatment regimens, it is imperative to differentiate patients who present with CACP from those whose chest pain is not associated with cocaine either by direct questioning or by screening of urine for cocaine metabolites. Once the presence of cocaine has been confirmed, guideline-based evaluation for potential ACS with serial electrocardiograms (ECG), cardiac biomarkers, and close monitoring of cardiac rhythms and hemodynamics is largely similar to standard management of all patients presenting with chest pain, with a few caveats.

Interpretation of the ECG can be challenging in the setting of cocaine. Studies have shown “abnormal” ECGs in 56% to 84% of patients, with many representing early repolarization or left ventricular hypertrophy.^9,10 Likewise, patients with MI are as likely to present with normal or nonspecific ECG findings as with ischemic findings.^7,11 ECG interpretation to diagnose ischemia or infarction in patients with CACP yields a sensitivity of 36% and specificity of 90%.⁷

Creatine kinase (CK), CK-MB fraction, and myoglobin have low specificity for the diagnosis of ischemia, as cocaine can induce skeletal muscle injury and rhabdomyolysis.^9,12 Cardiac troponins demonstrate a superior specificity compared to CK and CK-MB and are thus the preferred cardiac biomarkers in diagnosing cocaine-associated MI.¹²

Initial management and disposition. Patients at high risk for cardiovascular events are generally admitted to a monitored bed.¹³ Immediate reperfusion therapy with primary percutaneous coronary intervention is recommended in patients with ST-elevation MI (STEMI). Treatment with thrombolytic agents is associated with an increased risk of intracerebral hemorrhage and lacks documented efficacy in patients with CACP. Thrombolysis should therefore only be utilized if the diagnosis of STEMI is unequivocal and an experienced cardiac catheterization laboratory is unavailable.^14,15

Patients with unstable angina (UA) or non-ST-elevation MI (NSTEMI) are at higher risk for further cardiac events in a similar manner to those with ACS unrelated to cocaine. These cases might benefit from early cardiac catheterization and revascularization.¹⁶ Because of the increased risk of stent thrombosis in cocaine-users, thought to be due to recidivism, a detailed risk-benefit analysis should be undertaken prior to the implantation of cardiac stents.

Other diagnostic tests, such as stress testing and myocardial imaging, have not shown significant accuracy in diagnosing MI in this setting; moreover, these patients are at low overall risk for cardiac events and mortality. Consequently, an extensive diagnostic evaluation might not be cost-effective.^7,10,13,17 Patients who have CACP without MI have a very low frequency of delayed complications.^3,17 As such, cost-effective evaluation strategies, such as nine- or 12-hour observation periods in a chest pain unit, are appropriate for many of these low- to moderate-risk patients.¹³ For all CACP patients, the most critical post-discharge interventions are cardiac risk modification and cocaine cessation.¹³

Normalizing the toxic effects of cocaine with medications.

Aspirin: While no specific study has been performed in patients with CACP and aspirin, CACP guidelines, based on data supporting ACS guidelines for all patients, recommend administration of full-dose aspirin given its associated reduction in morbidity and mortality.^18,19 Furthermore, given the platelet-stimulating effects of cocaine, using aspirin in this setting seems very reasonable.

Benzodiazepines: CACP guidelines support the use of benzodiazepines early in management to indirectly combat the agitation, hypertension, and tachycardia resulting from the stimulatory effects of cocaine.^18,20 These recommendations are based on several animal and human studies that demonstrate significant reduction in heart rate and systemic arterial pressure with the use of these agents.^21,22

Nitroglycerin: Cardiac catheterization studies have shown reversal of vasoconstriction with administration of nitroglycerin. One study demonstrated a benefit of the drug in 49% of participants.²³ Additional investigation into the benefit of benzodiazepine and nitroglycerin combination therapy revealed mixed results. In one study, lorazepam plus nitroglycerin was found to be more efficacious than nitroglycerin alone.²⁴ In another, however, use of diazepam in combination with nitroglycerin did not show benefit when evaluating pain relief, cardiac dynamics, and left ventricular function.²⁵

Phentolamine: Phentolamine administration has been studied much less in the literature. This nonselective alpha-adrenergic antagonist exerts a dose-dependent reversal of cocaine’s vasoconstrictive properties in monkeys and humans.^26,27 International guidelines for Emergency Cardiovascular Care recommend its use in treatment of cocaine-associated ACS;27 however, the AHA recommends it less strongly.¹⁸

Calcium channel blockers: Calcium channel blockers (CCBs) have not shown promise as first-line agents. While catheterization studies demonstrate the vasodilatory properties of verapamil, larger studies looking at all-cause mortality conclude that CCBs might worsen mortality rates,²⁸ and animal studies indicate an increased risk of seizures.²⁹ At this time, CCBs are recommended only if cardiac symptoms continue after both benzodiazepines and nitroglycerin are administered.¹⁸

The beta-blocker controversy: The use of beta-blockers in patients with CACP remains controversial given the theoretical risk of unopposed alpha-adrenergic activation. Coronary vasospasm, decreased myocardial oxygen delivery, and increased systemic vascular resistance can result from their use.³⁰

Propranolol, a nonselective beta-blocker, was shown in catheterization studies to potentiate the coronary vasoconstriction of cocaine.³¹ Labetalol, a combined alpha/beta-blocker, reduced mean arterial pressure after cocaine administration during cardiac catheterization but did not reverse coronary vasoconstriction.³² This was attributed to the predominating beta greater than alpha blockade at doses administered. The selective beta-1 antagonists esmolol and metoprolol have shown no benefit in CACP.³³ Carvedilol, a combined alpha/beta-blocker with both peripheral and central nervous system activity, has potential to attenuate both physiologic and behavioral response to cocaine, but it has not been well studied in this patient subset.³⁴

click for large version

The 2005 ACC/AHA STEMI guidelines recommended against beta-blockers in the setting of STEMI precipitated by cocaine use due to the potential of exacerbating coronary vasoconstriction.³⁵ The 2007 ACC/AHA UA/NSTEMI guidelines stated that the use of a combined alpha/beta-blocker in patients with cocaine-induced ACS may be reasonable for patients with hypertension or tachycardia if pre-treated with a vasodilator.¹⁹ The 2008 ACC/AHA guidelines on the management of cocaine-related chest pain and MI recommended against the use of beta-blockers in the acute setting given the low incidence of cocaine-related MI and death.¹⁸

In a more recent study, Dattilo et al showed that beta-blockers administered to patients admitted with positive urine toxicology for cocaine significantly reduced MI and in-hospital mortality. Reduction of MI was of borderline significance in those admitted with a chief complaint of chest pain.³⁶ Limitations of this study include unknown time of cocaine ingestion, lack of follow-up on discharge mortality, and a small sample size of 348 patients lacking statistical power.

Another retrospective cohort study examined patients admitted with chest pain and urine toxicology positive for cocaine and found that beta-blocker administration during hospitalization was not associated with increased incident mortality. Further, after a mean follow-up of 2.5 years, there was a statistically significant decrease in cardiovascular death.³⁷ Drawbacks of this study included an older patient population, greater proportion of coronary artery disease, and higher follow-up of cardiovascular mortality rates than in previous studies, suggesting this subset might have received greater benefit from beta-blockers as a result of these characteristics.

