Surveys are common in medical research. Although survey research may be subject to inherent self-report bias, surveys have a great impact on policies and practices in medicine, often forming the basis for recommendations or new guidelines.^1,2 To interpret and use survey research results, clinicians should be familiar with key elements involved in the creation and validation of surveys.

The purpose of this article is to provide readers with a basic framework for evaluating surveys to allow them to be more informed as consumers of survey research.

IMPORTANT TOOLS IN MEDICAL RESEARCH

Surveys are important tools for answering questions on topics that are difficult to assess using other methods.³ They allow us to gather data systematically from subjects by asking questions, in order to make inferences about a larger population.^3,4 Clinicians use surveys to explore the opinions, beliefs, and perceptions of a group, or to investigate physician practice patterns and adherence to clinical guidelines. They may also use surveys to better understand why patients are not engaging in recommended behavioral or lifestyle changes.

Survey methods include interviews (in person, by phone) and questionnaires (paper-and-pencil, e-mailed, online).⁴

A well-constructed, validated survey can provide powerful data that may influence clinical practice, guide future research development, or drive the development and provision of needed programs and services. Surveys have the potential to transform the ways in which we think about and practice medicine.

READER BEWARE

While survey research in health care appears to have grown exponentially, the quality of reported survey research has not necessarily increased over time.

For consumers of survey research, the adage “reader beware” is apt. Although a considerable number of studies have examined the effects of survey methodology on the validity, reliability, and generalizability of the results,⁴ medical journals differ in their requirements for reporting survey methods.

In an analysis of 117 articles, Bennett et al³ found that more than 80% did not fully describe the survey development process or pretesting methods. They also found limited guidance and lack of consensus about the best way to report survey research. Of 95 surveys requiring scoring, 66% did not report scoring practices.

Duffett et al⁵ noted that of 127 critical care medicine surveys, only 36% had been pretested or pilot-tested, and half of all surveys reviewed did not include participant demographics or included only minimal information.

Because journal reporting practices differ, physicians may be unaware of the steps involved in survey construction and validation. Knowledge of these steps is helpful not only in constructing surveys but also in assessing published articles that used survey research.

LIMITATIONS OF SURVEY RESEARCH

Indirect measures of attitudes and behaviors

Surveys that rely on participants’ self-reports of behaviors, attitudes, beliefs, or actions are indirect measures and are susceptible to self-report and social-desirability biases. Participants may overestimate their own expertise or knowledge in self-report surveys. They may wish to reduce embarrassment⁶ or answer in ways that would make them “look better,”⁷ resulting in social-desirability bias. These issues need to be mentioned in the limitations section in papers reporting survey research.

Questions and response choices

The data derived from surveys are only as good as the questions that are asked.⁸ Stone⁹ noted that questions should be intelligible, unambiguous, and unbiased. If respondents do not comprehend questions as researchers intended, if questionnaire response choices are inadequate, or if questions trigger unintended emotional responses,^10–14 researchers may unwittingly introduce error, which will affect the validity of results. Even seemingly objective questions, such as those related to clinical algorithm use, practice patterns, or equipment available to hospital staff, may be interpreted differently by different respondents.

In their eagerness to launch a survey, clinician researchers may not realize that it must be carefully constructed. A focus on question development and validation is critical, as the questions determine the quality of the data derived from the survey.⁸ Even the position of the question or answer in the survey can affect how participants respond,¹⁵ as they may be guided to a response choice by preceding questions.¹⁶

WHAT DO YOU NEED TO KNOW ABOUT ASSESSING SURVEY RESEARCH?

What follows are questions and a basic framework that can be used to evaluate published survey research. Recommendations are based on the work of survey scientists,^{4,7,10,14,15,17,18} survey researchers in medicine and the social sciences, and national standards for test and questionnaire construction and validation (Table 1).^4,19,20

Who created the survey? How did they do it?

How the survey was created should be sufficiently described to allow readers to judge the adequacy of instrument development.^3–5 It is generally recommended that feedback from multiple sources be solicited during survey creation. Both questionnaire-design experts and subject-matter experts are considered critical in the process.⁴

What question was the survey designed to answer?

Is the objective of the study articulated in the paper? ^3,20 To judge survey research, readers need to know if the survey appears to adequately address the research question or questions and the objectives of the study in terms of methods used.⁴

Instrument pretesting and field testing are considered best practices by the American Association for Public Opinion Research, a professional organization for US survey scientists.⁴

Pretesting can include cognitive interviewing, the use of questionnaire appraisal tools, and hybrid methods, all of which are aimed at addressing validity issues.²¹ Pretesting with a group of participants similar to the target population allows for assessment of item ambiguity, instrument ease of use, adequacy of response categories (response choices), and time to completion.^4,12

Cognitive interviewing is designed to explore respondents’ comprehension of questions, response processes, and decision processes governing how they answer questions.^4,7,10,11 In cognitive interviewing, respondents are generally interviewed one on one. Techniques vary, but typically include “think alouds” (in which a respondent is asked to verbalize thoughts while responding to questions) and “verbal probing” (in which the respondent answers a question, then is asked follow-up questions as the interviewer probes for information related to the response choice or question itself).⁷ These techniques can provide evidence that researchers are actually measuring what they set out to measure and not an unrelated construct.^4,19

Field testing of a survey under realistic conditions can help to uncover problems in administration, such as issues in standardization of key procedures, and to ensure that the survey was administered as the researchers intended.^21,22 Field testing is vital before phone or in-person interviews to ensure standardization of any critical procedures. Pilot testing in a sample similar to the intended population allows for further refinement, with deletion of problem items, before the survey is launched.¹⁵

Because even “objective” questions can be somewhat subjective, all research surveys should go through some type of pretesting.^4,21 Based on the results of pretesting and field testing, surveys should then be revised before launch.^4,21 If an article on a self-report survey makes no mention of survey validation steps, readers may well question the validity of the results.

Are the survey questions and response choices understandable?

Is the meaning of each question unambiguous? Is the reading level appropriate for the sample population (a critical consideration in patient surveys)? Do any of the items actually ask two different questions?¹³ An example would be: “Was the representative courteous and prompt?” as it is possible to be courteous, but not prompt, and vice versa. If so, respondents may be confused or frustrated in attempting to answer it. If a rating scale is used throughout the questionnaire, are the anchors appropriate? For example, a question may be written in such a way that respondents want to answer “yes/no” or “agree/disagree,” but the scale used may include response options such as “poor,” “marginal,” “good,” and “excellent.” Items with Likert-response formats are commonly used in self-report surveys and allow participants to respond to a statement by choosing from a range of responses (eg, strongly disagree to strongly agree), often spaced horizontally under a line.

It is recommended that surveys also include options for answers beyond the response choices provided,²⁰ such as comment boxes or fill-in-the-blank items. Surveys with a closed-response format may constrain the quality of data collected because investigators may not foresee all possible answers. Surveys need to be available for review either within the article itself, in an appendix, or as supplementary material that is available elsewhere.

Does the sample appear to be appropriate?

Articles that report the results of surveys should describe the target population, the sample design, and, in a demographic table, respondents and nonrespondents. To judge appropriateness, several questions can be asked regarding sampling:

Target population. Is the population of interest (ie, the target population) described, including regional demographics, if applicable? The relationship between the sample and the target population is important, as a nonrepresentative sample may result in misleading conclusions about the population of interest.

Sampling frame. Who had an opportunity to participate in the survey? At its simplest, the sampling frame establishes who (or what, in the case of institutions) should be included within the sample. This is typically a list of elements (Groves et al⁴) that acts to “frame” or define the sample to be selected. Where the target population may be all academic internal medicine physicians in the United States, the sampling frame may be all male and female US physicians who are members of particular internal medicine professional organizations, identified by their directory email addresses.

Sample design. How was the sample actually selected?⁴ For example, did investigators use a convenience sample of colleagues at other institutions or use a stratified random sample, ensuring adequate representation of respondents with certain characteristics?

Description of respondents. How is the sample of respondents described? Are demographic features reported, including statistics on regional or national representativeness?⁵ Does the sample of survey respondents appear to be representative of the researcher’s population of interest (ie, the target population)?^3,23 If not, is this adequately described in the limitations section? Although outcomes will not be available on nonrespondents, demographic and baseline data often are available and should be reported. Are there systematic differences between respondents and nonrespondents?

Was the response rate adequate?

Was the response rate adequate, given the number of participants initially recruited? If the response rate was not adequate, did the researchers discuss this limitation?

Maximum response rate, defined as the total number of surveys returned divided by the total number of surveys sent,¹⁸ may be difficult to calculate with electronic or Web-based survey platforms. When the maximum response rate cannot be calculated, this issue needs to be addressed in the article’s limitations section.

The number of surveys has increased across fields over the past few decades, but survey response rates in general have decreased.^17,21,24,25 In fields outside of clinical medicine, response rates in the 40% range are common.¹⁷ In the 1990s, the mean response rate for surveys published in medical journals (mailed surveys) was approximately 60%.²⁶ A 2001 review of physician questionnaire studies found a similar average response rate (61%), with a 52% response rate for large-sample surveys.²⁷ In 2002, Field et al²⁸ examined the impact of incentives in physician survey studies and found response rates ranging from 8.5% to 80%.

Importantly, electronically delivered surveys (e-mail, Web-based) often have lower response rates than mailed surveys.^24,29 Nominal financial incentives have been associated with enhanced response rates.²⁸

A relatively low response rate does not necessarily mean you cannot trust the data. Survey scientists note that the representativeness of the sample may be more critical than response rate alone.¹⁷ Studies with small sample sizes may be more representative—and findings more valid—than those with large samples, if large samples are nonrepresentative when considering the target population.¹⁷

Do the conclusions go beyond the data?