The 2008 ACC/AHA guidelines instruct individualized consideration of the risk/benefit ratio for beta-blocker use in patients with CACP given the high rate of recidivism in cocaine abusers. The strongest indication is given to those with documented MI, left ventricular systolic dysfunction, or ventricular arrhythmias.¹⁸

It is important to note that these recommendations are based on cardiac catheterization laboratory studies, case reports, retrospective analyses, and animal experiments. No prospective controlled trials evaluating the role of beta-blockers in CACP and MI exist, and no trials regarding therapies to improve outcomes of patients sustaining a cocaine-associated MI have been reported.¹⁸

Back to the Case

This patient was experiencing cocaine-associated chest pain, which was confirmed with positive urine toxicology. Initial diagnostic workup with basic laboratory studies and cardiac biomarkers showed mild elevation in CK with troponin levels within normal limits. His ECG showed changes consistent with left ventricular hypertrophy. Chest radiograph was unremarkable.

He received aspirin, benzodiazepines, and nitroglycerin with normalization of vital signs, as well as subjective improvement in chest pain and anxiety. He was deemed to be at low risk for potential cardiac complications; thus, further cardiac testing was not pursued. Rather, he was admitted to an overnight observation unit with telemetry monitoring, where his chest pain did not recur.

He was seen in consultation with social work staff who arranged for drug abuse counseling after discharge. Given the uncertainty of relapse to cocaine use, as well as lack of known cardiac risk factors, he was not discharged on any new medications.

Bottom Line

The treatment of CACP includes normalizing the toxic systemic effects of the drug and minimizing the direct ischemic damage to the myocardium. Management varies slightly from traditional chest pain algorithms and includes benzodiazepines as well as antiplatelet agents and vasodilators to achieve this goal. Initial therapy with beta-blockers remains undefined and is largely discouraged in the acute setting. The role of beta-blockade upon discharge, however, can be beneficial in specific populations, especially those found to have underlying coronary disease.

Dr. Houchens and Dr. Czarnik are clinical instructors and Dr. Mack is a clinical lecturer at the University of Michigan Health System in Ann Arbor.

References

1. Hughes A, Sathe N, Spagnola K. State Estimates of Substance Use from the 2005-2006 National Surveys on Drug Use and Health. DHHS Publication No. SMA 08-4311, NSDUH Series H-33. Rockville, MD: Substance Abuse and Mental Health Services Administration, Office of Applied Studies; 2008.
2. Volkow ND. Cocaine: Abuse and Addiction. National Institute on Drug Abuse. Washington, DC: U.S. Department of Health and Human Services; 2009.
3. Brody SL, Slovis CM, Wrenn KD. Cocaine-related medical problems: consecutive series of 233 patients. Am J Med. 1990;88:325-331.
4. Levis JT, Garmel GM. Cocaine-associated chest pain. Emerg Med Clin North Am. 2005;23:1083-1103.
5. Eagle KA, Isselbacher EM, DeSanctis RW. Cocaine-related aortic dissection in perspective. Circulation. 2002;105:1529-1530.
6. Feldman JA, Fish SS, Beshansky JR, Griffith JL, Woolard RH, Selker HP. Acute cardiac ischemia in patients with cocaine-associated complaints: results of a multicenter trial. Ann Emerg Med. 2000;36:469-476.
7. Hollander JE, Hoffman RS, Gennis P, et al. Prospective multicenter evaluation of cocaine associated chest pain. Cocaine Associated Chest Pain (COCHPA) Study Group. Acad Emerg Med. 1994;1:330-339.
8. Schwartz BG, Rezkalla S, Kloner RA. Cardiovascular effects of cocaine. Circulation. 2010;122:2558-2569.
9. Gitter MJ, Goldsmith SR, Dunbar DN, et al. Cocaine and chest pain: clinical features and outcomes of patients hospitalized to rule out myocardial infarction. Ann Intern Med. 1991;115:277-282.
10. Amin M, Gabelman G, Karpel J, et al. Acute myocardial infarction and chest pain syndromes after cocaine use. Am J Cardiol. 1990;66:1434-1437.
11. Tokarski GF, Paganussi P, Urbanski R, et al. An evaluation of cocaine-induced chest pain. Ann Emerg Med. 1990;19:1088-1092.
12. Hollander JE, Levitt MA, Young GP, Briglia E, Wetli CV, Gawad Y. Effect of recent cocaine use on the specificity of cardiac markers for diagnosis of acute myocardial infarction. Am Heart J. 1998;135(2 Pt 1):245-252.
13. Weber JE, Shofer FS, Larkin GL, Kalaria AS, Hollander JE. Validation of a brief observation period for patients with cocaine-associated chest pain. N Engl J Med. 2003;348:510-517.
14. Hahn IH, Hoffman RS. Diagnosis and treatment of acute myocardial infarction: cocaine use and acute myocardial infarction. Emerg Med Clin North Am. 2001;19(2):1-18.
15. Hoffman RS, Hollander JE. Evaluation of patients with chest pain after cocaine use. Crit Care Clin. 1997;13:809-828. Cannon CP, Weintraub WS, Demopoulos LA, et al.
16. Comparison of early invasive and conservative strategies in patients with unstable coronary syndromes treated with the glycoprotein IIb/IIIa inhibitor tirofiban. N Engl J Med. 2001;344:1879-1887.
17. Hollander JE, Hoffman RS. Cocaine-induced myocardial infarction: an analysis and review of the literature. J Emerg Med. 1992;10:169-177.
18. McCord J, Jneid H, Hollander JE, et al. Management of cocaine-associated chest pain and myocardial infarction: a scientific statement from the American Heart Association Acute Cardiac Care Committee of the Council on Clinical Cardiology. Circulation. 2008;117:1897-1907.
19. Anderson JL, Adams CD, Antman EM, et al. ACC/AHA 2007 guidelines for the management of patients with unstable angina/non-ST-Elevation myocardial infarction: a report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines (Writing Committee to Revise the 2002 Guidelines for the Management of Patients With Unstable Angina/Non-ST-Elevation Myocardial Infarction) developed in collaboration with the American College of Emergency Physicians, the Society for Cardiovascular Angiography and Interventions, and the Society of Thoracic Surgeons endorsed by the American Association of Cardiovascular and Pulmonary Rehabilitation and the Society for Academic Emergency Medicine. J Am Coll Cardiol. 2007;50:E1-E157.
20. Hollander JE. Management of cocaine-associated myocardial ischemia. N Engl J Med. 1995;333:1267-1272.
21. Brubacher JR, Hoffman RS. Cocaine toxicity. Top Emerg Med. 1997;19(4):1-16.
22. Catavas JD, Waters IW. Acute cocaine intoxication in the conscious dog: studies on the mechanism of lethality. J Pharmacol Exp Ther. 1981;217:350-356.
23. Hollander JE, Hoffman RS, Gennis P, et al. Nitroglycerin in the treatment of cocaine associated chest pain—clinical safety and efficacy. J Toxicol Clin Toxicol. 1994;32(3): 243-256.
24. Honderick T, Williams D, Seaberg D, Wears R. A prospective, randomized, controlled trial of benzodiazepines and nitroglycerin or nitroglycerin alone in the treatment of cocaine-associated acute coronary syndromes. Am J Emerg Med. 2003;21(1):39-42.
25. Baumann BM, Perrone J, Hornig SE, Shofer FS, Hollander JE. Randomized, double-blind, placebo-controlled trial of diazepam, nitroglycerin, or both for treatment of patients with potential cocaine-associated acute coronary syndromes. Acad Emerg Med. 2000;7:878-885.
26. Schindler CW, Tella SR, Goldberg SR. Adrenoceptor mechanisms in the cardiovascular effects of cocaine in conscious squirrel monkeys. Life Sci. 1992;51(9):653-660.
27. Lange RA, Cigarroa RG, Yancy CW Jr., et al. Cocaine-induced coronary-artery vasoconstriction. N Engl J Med. 1989;321(23):1557-1562.
28. Furberg CD, Psaty BM, Meyer JV. Nifedipine. Dose-related increase in mortality in patients with coronary heart disease. Circulation. 1995;92:1326-1331.
29. Derlet RW, Albertson TE. Potentiation of cocaine toxicity with calcium channel blockers. Am J Emerg Med. 1989;7:464-468.
30. Lange RA, Hillis LD. Cardiovascular complications of cocaine use. N Engl J Med. 2001;345:351-358.
31. Lange RA, Cigarroa RG, Flores ED, et al. Potentiation of cocaine-induced coronary vasoconstriction by beta-adrenergic blockade. Ann Intern Med. 1990;112:897-903.
32. Boehrer JD, Moliterno DJ, Willard JE, Hillis LD, Lange RA. Influence of labetalol on cocaine-induced coronary vasoconstriction in humans. Am J Med. 1993;94:608-610.
33. Sand IC, Brody SL, Wrenn KD, Slovis CM. Experience with esmolol for the treatment of cocaine-associated cardiovascular complications. Am J Emerg Med. 1991;9:161-163.
34. Sofuoglo M, Brown S, Babb DA, Pentel PR, Hatsukami DK. Carvedilol affects the physiological and behavioral response to smoked cocaine in humans. Drug Alcohol Depend. 2000;60:69-76.
35. Antman EM, Anbe DT, Armstrong PW, et al. ACC/AHA guidelines for the management of patients with ST-elevation myocardial infarction: a report of the American College of Cardiology/American Heart Association Task Force of Practice Guidelines (Committee to Revise the 1999 Guidelines for the Management of patients with acute myocardial infarction). J Am Coll Cardiol. 2004;44:E1-E211.
36. Dattilo PB, Hailpern SM, Fearon K, Sohal D, Nordin C. Beta-blockers are associated with reduced risk of myocardial infarction after cocaine use. Ann Emerg Med. 2008;51:117-125.
37. Rangel C, Shu RG, Lazar LD, Vittinghoff E, Hsue PY, Marcus GM. Beta-blockers for chest pain associated with recent cocaine use. Arch Intern Med. 2010;170:874-879.