Are the inferences overreaching, in view of the survey design? In studies with low response rates and nonrepresentative samples, researchers must be careful in interpreting the results. If the results cannot be generalized beyond the research sample, is this clear from the limitations, discussion, and conclusion sections?

In this review, we have summarized the findings of three published surveys^1,2,30 and commented on how they appear to meet—or don’t quite meet—recommendations for survey development, validation, and use. The papers chosen were deemed strong examples in particular categories, such as description of survey authorship,¹ instrument validation,³⁰ sampling methodology,² and response rate.¹

It should be noted that even when surveys are conducted with the utmost rigor, survey reporting may leave out critical details. Survey methodology may not be adequately described for a variety of reasons, including researchers’ training in survey design and methodology; a lack of universally accepted journal-reporting guidelines³; and even journals’ space limitations. At times, journals may excise descriptions of survey development and validation, deeming these sections superfluous. Limitations sections can be critical to interpreting the results of survey research and evaluating the scope of conclusions.

Surveys are common in medical research. Although survey research may be subject to inherent self-report bias, surveys have a great impact on policies and practices in medicine, often forming the basis for recommendations or new guidelines.^1,2 To interpret and use survey research results, clinicians should be familiar with key elements involved in the creation and validation of surveys.

The purpose of this article is to provide readers with a basic framework for evaluating surveys to allow them to be more informed as consumers of survey research.

IMPORTANT TOOLS IN MEDICAL RESEARCH

Surveys are important tools for answering questions on topics that are difficult to assess using other methods.³ They allow us to gather data systematically from subjects by asking questions, in order to make inferences about a larger population.^3,4 Clinicians use surveys to explore the opinions, beliefs, and perceptions of a group, or to investigate physician practice patterns and adherence to clinical guidelines. They may also use surveys to better understand why patients are not engaging in recommended behavioral or lifestyle changes.

Survey methods include interviews (in person, by phone) and questionnaires (paper-and-pencil, e-mailed, online).⁴

A well-constructed, validated survey can provide powerful data that may influence clinical practice, guide future research development, or drive the development and provision of needed programs and services. Surveys have the potential to transform the ways in which we think about and practice medicine.

READER BEWARE

While survey research in health care appears to have grown exponentially, the quality of reported survey research has not necessarily increased over time.

For consumers of survey research, the adage “reader beware” is apt. Although a considerable number of studies have examined the effects of survey methodology on the validity, reliability, and generalizability of the results,⁴ medical journals differ in their requirements for reporting survey methods.

In an analysis of 117 articles, Bennett et al³ found that more than 80% did not fully describe the survey development process or pretesting methods. They also found limited guidance and lack of consensus about the best way to report survey research. Of 95 surveys requiring scoring, 66% did not report scoring practices.

Duffett et al⁵ noted that of 127 critical care medicine surveys, only 36% had been pretested or pilot-tested, and half of all surveys reviewed did not include participant demographics or included only minimal information.

Because journal reporting practices differ, physicians may be unaware of the steps involved in survey construction and validation. Knowledge of these steps is helpful not only in constructing surveys but also in assessing published articles that used survey research.

LIMITATIONS OF SURVEY RESEARCH

Indirect measures of attitudes and behaviors

Surveys that rely on participants’ self-reports of behaviors, attitudes, beliefs, or actions are indirect measures and are susceptible to self-report and social-desirability biases. Participants may overestimate their own expertise or knowledge in self-report surveys. They may wish to reduce embarrassment⁶ or answer in ways that would make them “look better,”⁷ resulting in social-desirability bias. These issues need to be mentioned in the limitations section in papers reporting survey research.

Questions and response choices

The data derived from surveys are only as good as the questions that are asked.⁸ Stone⁹ noted that questions should be intelligible, unambiguous, and unbiased. If respondents do not comprehend questions as researchers intended, if questionnaire response choices are inadequate, or if questions trigger unintended emotional responses,^10–14 researchers may unwittingly introduce error, which will affect the validity of results. Even seemingly objective questions, such as those related to clinical algorithm use, practice patterns, or equipment available to hospital staff, may be interpreted differently by different respondents.

In their eagerness to launch a survey, clinician researchers may not realize that it must be carefully constructed. A focus on question development and validation is critical, as the questions determine the quality of the data derived from the survey.⁸ Even the position of the question or answer in the survey can affect how participants respond,¹⁵ as they may be guided to a response choice by preceding questions.¹⁶

WHAT DO YOU NEED TO KNOW ABOUT ASSESSING SURVEY RESEARCH?

What follows are questions and a basic framework that can be used to evaluate published survey research. Recommendations are based on the work of survey scientists,^{4,7,10,14,15,17,18} survey researchers in medicine and the social sciences, and national standards for test and questionnaire construction and validation (Table 1).^4,19,20

Who created the survey? How did they do it?

How the survey was created should be sufficiently described to allow readers to judge the adequacy of instrument development.^3–5 It is generally recommended that feedback from multiple sources be solicited during survey creation. Both questionnaire-design experts and subject-matter experts are considered critical in the process.⁴

What question was the survey designed to answer?

Is the objective of the study articulated in the paper? ^3,20 To judge survey research, readers need to know if the survey appears to adequately address the research question or questions and the objectives of the study in terms of methods used.⁴

Instrument pretesting and field testing are considered best practices by the American Association for Public Opinion Research, a professional organization for US survey scientists.⁴

Pretesting can include cognitive interviewing, the use of questionnaire appraisal tools, and hybrid methods, all of which are aimed at addressing validity issues.²¹ Pretesting with a group of participants similar to the target population allows for assessment of item ambiguity, instrument ease of use, adequacy of response categories (response choices), and time to completion.^4,12

Cognitive interviewing is designed to explore respondents’ comprehension of questions, response processes, and decision processes governing how they answer questions.^4,7,10,11 In cognitive interviewing, respondents are generally interviewed one on one. Techniques vary, but typically include “think alouds” (in which a respondent is asked to verbalize thoughts while responding to questions) and “verbal probing” (in which the respondent answers a question, then is asked follow-up questions as the interviewer probes for information related to the response choice or question itself).⁷ These techniques can provide evidence that researchers are actually measuring what they set out to measure and not an unrelated construct.^4,19

Field testing of a survey under realistic conditions can help to uncover problems in administration, such as issues in standardization of key procedures, and to ensure that the survey was administered as the researchers intended.^21,22 Field testing is vital before phone or in-person interviews to ensure standardization of any critical procedures. Pilot testing in a sample similar to the intended population allows for further refinement, with deletion of problem items, before the survey is launched.¹⁵

Because even “objective” questions can be somewhat subjective, all research surveys should go through some type of pretesting.^4,21 Based on the results of pretesting and field testing, surveys should then be revised before launch.^4,21 If an article on a self-report survey makes no mention of survey validation steps, readers may well question the validity of the results.

Are the survey questions and response choices understandable?

Is the meaning of each question unambiguous? Is the reading level appropriate for the sample population (a critical consideration in patient surveys)? Do any of the items actually ask two different questions?¹³ An example would be: “Was the representative courteous and prompt?” as it is possible to be courteous, but not prompt, and vice versa. If so, respondents may be confused or frustrated in attempting to answer it. If a rating scale is used throughout the questionnaire, are the anchors appropriate? For example, a question may be written in such a way that respondents want to answer “yes/no” or “agree/disagree,” but the scale used may include response options such as “poor,” “marginal,” “good,” and “excellent.” Items with Likert-response formats are commonly used in self-report surveys and allow participants to respond to a statement by choosing from a range of responses (eg, strongly disagree to strongly agree), often spaced horizontally under a line.

It is recommended that surveys also include options for answers beyond the response choices provided,²⁰ such as comment boxes or fill-in-the-blank items. Surveys with a closed-response format may constrain the quality of data collected because investigators may not foresee all possible answers. Surveys need to be available for review either within the article itself, in an appendix, or as supplementary material that is available elsewhere.

Does the sample appear to be appropriate?

Articles that report the results of surveys should describe the target population, the sample design, and, in a demographic table, respondents and nonrespondents. To judge appropriateness, several questions can be asked regarding sampling:

Target population. Is the population of interest (ie, the target population) described, including regional demographics, if applicable? The relationship between the sample and the target population is important, as a nonrepresentative sample may result in misleading conclusions about the population of interest.

Sampling frame. Who had an opportunity to participate in the survey? At its simplest, the sampling frame establishes who (or what, in the case of institutions) should be included within the sample. This is typically a list of elements (Groves et al⁴) that acts to “frame” or define the sample to be selected. Where the target population may be all academic internal medicine physicians in the United States, the sampling frame may be all male and female US physicians who are members of particular internal medicine professional organizations, identified by their directory email addresses.

Sample design. How was the sample actually selected?⁴ For example, did investigators use a convenience sample of colleagues at other institutions or use a stratified random sample, ensuring adequate representation of respondents with certain characteristics?

Description of respondents. How is the sample of respondents described? Are demographic features reported, including statistics on regional or national representativeness?⁵ Does the sample of survey respondents appear to be representative of the researcher’s population of interest (ie, the target population)?^3,23 If not, is this adequately described in the limitations section? Although outcomes will not be available on nonrespondents, demographic and baseline data often are available and should be reported. Are there systematic differences between respondents and nonrespondents?

Was the response rate adequate?

Was the response rate adequate, given the number of participants initially recruited? If the response rate was not adequate, did the researchers discuss this limitation?

Maximum response rate, defined as the total number of surveys returned divided by the total number of surveys sent,¹⁸ may be difficult to calculate with electronic or Web-based survey platforms. When the maximum response rate cannot be calculated, this issue needs to be addressed in the article’s limitations section.