click for large version

Case

A 38-year-old man with a history of tobacco use presents to the emergency department complaining of constant substernal chest pain for three hours. His temperature is 37.7°C, his heart rate is 110 beats per minute, and his blood pressure is 155/95 mmHg. He appears anxious and diaphoretic but examination is otherwise unremarkable. He admits to cocaine use one hour before the onset of symptoms. What are the appropriate treatments for his condition?

Overview

Cocaine is the second-most-commonly used illicit drug in the U.S. and represents 31% of all ED visits related to substance abuse.^1,2 According to recent survey results, 2.1 million people report recent cocaine use, and 1.6 million engage in cocaine abuse or dependence.² Acute cardiopulmonary complaints are common in individuals who present to the ED after cocaine use, with chest pain being the most frequently reported symptom in 40%.³

Numerous etiologies for cocaine-associated chest pain (CACP) have been discovered, including musculoskeletal pain, pulmonary hypertension, cardiomyopathy, arrhythmias, and endocarditis.⁴ Only 0.5% of patients with aortic dissection over a four-year period had a recent history of cocaine use, making cocaine a rare cause of a rare condition.⁵ Cardiac chest pain remains the most frequent underlying etiology, resulting in the most common complication of myocardial infarction (MI) in up to 6% of patients.^6,7

The ways in which cocaine use can cause myocardial ischemia and MI are multifactorial. A vigorous central sympathomimetic effect, coronary artery vasoconstriction, stimulation of platelets, and enhanced atherosclerosis all lead to a myocardial oxygen supply-demand imbalance.⁸ Other key interactions in the cardiovascular system are displayed in Figure 1. Understanding the role of these mechanisms in CACP is crucial to patient care.

Clinician goals in the management of CACP are to rapidly and accurately exclude life-threatening etiologies; assess the need for urgent acute coronary syndrome (ACS) evaluation; risk-stratify patients and ensure appropriate disposition; normalize the toxic effects of cocaine; treat resultant organ damage; and prevent long-term complications. An algorithm detailing this approach is provided in Figure 2.

click for large version

click for large version

Review of the Data

Diagnostic evaluation. Given potential differences in treatment regimens, it is imperative to differentiate patients who present with CACP from those whose chest pain is not associated with cocaine either by direct questioning or by screening of urine for cocaine metabolites. Once the presence of cocaine has been confirmed, guideline-based evaluation for potential ACS with serial electrocardiograms (ECG), cardiac biomarkers, and close monitoring of cardiac rhythms and hemodynamics is largely similar to standard management of all patients presenting with chest pain, with a few caveats.

Interpretation of the ECG can be challenging in the setting of cocaine. Studies have shown “abnormal” ECGs in 56% to 84% of patients, with many representing early repolarization or left ventricular hypertrophy.^9,10 Likewise, patients with MI are as likely to present with normal or nonspecific ECG findings as with ischemic findings.^7,11 ECG interpretation to diagnose ischemia or infarction in patients with CACP yields a sensitivity of 36% and specificity of 90%.⁷

Creatine kinase (CK), CK-MB fraction, and myoglobin have low specificity for the diagnosis of ischemia, as cocaine can induce skeletal muscle injury and rhabdomyolysis.^9,12 Cardiac troponins demonstrate a superior specificity compared to CK and CK-MB and are thus the preferred cardiac biomarkers in diagnosing cocaine-associated MI.¹²

Initial management and disposition. Patients at high risk for cardiovascular events are generally admitted to a monitored bed.¹³ Immediate reperfusion therapy with primary percutaneous coronary intervention is recommended in patients with ST-elevation MI (STEMI). Treatment with thrombolytic agents is associated with an increased risk of intracerebral hemorrhage and lacks documented efficacy in patients with CACP. Thrombolysis should therefore only be utilized if the diagnosis of STEMI is unequivocal and an experienced cardiac catheterization laboratory is unavailable.^14,15

Patients with unstable angina (UA) or non-ST-elevation MI (NSTEMI) are at higher risk for further cardiac events in a similar manner to those with ACS unrelated to cocaine. These cases might benefit from early cardiac catheterization and revascularization.¹⁶ Because of the increased risk of stent thrombosis in cocaine-users, thought to be due to recidivism, a detailed risk-benefit analysis should be undertaken prior to the implantation of cardiac stents.