The number of surveys has increased across fields over the past few decades, but survey response rates in general have decreased.^17,21,24,25 In fields outside of clinical medicine, response rates in the 40% range are common.¹⁷ In the 1990s, the mean response rate for surveys published in medical journals (mailed surveys) was approximately 60%.²⁶ A 2001 review of physician questionnaire studies found a similar average response rate (61%), with a 52% response rate for large-sample surveys.²⁷ In 2002, Field et al²⁸ examined the impact of incentives in physician survey studies and found response rates ranging from 8.5% to 80%.

Importantly, electronically delivered surveys (e-mail, Web-based) often have lower response rates than mailed surveys.^24,29 Nominal financial incentives have been associated with enhanced response rates.²⁸

A relatively low response rate does not necessarily mean you cannot trust the data. Survey scientists note that the representativeness of the sample may be more critical than response rate alone.¹⁷ Studies with small sample sizes may be more representative—and findings more valid—than those with large samples, if large samples are nonrepresentative when considering the target population.¹⁷

Do the conclusions go beyond the data?

Are the inferences overreaching, in view of the survey design? In studies with low response rates and nonrepresentative samples, researchers must be careful in interpreting the results. If the results cannot be generalized beyond the research sample, is this clear from the limitations, discussion, and conclusion sections?

In this review, we have summarized the findings of three published surveys^1,2,30 and commented on how they appear to meet—or don’t quite meet—recommendations for survey development, validation, and use. The papers chosen were deemed strong examples in particular categories, such as description of survey authorship,¹ instrument validation,³⁰ sampling methodology,² and response rate.¹

It should be noted that even when surveys are conducted with the utmost rigor, survey reporting may leave out critical details. Survey methodology may not be adequately described for a variety of reasons, including researchers’ training in survey design and methodology; a lack of universally accepted journal-reporting guidelines³; and even journals’ space limitations. At times, journals may excise descriptions of survey development and validation, deeming these sections superfluous. Limitations sections can be critical to interpreting the results of survey research and evaluating the scope of conclusions.

Surveys are common in medical research. Although survey research may be subject to inherent self-report bias, surveys have a great impact on policies and practices in medicine, often forming the basis for recommendations or new guidelines.^1,2 To interpret and use survey research results, clinicians should be familiar with key elements involved in the creation and validation of surveys.

The purpose of this article is to provide readers with a basic framework for evaluating surveys to allow them to be more informed as consumers of survey research.

IMPORTANT TOOLS IN MEDICAL RESEARCH

Surveys are important tools for answering questions on topics that are difficult to assess using other methods.³ They allow us to gather data systematically from subjects by asking questions, in order to make inferences about a larger population.^3,4 Clinicians use surveys to explore the opinions, beliefs, and perceptions of a group, or to investigate physician practice patterns and adherence to clinical guidelines. They may also use surveys to better understand why patients are not engaging in recommended behavioral or lifestyle changes.

Survey methods include interviews (in person, by phone) and questionnaires (paper-and-pencil, e-mailed, online).⁴

A well-constructed, validated survey can provide powerful data that may influence clinical practice, guide future research development, or drive the development and provision of needed programs and services. Surveys have the potential to transform the ways in which we think about and practice medicine.

READER BEWARE

While survey research in health care appears to have grown exponentially, the quality of reported survey research has not necessarily increased over time.

For consumers of survey research, the adage “reader beware” is apt. Although a considerable number of studies have examined the effects of survey methodology on the validity, reliability, and generalizability of the results,⁴ medical journals differ in their requirements for reporting survey methods.

In an analysis of 117 articles, Bennett et al³ found that more than 80% did not fully describe the survey development process or pretesting methods. They also found limited guidance and lack of consensus about the best way to report survey research. Of 95 surveys requiring scoring, 66% did not report scoring practices.

Duffett et al⁵ noted that of 127 critical care medicine surveys, only 36% had been pretested or pilot-tested, and half of all surveys reviewed did not include participant demographics or included only minimal information.

Because journal reporting practices differ, physicians may be unaware of the steps involved in survey construction and validation. Knowledge of these steps is helpful not only in constructing surveys but also in assessing published articles that used survey research.

LIMITATIONS OF SURVEY RESEARCH

Indirect measures of attitudes and behaviors

Surveys that rely on participants’ self-reports of behaviors, attitudes, beliefs, or actions are indirect measures and are susceptible to self-report and social-desirability biases. Participants may overestimate their own expertise or knowledge in self-report surveys. They may wish to reduce embarrassment⁶ or answer in ways that would make them “look better,”⁷ resulting in social-desirability bias. These issues need to be mentioned in the limitations section in papers reporting survey research.

Questions and response choices

The data derived from surveys are only as good as the questions that are asked.⁸ Stone⁹ noted that questions should be intelligible, unambiguous, and unbiased. If respondents do not comprehend questions as researchers intended, if questionnaire response choices are inadequate, or if questions trigger unintended emotional responses,^10–14 researchers may unwittingly introduce error, which will affect the validity of results. Even seemingly objective questions, such as those related to clinical algorithm use, practice patterns, or equipment available to hospital staff, may be interpreted differently by different respondents.