Other diagnostic tests, such as stress testing and myocardial imaging, have not shown significant accuracy in diagnosing MI in this setting; moreover, these patients are at low overall risk for cardiac events and mortality. Consequently, an extensive diagnostic evaluation might not be cost-effective.^7,10,13,17 Patients who have CACP without MI have a very low frequency of delayed complications.^3,17 As such, cost-effective evaluation strategies, such as nine- or 12-hour observation periods in a chest pain unit, are appropriate for many of these low- to moderate-risk patients.¹³ For all CACP patients, the most critical post-discharge interventions are cardiac risk modification and cocaine cessation.¹³

Normalizing the toxic effects of cocaine with medications.

Aspirin: While no specific study has been performed in patients with CACP and aspirin, CACP guidelines, based on data supporting ACS guidelines for all patients, recommend administration of full-dose aspirin given its associated reduction in morbidity and mortality.^18,19 Furthermore, given the platelet-stimulating effects of cocaine, using aspirin in this setting seems very reasonable.

Benzodiazepines: CACP guidelines support the use of benzodiazepines early in management to indirectly combat the agitation, hypertension, and tachycardia resulting from the stimulatory effects of cocaine.^18,20 These recommendations are based on several animal and human studies that demonstrate significant reduction in heart rate and systemic arterial pressure with the use of these agents.^21,22

Nitroglycerin: Cardiac catheterization studies have shown reversal of vasoconstriction with administration of nitroglycerin. One study demonstrated a benefit of the drug in 49% of participants.²³ Additional investigation into the benefit of benzodiazepine and nitroglycerin combination therapy revealed mixed results. In one study, lorazepam plus nitroglycerin was found to be more efficacious than nitroglycerin alone.²⁴ In another, however, use of diazepam in combination with nitroglycerin did not show benefit when evaluating pain relief, cardiac dynamics, and left ventricular function.²⁵

Phentolamine: Phentolamine administration has been studied much less in the literature. This nonselective alpha-adrenergic antagonist exerts a dose-dependent reversal of cocaine’s vasoconstrictive properties in monkeys and humans.^26,27 International guidelines for Emergency Cardiovascular Care recommend its use in treatment of cocaine-associated ACS;27 however, the AHA recommends it less strongly.¹⁸

Calcium channel blockers: Calcium channel blockers (CCBs) have not shown promise as first-line agents. While catheterization studies demonstrate the vasodilatory properties of verapamil, larger studies looking at all-cause mortality conclude that CCBs might worsen mortality rates,²⁸ and animal studies indicate an increased risk of seizures.²⁹ At this time, CCBs are recommended only if cardiac symptoms continue after both benzodiazepines and nitroglycerin are administered.¹⁸

The beta-blocker controversy: The use of beta-blockers in patients with CACP remains controversial given the theoretical risk of unopposed alpha-adrenergic activation. Coronary vasospasm, decreased myocardial oxygen delivery, and increased systemic vascular resistance can result from their use.³⁰

Propranolol, a nonselective beta-blocker, was shown in catheterization studies to potentiate the coronary vasoconstriction of cocaine.³¹ Labetalol, a combined alpha/beta-blocker, reduced mean arterial pressure after cocaine administration during cardiac catheterization but did not reverse coronary vasoconstriction.³² This was attributed to the predominating beta greater than alpha blockade at doses administered. The selective beta-1 antagonists esmolol and metoprolol have shown no benefit in CACP.³³ Carvedilol, a combined alpha/beta-blocker with both peripheral and central nervous system activity, has potential to attenuate both physiologic and behavioral response to cocaine, but it has not been well studied in this patient subset.³⁴

click for large version

The 2005 ACC/AHA STEMI guidelines recommended against beta-blockers in the setting of STEMI precipitated by cocaine use due to the potential of exacerbating coronary vasoconstriction.³⁵ The 2007 ACC/AHA UA/NSTEMI guidelines stated that the use of a combined alpha/beta-blocker in patients with cocaine-induced ACS may be reasonable for patients with hypertension or tachycardia if pre-treated with a vasodilator.¹⁹ The 2008 ACC/AHA guidelines on the management of cocaine-related chest pain and MI recommended against the use of beta-blockers in the acute setting given the low incidence of cocaine-related MI and death.¹⁸

In a more recent study, Dattilo et al showed that beta-blockers administered to patients admitted with positive urine toxicology for cocaine significantly reduced MI and in-hospital mortality. Reduction of MI was of borderline significance in those admitted with a chief complaint of chest pain.³⁶ Limitations of this study include unknown time of cocaine ingestion, lack of follow-up on discharge mortality, and a small sample size of 348 patients lacking statistical power.

Another retrospective cohort study examined patients admitted with chest pain and urine toxicology positive for cocaine and found that beta-blocker administration during hospitalization was not associated with increased incident mortality. Further, after a mean follow-up of 2.5 years, there was a statistically significant decrease in cardiovascular death.³⁷ Drawbacks of this study included an older patient population, greater proportion of coronary artery disease, and higher follow-up of cardiovascular mortality rates than in previous studies, suggesting this subset might have received greater benefit from beta-blockers as a result of these characteristics.

The 2008 ACC/AHA guidelines instruct individualized consideration of the risk/benefit ratio for beta-blocker use in patients with CACP given the high rate of recidivism in cocaine abusers. The strongest indication is given to those with documented MI, left ventricular systolic dysfunction, or ventricular arrhythmias.¹⁸

It is important to note that these recommendations are based on cardiac catheterization laboratory studies, case reports, retrospective analyses, and animal experiments. No prospective controlled trials evaluating the role of beta-blockers in CACP and MI exist, and no trials regarding therapies to improve outcomes of patients sustaining a cocaine-associated MI have been reported.¹⁸

Back to the Case

This patient was experiencing cocaine-associated chest pain, which was confirmed with positive urine toxicology. Initial diagnostic workup with basic laboratory studies and cardiac biomarkers showed mild elevation in CK with troponin levels within normal limits. His ECG showed changes consistent with left ventricular hypertrophy. Chest radiograph was unremarkable.

He received aspirin, benzodiazepines, and nitroglycerin with normalization of vital signs, as well as subjective improvement in chest pain and anxiety. He was deemed to be at low risk for potential cardiac complications; thus, further cardiac testing was not pursued. Rather, he was admitted to an overnight observation unit with telemetry monitoring, where his chest pain did not recur.

He was seen in consultation with social work staff who arranged for drug abuse counseling after discharge. Given the uncertainty of relapse to cocaine use, as well as lack of known cardiac risk factors, he was not discharged on any new medications.