In their eagerness to launch a survey, clinician researchers may not realize that it must be carefully constructed. A focus on question development and validation is critical, as the questions determine the quality of the data derived from the survey.⁸ Even the position of the question or answer in the survey can affect how participants respond,¹⁵ as they may be guided to a response choice by preceding questions.¹⁶

WHAT DO YOU NEED TO KNOW ABOUT ASSESSING SURVEY RESEARCH?

What follows are questions and a basic framework that can be used to evaluate published survey research. Recommendations are based on the work of survey scientists,^{4,7,10,14,15,17,18} survey researchers in medicine and the social sciences, and national standards for test and questionnaire construction and validation (Table 1).^4,19,20

Who created the survey? How did they do it?

How the survey was created should be sufficiently described to allow readers to judge the adequacy of instrument development.^3–5 It is generally recommended that feedback from multiple sources be solicited during survey creation. Both questionnaire-design experts and subject-matter experts are considered critical in the process.⁴

What question was the survey designed to answer?

Is the objective of the study articulated in the paper? ^3,20 To judge survey research, readers need to know if the survey appears to adequately address the research question or questions and the objectives of the study in terms of methods used.⁴

Instrument pretesting and field testing are considered best practices by the American Association for Public Opinion Research, a professional organization for US survey scientists.⁴

Pretesting can include cognitive interviewing, the use of questionnaire appraisal tools, and hybrid methods, all of which are aimed at addressing validity issues.²¹ Pretesting with a group of participants similar to the target population allows for assessment of item ambiguity, instrument ease of use, adequacy of response categories (response choices), and time to completion.^4,12

Cognitive interviewing is designed to explore respondents’ comprehension of questions, response processes, and decision processes governing how they answer questions.^4,7,10,11 In cognitive interviewing, respondents are generally interviewed one on one. Techniques vary, but typically include “think alouds” (in which a respondent is asked to verbalize thoughts while responding to questions) and “verbal probing” (in which the respondent answers a question, then is asked follow-up questions as the interviewer probes for information related to the response choice or question itself).⁷ These techniques can provide evidence that researchers are actually measuring what they set out to measure and not an unrelated construct.^4,19

Field testing of a survey under realistic conditions can help to uncover problems in administration, such as issues in standardization of key procedures, and to ensure that the survey was administered as the researchers intended.^21,22 Field testing is vital before phone or in-person interviews to ensure standardization of any critical procedures. Pilot testing in a sample similar to the intended population allows for further refinement, with deletion of problem items, before the survey is launched.¹⁵

Because even “objective” questions can be somewhat subjective, all research surveys should go through some type of pretesting.^4,21 Based on the results of pretesting and field testing, surveys should then be revised before launch.^4,21 If an article on a self-report survey makes no mention of survey validation steps, readers may well question the validity of the results.

Are the survey questions and response choices understandable?

Is the meaning of each question unambiguous? Is the reading level appropriate for the sample population (a critical consideration in patient surveys)? Do any of the items actually ask two different questions?¹³ An example would be: “Was the representative courteous and prompt?” as it is possible to be courteous, but not prompt, and vice versa. If so, respondents may be confused or frustrated in attempting to answer it. If a rating scale is used throughout the questionnaire, are the anchors appropriate? For example, a question may be written in such a way that respondents want to answer “yes/no” or “agree/disagree,” but the scale used may include response options such as “poor,” “marginal,” “good,” and “excellent.” Items with Likert-response formats are commonly used in self-report surveys and allow participants to respond to a statement by choosing from a range of responses (eg, strongly disagree to strongly agree), often spaced horizontally under a line.

It is recommended that surveys also include options for answers beyond the response choices provided,²⁰ such as comment boxes or fill-in-the-blank items. Surveys with a closed-response format may constrain the quality of data collected because investigators may not foresee all possible answers. Surveys need to be available for review either within the article itself, in an appendix, or as supplementary material that is available elsewhere.

Does the sample appear to be appropriate?

Articles that report the results of surveys should describe the target population, the sample design, and, in a demographic table, respondents and nonrespondents. To judge appropriateness, several questions can be asked regarding sampling:

Target population. Is the population of interest (ie, the target population) described, including regional demographics, if applicable? The relationship between the sample and the target population is important, as a nonrepresentative sample may result in misleading conclusions about the population of interest.

Sampling frame. Who had an opportunity to participate in the survey? At its simplest, the sampling frame establishes who (or what, in the case of institutions) should be included within the sample. This is typically a list of elements (Groves et al⁴) that acts to “frame” or define the sample to be selected. Where the target population may be all academic internal medicine physicians in the United States, the sampling frame may be all male and female US physicians who are members of particular internal medicine professional organizations, identified by their directory email addresses.

Sample design. How was the sample actually selected?⁴ For example, did investigators use a convenience sample of colleagues at other institutions or use a stratified random sample, ensuring adequate representation of respondents with certain characteristics?

Description of respondents. How is the sample of respondents described? Are demographic features reported, including statistics on regional or national representativeness?⁵ Does the sample of survey respondents appear to be representative of the researcher’s population of interest (ie, the target population)?^3,23 If not, is this adequately described in the limitations section? Although outcomes will not be available on nonrespondents, demographic and baseline data often are available and should be reported. Are there systematic differences between respondents and nonrespondents?

Was the response rate adequate?

Was the response rate adequate, given the number of participants initially recruited? If the response rate was not adequate, did the researchers discuss this limitation?

Maximum response rate, defined as the total number of surveys returned divided by the total number of surveys sent,¹⁸ may be difficult to calculate with electronic or Web-based survey platforms. When the maximum response rate cannot be calculated, this issue needs to be addressed in the article’s limitations section.

The number of surveys has increased across fields over the past few decades, but survey response rates in general have decreased.^17,21,24,25 In fields outside of clinical medicine, response rates in the 40% range are common.¹⁷ In the 1990s, the mean response rate for surveys published in medical journals (mailed surveys) was approximately 60%.²⁶ A 2001 review of physician questionnaire studies found a similar average response rate (61%), with a 52% response rate for large-sample surveys.²⁷ In 2002, Field et al²⁸ examined the impact of incentives in physician survey studies and found response rates ranging from 8.5% to 80%.

Importantly, electronically delivered surveys (e-mail, Web-based) often have lower response rates than mailed surveys.^24,29 Nominal financial incentives have been associated with enhanced response rates.²⁸

A relatively low response rate does not necessarily mean you cannot trust the data. Survey scientists note that the representativeness of the sample may be more critical than response rate alone.¹⁷ Studies with small sample sizes may be more representative—and findings more valid—than those with large samples, if large samples are nonrepresentative when considering the target population.¹⁷

Do the conclusions go beyond the data?

Are the inferences overreaching, in view of the survey design? In studies with low response rates and nonrepresentative samples, researchers must be careful in interpreting the results. If the results cannot be generalized beyond the research sample, is this clear from the limitations, discussion, and conclusion sections?

In this review, we have summarized the findings of three published surveys^1,2,30 and commented on how they appear to meet—or don’t quite meet—recommendations for survey development, validation, and use. The papers chosen were deemed strong examples in particular categories, such as description of survey authorship,¹ instrument validation,³⁰ sampling methodology,² and response rate.¹

It should be noted that even when surveys are conducted with the utmost rigor, survey reporting may leave out critical details. Survey methodology may not be adequately described for a variety of reasons, including researchers’ training in survey design and methodology; a lack of universally accepted journal-reporting guidelines³; and even journals’ space limitations. At times, journals may excise descriptions of survey development and validation, deeming these sections superfluous. Limitations sections can be critical to interpreting the results of survey research and evaluating the scope of conclusions.

Paget disease of bone is a focal disorder of the aging skeleton that can be asymptomatic or can present with pain, bowing deformities, fractures, or nonspecific rheumatic complaints. Physicians often discover it in asymptomatic patients when serum alkaline phosphatase levels are elevated or as an incidental finding on radiography. Despite evidence of germline mutations and polymorphisms that predispose to Paget disease, the environmental determinants that permit disease expression in older people remain unknown.

A STRIKING GEOGRAPHIC DISTRIBUTION

Researchers have been studying the determinants and distribution of Paget disease ever since Sir James Paget first described it in 1877.¹

Paget disease has a predilection for the axial skeleton, particularly the lumbosacral spine and pelvis, as well as the skull, femur, and tibia.² Knowing this, investigators have used screening plain films of the abdomen (kidney-ureter-bladder views) to estimate its prevalence in different populations, as these images capture the lumbosacral spine, pelvis, and proximal femurs. Other means of assessing prevalence have included autopsy series, questionnaires, and screens for biochemical markers of bone turnover, such as elevated serum alkaline phosphatase from bone.^3–6

Using these methods, Paget disease has been estimated to occur in 1% to 3% of people over age 55, and in as many as 8% of people over age 80 in certain countries.⁷

This disease has a striking geographic distribution, being frequent in Europe, Canada, the United States, Australia, New Zealand, and cities of South America, but rare in Scandinavia and Japan. It seems to be equally rare in other countries of the Far East and in India, Russia, and Africa, although its prevalence in these areas has not been thoroughly investigated.⁸

That it is an ancient disease has been corroborated by excavations in churchyards in Great Britain.^9,10 It may be familial or sporadic, but its expression is delayed until late middle age in most persons, and it does not occur in children. For reasons unclear, the prevalence seems to be decreasing in many countries.^11–13

GENETICS IS NOT THE WHOLE STORY

The variable prevalence of Paget disease in different geographic regions and its sometimes-familial expression suggest a genetic predisposition, environmental factor, or both.