Bottom Line

The treatment of CACP includes normalizing the toxic systemic effects of the drug and minimizing the direct ischemic damage to the myocardium. Management varies slightly from traditional chest pain algorithms and includes benzodiazepines as well as antiplatelet agents and vasodilators to achieve this goal. Initial therapy with beta-blockers remains undefined and is largely discouraged in the acute setting. The role of beta-blockade upon discharge, however, can be beneficial in specific populations, especially those found to have underlying coronary disease.

Dr. Houchens and Dr. Czarnik are clinical instructors and Dr. Mack is a clinical lecturer at the University of Michigan Health System in Ann Arbor.

References

1. Hughes A, Sathe N, Spagnola K. State Estimates of Substance Use from the 2005-2006 National Surveys on Drug Use and Health. DHHS Publication No. SMA 08-4311, NSDUH Series H-33. Rockville, MD: Substance Abuse and Mental Health Services Administration, Office of Applied Studies; 2008.
2. Volkow ND. Cocaine: Abuse and Addiction. National Institute on Drug Abuse. Washington, DC: U.S. Department of Health and Human Services; 2009.
3. Brody SL, Slovis CM, Wrenn KD. Cocaine-related medical problems: consecutive series of 233 patients. Am J Med. 1990;88:325-331.
4. Levis JT, Garmel GM. Cocaine-associated chest pain. Emerg Med Clin North Am. 2005;23:1083-1103.
5. Eagle KA, Isselbacher EM, DeSanctis RW. Cocaine-related aortic dissection in perspective. Circulation. 2002;105:1529-1530.
6. Feldman JA, Fish SS, Beshansky JR, Griffith JL, Woolard RH, Selker HP. Acute cardiac ischemia in patients with cocaine-associated complaints: results of a multicenter trial. Ann Emerg Med. 2000;36:469-476.
7. Hollander JE, Hoffman RS, Gennis P, et al. Prospective multicenter evaluation of cocaine associated chest pain. Cocaine Associated Chest Pain (COCHPA) Study Group. Acad Emerg Med. 1994;1:330-339.
8. Schwartz BG, Rezkalla S, Kloner RA. Cardiovascular effects of cocaine. Circulation. 2010;122:2558-2569.
9. Gitter MJ, Goldsmith SR, Dunbar DN, et al. Cocaine and chest pain: clinical features and outcomes of patients hospitalized to rule out myocardial infarction. Ann Intern Med. 1991;115:277-282.
10. Amin M, Gabelman G, Karpel J, et al. Acute myocardial infarction and chest pain syndromes after cocaine use. Am J Cardiol. 1990;66:1434-1437.
11. Tokarski GF, Paganussi P, Urbanski R, et al. An evaluation of cocaine-induced chest pain. Ann Emerg Med. 1990;19:1088-1092.
12. Hollander JE, Levitt MA, Young GP, Briglia E, Wetli CV, Gawad Y. Effect of recent cocaine use on the specificity of cardiac markers for diagnosis of acute myocardial infarction. Am Heart J. 1998;135(2 Pt 1):245-252.
13. Weber JE, Shofer FS, Larkin GL, Kalaria AS, Hollander JE. Validation of a brief observation period for patients with cocaine-associated chest pain. N Engl J Med. 2003;348:510-517.
14. Hahn IH, Hoffman RS. Diagnosis and treatment of acute myocardial infarction: cocaine use and acute myocardial infarction. Emerg Med Clin North Am. 2001;19(2):1-18.
15. Hoffman RS, Hollander JE. Evaluation of patients with chest pain after cocaine use. Crit Care Clin. 1997;13:809-828. Cannon CP, Weintraub WS, Demopoulos LA, et al.
16. Comparison of early invasive and conservative strategies in patients with unstable coronary syndromes treated with the glycoprotein IIb/IIIa inhibitor tirofiban. N Engl J Med. 2001;344:1879-1887.
17. Hollander JE, Hoffman RS. Cocaine-induced myocardial infarction: an analysis and review of the literature. J Emerg Med. 1992;10:169-177.
18. McCord J, Jneid H, Hollander JE, et al. Management of cocaine-associated chest pain and myocardial infarction: a scientific statement from the American Heart Association Acute Cardiac Care Committee of the Council on Clinical Cardiology. Circulation. 2008;117:1897-1907.
19. Anderson JL, Adams CD, Antman EM, et al. ACC/AHA 2007 guidelines for the management of patients with unstable angina/non-ST-Elevation myocardial infarction: a report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines (Writing Committee to Revise the 2002 Guidelines for the Management of Patients With Unstable Angina/Non-ST-Elevation Myocardial Infarction) developed in collaboration with the American College of Emergency Physicians, the Society for Cardiovascular Angiography and Interventions, and the Society of Thoracic Surgeons endorsed by the American Association of Cardiovascular and Pulmonary Rehabilitation and the Society for Academic Emergency Medicine. J Am Coll Cardiol. 2007;50:E1-E157.
20. Hollander JE. Management of cocaine-associated myocardial ischemia. N Engl J Med. 1995;333:1267-1272.
21. Brubacher JR, Hoffman RS. Cocaine toxicity. Top Emerg Med. 1997;19(4):1-16.
22. Catavas JD, Waters IW. Acute cocaine intoxication in the conscious dog: studies on the mechanism of lethality. J Pharmacol Exp Ther. 1981;217:350-356.
23. Hollander JE, Hoffman RS, Gennis P, et al. Nitroglycerin in the treatment of cocaine associated chest pain—clinical safety and efficacy. J Toxicol Clin Toxicol. 1994;32(3): 243-256.
24. Honderick T, Williams D, Seaberg D, Wears R. A prospective, randomized, controlled trial of benzodiazepines and nitroglycerin or nitroglycerin alone in the treatment of cocaine-associated acute coronary syndromes. Am J Emerg Med. 2003;21(1):39-42.
25. Baumann BM, Perrone J, Hornig SE, Shofer FS, Hollander JE. Randomized, double-blind, placebo-controlled trial of diazepam, nitroglycerin, or both for treatment of patients with potential cocaine-associated acute coronary syndromes. Acad Emerg Med. 2000;7:878-885.
26. Schindler CW, Tella SR, Goldberg SR. Adrenoceptor mechanisms in the cardiovascular effects of cocaine in conscious squirrel monkeys. Life Sci. 1992;51(9):653-660.
27. Lange RA, Cigarroa RG, Yancy CW Jr., et al. Cocaine-induced coronary-artery vasoconstriction. N Engl J Med. 1989;321(23):1557-1562.
28. Furberg CD, Psaty BM, Meyer JV. Nifedipine. Dose-related increase in mortality in patients with coronary heart disease. Circulation. 1995;92:1326-1331.
29. Derlet RW, Albertson TE. Potentiation of cocaine toxicity with calcium channel blockers. Am J Emerg Med. 1989;7:464-468.
30. Lange RA, Hillis LD. Cardiovascular complications of cocaine use. N Engl J Med. 2001;345:351-358.
31. Lange RA, Cigarroa RG, Flores ED, et al. Potentiation of cocaine-induced coronary vasoconstriction by beta-adrenergic blockade. Ann Intern Med. 1990;112:897-903.
32. Boehrer JD, Moliterno DJ, Willard JE, Hillis LD, Lange RA. Influence of labetalol on cocaine-induced coronary vasoconstriction in humans. Am J Med. 1993;94:608-610.
33. Sand IC, Brody SL, Wrenn KD, Slovis CM. Experience with esmolol for the treatment of cocaine-associated cardiovascular complications. Am J Emerg Med. 1991;9:161-163.
34. Sofuoglo M, Brown S, Babb DA, Pentel PR, Hatsukami DK. Carvedilol affects the physiological and behavioral response to smoked cocaine in humans. Drug Alcohol Depend. 2000;60:69-76.
35. Antman EM, Anbe DT, Armstrong PW, et al. ACC/AHA guidelines for the management of patients with ST-elevation myocardial infarction: a report of the American College of Cardiology/American Heart Association Task Force of Practice Guidelines (Committee to Revise the 1999 Guidelines for the Management of patients with acute myocardial infarction). J Am Coll Cardiol. 2004;44:E1-E211.
36. Dattilo PB, Hailpern SM, Fearon K, Sohal D, Nordin C. Beta-blockers are associated with reduced risk of myocardial infarction after cocaine use. Ann Emerg Med. 2008;51:117-125.
37. Rangel C, Shu RG, Lazar LD, Vittinghoff E, Hsue PY, Marcus GM. Beta-blockers for chest pain associated with recent cocaine use. Arch Intern Med. 2010;170:874-879.

click for large version

Case

A 38-year-old man with a history of tobacco use presents to the emergency department complaining of constant substernal chest pain for three hours. His temperature is 37.7°C, his heart rate is 110 beats per minute, and his blood pressure is 155/95 mmHg. He appears anxious and diaphoretic but examination is otherwise unremarkable. He admits to cocaine use one hour before the onset of symptoms. What are the appropriate treatments for his condition?

Overview

Cocaine is the second-most-commonly used illicit drug in the U.S. and represents 31% of all ED visits related to substance abuse.^1,2 According to recent survey results, 2.1 million people report recent cocaine use, and 1.6 million engage in cocaine abuse or dependence.² Acute cardiopulmonary complaints are common in individuals who present to the ED after cocaine use, with chest pain being the most frequently reported symptom in 40%.³

Numerous etiologies for cocaine-associated chest pain (CACP) have been discovered, including musculoskeletal pain, pulmonary hypertension, cardiomyopathy, arrhythmias, and endocarditis.⁴ Only 0.5% of patients with aortic dissection over a four-year period had a recent history of cocaine use, making cocaine a rare cause of a rare condition.⁵ Cardiac chest pain remains the most frequent underlying etiology, resulting in the most common complication of myocardial infarction (MI) in up to 6% of patients.^6,7

The ways in which cocaine use can cause myocardial ischemia and MI are multifactorial. A vigorous central sympathomimetic effect, coronary artery vasoconstriction, stimulation of platelets, and enhanced atherosclerosis all lead to a myocardial oxygen supply-demand imbalance.⁸ Other key interactions in the cardiovascular system are displayed in Figure 1. Understanding the role of these mechanisms in CACP is crucial to patient care.

Clinician goals in the management of CACP are to rapidly and accurately exclude life-threatening etiologies; assess the need for urgent acute coronary syndrome (ACS) evaluation; risk-stratify patients and ensure appropriate disposition; normalize the toxic effects of cocaine; treat resultant organ damage; and prevent long-term complications. An algorithm detailing this approach is provided in Figure 2.

click for large version

click for large version

Review of the Data

Diagnostic evaluation. Given potential differences in treatment regimens, it is imperative to differentiate patients who present with CACP from those whose chest pain is not associated with cocaine either by direct questioning or by screening of urine for cocaine metabolites. Once the presence of cocaine has been confirmed, guideline-based evaluation for potential ACS with serial electrocardiograms (ECG), cardiac biomarkers, and close monitoring of cardiac rhythms and hemodynamics is largely similar to standard management of all patients presenting with chest pain, with a few caveats.

Interpretation of the ECG can be challenging in the setting of cocaine. Studies have shown “abnormal” ECGs in 56% to 84% of patients, with many representing early repolarization or left ventricular hypertrophy.^9,10 Likewise, patients with MI are as likely to present with normal or nonspecific ECG findings as with ischemic findings.^7,11 ECG interpretation to diagnose ischemia or infarction in patients with CACP yields a sensitivity of 36% and specificity of 90%.⁷

Creatine kinase (CK), CK-MB fraction, and myoglobin have low specificity for the diagnosis of ischemia, as cocaine can induce skeletal muscle injury and rhabdomyolysis.^9,12 Cardiac troponins demonstrate a superior specificity compared to CK and CK-MB and are thus the preferred cardiac biomarkers in diagnosing cocaine-associated MI.¹²

Initial management and disposition. Patients at high risk for cardiovascular events are generally admitted to a monitored bed.¹³ Immediate reperfusion therapy with primary percutaneous coronary intervention is recommended in patients with ST-elevation MI (STEMI). Treatment with thrombolytic agents is associated with an increased risk of intracerebral hemorrhage and lacks documented efficacy in patients with CACP. Thrombolysis should therefore only be utilized if the diagnosis of STEMI is unequivocal and an experienced cardiac catheterization laboratory is unavailable.^14,15

Patients with unstable angina (UA) or non-ST-elevation MI (NSTEMI) are at higher risk for further cardiac events in a similar manner to those with ACS unrelated to cocaine. These cases might benefit from early cardiac catheterization and revascularization.¹⁶ Because of the increased risk of stent thrombosis in cocaine-users, thought to be due to recidivism, a detailed risk-benefit analysis should be undertaken prior to the implantation of cardiac stents.

Other diagnostic tests, such as stress testing and myocardial imaging, have not shown significant accuracy in diagnosing MI in this setting; moreover, these patients are at low overall risk for cardiac events and mortality. Consequently, an extensive diagnostic evaluation might not be cost-effective.^7,10,13,17 Patients who have CACP without MI have a very low frequency of delayed complications.^3,17 As such, cost-effective evaluation strategies, such as nine- or 12-hour observation periods in a chest pain unit, are appropriate for many of these low- to moderate-risk patients.¹³ For all CACP patients, the most critical post-discharge interventions are cardiac risk modification and cocaine cessation.¹³

Normalizing the toxic effects of cocaine with medications.

Aspirin: While no specific study has been performed in patients with CACP and aspirin, CACP guidelines, based on data supporting ACS guidelines for all patients, recommend administration of full-dose aspirin given its associated reduction in morbidity and mortality.^18,19 Furthermore, given the platelet-stimulating effects of cocaine, using aspirin in this setting seems very reasonable.

Benzodiazepines: CACP guidelines support the use of benzodiazepines early in management to indirectly combat the agitation, hypertension, and tachycardia resulting from the stimulatory effects of cocaine.^18,20 These recommendations are based on several animal and human studies that demonstrate significant reduction in heart rate and systemic arterial pressure with the use of these agents.^21,22

Nitroglycerin: Cardiac catheterization studies have shown reversal of vasoconstriction with administration of nitroglycerin. One study demonstrated a benefit of the drug in 49% of participants.²³ Additional investigation into the benefit of benzodiazepine and nitroglycerin combination therapy revealed mixed results. In one study, lorazepam plus nitroglycerin was found to be more efficacious than nitroglycerin alone.²⁴ In another, however, use of diazepam in combination with nitroglycerin did not show benefit when evaluating pain relief, cardiac dynamics, and left ventricular function.²⁵

Phentolamine: Phentolamine administration has been studied much less in the literature. This nonselective alpha-adrenergic antagonist exerts a dose-dependent reversal of cocaine’s vasoconstrictive properties in monkeys and humans.^26,27 International guidelines for Emergency Cardiovascular Care recommend its use in treatment of cocaine-associated ACS;27 however, the AHA recommends it less strongly.¹⁸

Calcium channel blockers: Calcium channel blockers (CCBs) have not shown promise as first-line agents. While catheterization studies demonstrate the vasodilatory properties of verapamil, larger studies looking at all-cause mortality conclude that CCBs might worsen mortality rates,²⁸ and animal studies indicate an increased risk of seizures.²⁹ At this time, CCBs are recommended only if cardiac symptoms continue after both benzodiazepines and nitroglycerin are administered.¹⁸

The beta-blocker controversy: The use of beta-blockers in patients with CACP remains controversial given the theoretical risk of unopposed alpha-adrenergic activation. Coronary vasospasm, decreased myocardial oxygen delivery, and increased systemic vascular resistance can result from their use.³⁰

Propranolol, a nonselective beta-blocker, was shown in catheterization studies to potentiate the coronary vasoconstriction of cocaine.³¹ Labetalol, a combined alpha/beta-blocker, reduced mean arterial pressure after cocaine administration during cardiac catheterization but did not reverse coronary vasoconstriction.³² This was attributed to the predominating beta greater than alpha blockade at doses administered. The selective beta-1 antagonists esmolol and metoprolol have shown no benefit in CACP.³³ Carvedilol, a combined alpha/beta-blocker with both peripheral and central nervous system activity, has potential to attenuate both physiologic and behavioral response to cocaine, but it has not been well studied in this patient subset.³⁴

click for large version

The 2005 ACC/AHA STEMI guidelines recommended against beta-blockers in the setting of STEMI precipitated by cocaine use due to the potential of exacerbating coronary vasoconstriction.³⁵ The 2007 ACC/AHA UA/NSTEMI guidelines stated that the use of a combined alpha/beta-blocker in patients with cocaine-induced ACS may be reasonable for patients with hypertension or tachycardia if pre-treated with a vasodilator.¹⁹ The 2008 ACC/AHA guidelines on the management of cocaine-related chest pain and MI recommended against the use of beta-blockers in the acute setting given the low incidence of cocaine-related MI and death.¹⁸

In a more recent study, Dattilo et al showed that beta-blockers administered to patients admitted with positive urine toxicology for cocaine significantly reduced MI and in-hospital mortality. Reduction of MI was of borderline significance in those admitted with a chief complaint of chest pain.³⁶ Limitations of this study include unknown time of cocaine ingestion, lack of follow-up on discharge mortality, and a small sample size of 348 patients lacking statistical power.

Another retrospective cohort study examined patients admitted with chest pain and urine toxicology positive for cocaine and found that beta-blocker administration during hospitalization was not associated with increased incident mortality. Further, after a mean follow-up of 2.5 years, there was a statistically significant decrease in cardiovascular death.³⁷ Drawbacks of this study included an older patient population, greater proportion of coronary artery disease, and higher follow-up of cardiovascular mortality rates than in previous studies, suggesting this subset might have received greater benefit from beta-blockers as a result of these characteristics.

The 2008 ACC/AHA guidelines instruct individualized consideration of the risk/benefit ratio for beta-blocker use in patients with CACP given the high rate of recidivism in cocaine abusers. The strongest indication is given to those with documented MI, left ventricular systolic dysfunction, or ventricular arrhythmias.¹⁸

It is important to note that these recommendations are based on cardiac catheterization laboratory studies, case reports, retrospective analyses, and animal experiments. No prospective controlled trials evaluating the role of beta-blockers in CACP and MI exist, and no trials regarding therapies to improve outcomes of patients sustaining a cocaine-associated MI have been reported.¹⁸

Back to the Case

This patient was experiencing cocaine-associated chest pain, which was confirmed with positive urine toxicology. Initial diagnostic workup with basic laboratory studies and cardiac biomarkers showed mild elevation in CK with troponin levels within normal limits. His ECG showed changes consistent with left ventricular hypertrophy. Chest radiograph was unremarkable.

He received aspirin, benzodiazepines, and nitroglycerin with normalization of vital signs, as well as subjective improvement in chest pain and anxiety. He was deemed to be at low risk for potential cardiac complications; thus, further cardiac testing was not pursued. Rather, he was admitted to an overnight observation unit with telemetry monitoring, where his chest pain did not recur.

He was seen in consultation with social work staff who arranged for drug abuse counseling after discharge. Given the uncertainty of relapse to cocaine use, as well as lack of known cardiac risk factors, he was not discharged on any new medications.

Bottom Line

The treatment of CACP includes normalizing the toxic systemic effects of the drug and minimizing the direct ischemic damage to the myocardium. Management varies slightly from traditional chest pain algorithms and includes benzodiazepines as well as antiplatelet agents and vasodilators to achieve this goal. Initial therapy with beta-blockers remains undefined and is largely discouraged in the acute setting. The role of beta-blockade upon discharge, however, can be beneficial in specific populations, especially those found to have underlying coronary disease.

Dr. Houchens and Dr. Czarnik are clinical instructors and Dr. Mack is a clinical lecturer at the University of Michigan Health System in Ann Arbor.

References

1. Hughes A, Sathe N, Spagnola K. State Estimates of Substance Use from the 2005-2006 National Surveys on Drug Use and Health. DHHS Publication No. SMA 08-4311, NSDUH Series H-33. Rockville, MD: Substance Abuse and Mental Health Services Administration, Office of Applied Studies; 2008.
2. Volkow ND. Cocaine: Abuse and Addiction. National Institute on Drug Abuse. Washington, DC: U.S. Department of Health and Human Services; 2009.
3. Brody SL, Slovis CM, Wrenn KD. Cocaine-related medical problems: consecutive series of 233 patients. Am J Med. 1990;88:325-331.
4. Levis JT, Garmel GM. Cocaine-associated chest pain. Emerg Med Clin North Am. 2005;23:1083-1103.
5. Eagle KA, Isselbacher EM, DeSanctis RW. Cocaine-related aortic dissection in perspective. Circulation. 2002;105:1529-1530.
6. Feldman JA, Fish SS, Beshansky JR, Griffith JL, Woolard RH, Selker HP. Acute cardiac ischemia in patients with cocaine-associated complaints: results of a multicenter trial. Ann Emerg Med. 2000;36:469-476.
7. Hollander JE, Hoffman RS, Gennis P, et al. Prospective multicenter evaluation of cocaine associated chest pain. Cocaine Associated Chest Pain (COCHPA) Study Group. Acad Emerg Med. 1994;1:330-339.
8. Schwartz BG, Rezkalla S, Kloner RA. Cardiovascular effects of cocaine. Circulation. 2010;122:2558-2569.
9. Gitter MJ, Goldsmith SR, Dunbar DN, et al. Cocaine and chest pain: clinical features and outcomes of patients hospitalized to rule out myocardial infarction. Ann Intern Med. 1991;115:277-282.
10. Amin M, Gabelman G, Karpel J, et al. Acute myocardial infarction and chest pain syndromes after cocaine use. Am J Cardiol. 1990;66:1434-1437.
11. Tokarski GF, Paganussi P, Urbanski R, et al. An evaluation of cocaine-induced chest pain. Ann Emerg Med. 1990;19:1088-1092.
12. Hollander JE, Levitt MA, Young GP, Briglia E, Wetli CV, Gawad Y. Effect of recent cocaine use on the specificity of cardiac markers for diagnosis of acute myocardial infarction. Am Heart J. 1998;135(2 Pt 1):245-252.
13. Weber JE, Shofer FS, Larkin GL, Kalaria AS, Hollander JE. Validation of a brief observation period for patients with cocaine-associated chest pain. N Engl J Med. 2003;348:510-517.
14. Hahn IH, Hoffman RS. Diagnosis and treatment of acute myocardial infarction: cocaine use and acute myocardial infarction. Emerg Med Clin North Am. 2001;19(2):1-18.
15. Hoffman RS, Hollander JE. Evaluation of patients with chest pain after cocaine use. Crit Care Clin. 1997;13:809-828. Cannon CP, Weintraub WS, Demopoulos LA, et al.
16. Comparison of early invasive and conservative strategies in patients with unstable coronary syndromes treated with the glycoprotein IIb/IIIa inhibitor tirofiban. N Engl J Med. 2001;344:1879-1887.
17. Hollander JE, Hoffman RS. Cocaine-induced myocardial infarction: an analysis and review of the literature. J Emerg Med. 1992;10:169-177.
18. McCord J, Jneid H, Hollander JE, et al. Management of cocaine-associated chest pain and myocardial infarction: a scientific statement from the American Heart Association Acute Cardiac Care Committee of the Council on Clinical Cardiology. Circulation. 2008;117:1897-1907.
19. Anderson JL, Adams CD, Antman EM, et al. ACC/AHA 2007 guidelines for the management of patients with unstable angina/non-ST-Elevation myocardial infarction: a report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines (Writing Committee to Revise the 2002 Guidelines for the Management of Patients With Unstable Angina/Non-ST-Elevation Myocardial Infarction) developed in collaboration with the American College of Emergency Physicians, the Society for Cardiovascular Angiography and Interventions, and the Society of Thoracic Surgeons endorsed by the American Association of Cardiovascular and Pulmonary Rehabilitation and the Society for Academic Emergency Medicine. J Am Coll Cardiol. 2007;50:E1-E157.
20. Hollander JE. Management of cocaine-associated myocardial ischemia. N Engl J Med. 1995;333:1267-1272.
21. Brubacher JR, Hoffman RS. Cocaine toxicity. Top Emerg Med. 1997;19(4):1-16.
22. Catavas JD, Waters IW. Acute cocaine intoxication in the conscious dog: studies on the mechanism of lethality. J Pharmacol Exp Ther. 1981;217:350-356.
23. Hollander JE, Hoffman RS, Gennis P, et al. Nitroglycerin in the treatment of cocaine associated chest pain—clinical safety and efficacy. J Toxicol Clin Toxicol. 1994;32(3): 243-256.
24. Honderick T, Williams D, Seaberg D, Wears R. A prospective, randomized, controlled trial of benzodiazepines and nitroglycerin or nitroglycerin alone in the treatment of cocaine-associated acute coronary syndromes. Am J Emerg Med. 2003;21(1):39-42.
25. Baumann BM, Perrone J, Hornig SE, Shofer FS, Hollander JE. Randomized, double-blind, placebo-controlled trial of diazepam, nitroglycerin, or both for treatment of patients with potential cocaine-associated acute coronary syndromes. Acad Emerg Med. 2000;7:878-885.
26. Schindler CW, Tella SR, Goldberg SR. Adrenoceptor mechanisms in the cardiovascular effects of cocaine in conscious squirrel monkeys. Life Sci. 1992;51(9):653-660.
27. Lange RA, Cigarroa RG, Yancy CW Jr., et al. Cocaine-induced coronary-artery vasoconstriction. N Engl J Med. 1989;321(23):1557-1562.
28. Furberg CD, Psaty BM, Meyer JV. Nifedipine. Dose-related increase in mortality in patients with coronary heart disease. Circulation. 1995;92:1326-1331.
29. Derlet RW, Albertson TE. Potentiation of cocaine toxicity with calcium channel blockers. Am J Emerg Med. 1989;7:464-468.
30. Lange RA, Hillis LD. Cardiovascular complications of cocaine use. N Engl J Med. 2001;345:351-358.
31. Lange RA, Cigarroa RG, Flores ED, et al. Potentiation of cocaine-induced coronary vasoconstriction by beta-adrenergic blockade. Ann Intern Med. 1990;112:897-903.
32. Boehrer JD, Moliterno DJ, Willard JE, Hillis LD, Lange RA. Influence of labetalol on cocaine-induced coronary vasoconstriction in humans. Am J Med. 1993;94:608-610.
33. Sand IC, Brody SL, Wrenn KD, Slovis CM. Experience with esmolol for the treatment of cocaine-associated cardiovascular complications. Am J Emerg Med. 1991;9:161-163.
34. Sofuoglo M, Brown S, Babb DA, Pentel PR, Hatsukami DK. Carvedilol affects the physiological and behavioral response to smoked cocaine in humans. Drug Alcohol Depend. 2000;60:69-76.
35. Antman EM, Anbe DT, Armstrong PW, et al. ACC/AHA guidelines for the management of patients with ST-elevation myocardial infarction: a report of the American College of Cardiology/American Heart Association Task Force of Practice Guidelines (Committee to Revise the 1999 Guidelines for the Management of patients with acute myocardial infarction). J Am Coll Cardiol. 2004;44:E1-E211.
36. Dattilo PB, Hailpern SM, Fearon K, Sohal D, Nordin C. Beta-blockers are associated with reduced risk of myocardial infarction after cocaine use. Ann Emerg Med. 2008;51:117-125.
37. Rangel C, Shu RG, Lazar LD, Vittinghoff E, Hsue PY, Marcus GM. Beta-blockers for chest pain associated with recent cocaine use. Arch Intern Med. 2010;170:874-879.