Mutations in SQSTM1

In 2002, scientists investigating a cohort of French Canadian families found a mutation in the SQSTM1 gene that was present in almost 50% of people with familial Paget disease and in 16% of those with sporadic Paget disease.¹⁴ Hocking and his colleagues in the United Kingdom subsequently found the same mutation in 19% of cases of familial Paget disease and in 9% of sporadic cases.¹⁵

Further, investigators noted that the mutation was often present on a conserved haplotype, consistent with a stable genetic change occurring in the affected population.¹⁶ This observation of a “founder effect” dovetailed with the epidemiology of Paget disease,¹⁷ but only with this SQSTM1 mutation.

Throughout Europe, Australia, and the United States, comparable rates of the SQSTM1 mutation were reported in or around the ubiquitin-associated domain. Several specific mutations exist, the most common one being P392L, ie, a prolineto-leucine substitution at amino acid 392. Scientists have tried to correlate severity of disease with genotype, but the findings have been inconsistent.^18–21

Investigations into the mechanism of disease have pointed to the role of p62, the product of SQSTM1, in signaling osteoclast activation via nuclear factor kappa B. Since this initial discovery, polymorphisms in the genes affecting osteoclast maturation, activation, and fusion pathways have been shown to predispose to Paget disease. Examples:

• TNFRSF11A, which codes for receptor activator of nuclear factor kappa B, or RANK
• TNFRSF11B, which codes for osteoprotegerin, or OPG
• CSF1, which codes for macrophage colony-stimulating factor 1, and
• OPTN, which codes for optineurin, a member of the nuclear factor kappa B-modulating protein family.

Clinicians interested in these details can read an excellent review of the pathogenesis of Paget disease.²²

Other possible factors

Although there is good evidence that measles and canine distemper virus can infect osteoclasts and modify their phenotype, there is no good evidence that these infections by themselves cause Paget disease.^23–25 It is, however, tempting to think of these RNA paramyxoviruses as precipitating factors; conceivably, an infectious agent might seed the ends of long bones, accounting for the fixed distribution of Paget disease and its late expression.

Epidemiologic studies from around the world have failed to identify conclusively any environmental exposure that predisposes to Paget disease, although a rural setting, trauma, infection, and milk ingestion have all been proposed.^26–28 It is also possible that as bone ages and the marrow becomes less cellular and more fatty, these changes may permit the disease to develop.

The greatest risk factor for Paget disease is perhaps aging, followed by ancestry and a known family history of it. That genetics is not the whole story is evident by reports of people with SQSTM1 mutations who show no clinical evidence of Paget disease in their old age, and patients with Paget disease who have no SQSTM1 mutation.^20,29

CLINICAL PRESENTATION

Most patients with Paget disease have no symptoms and come to medical attention because of an elevated serum alkaline phosphatase level or characteristic findings on radiographs ordered for other indications.¹¹ Paget disease is the second most common disorder of aging bone after osteoporosis. Yet unlike osteoporosis, which presents as a systemic fragility of bone, the clinical manifestations of Paget disease depend on which bones are affected and how enlarged or misshapen they have become.

Common complications

As a consequence of this abnormal bone remodeling and overgrowth, many patients present with bone pain. Bone deformity, headache, and hearing loss may also occur (Figure 1), as well as fractures and nerve compression syndromes (eg, spinal stenosis, sciatica, cauda equina syndrome).

It is important to remember that “pagetic” bone may not be the source of pain, and that functional impairment caused by degenerative changes at affected sites is common (Figure  2).^30,31

In a study from the New England Registry for Paget’s Disease,³² most patients knew fairly well which bones were affected and what complications resulted from this when deformity, fracture, or total joint replacement had occurred.³² Although Paget disease did affect their quality of life as measured by physical functioning on the Short Form-12 assessment, these impairments did not seem to affect their outlook, which was as good as or better than that in other people their age.

Metabolic complications

Metabolic complications of Paget disease are rare today but can occur in an elderly patient who has active, polyostotic (multibone) disease.³³ The accelerated rate of bone remodeling and the increased vascularity of pagetic bone have been reported to lead to high-output heart failure. In theory, treatment should ease this by diminishing blood flow to pagetic bone and restoring bone turnover to more normal levels.³⁴

Hypercalcemia can occur when patients with Paget disease are immobilized for any reason, and there is probably a higher incidence of renal stones in patients with Paget disease.^35,36

Malignant complications

Osteosarcoma rarely arises in pagetic bone. Yet Paget disease may account for a significant number of cases of this cancer in the elderly.³⁷ In these cases, osteosarcoma is presumed to be driven by a second genetic mutation, has a genetic signature distinct from that in osteosarcomas occurring in youth, and is quite resistant to treatment.³⁸ In Scandinavia and Japan, where Paget disease is rare, the second peak of osteosarcoma that occurs with aging seems muted as well.^39,40 These cancers present with pain, soft-tissue swelling, and variable elevations in serum alkaline phosphatase. Investigations to date suggest that pagetic lesions and osteosarcomas arising in pagetic bone are probably both driven to some extent by stromal cells overexpressing RANK ligand and may not represent defects intrinsic to the osteoclast.⁴¹

Giant-cell tumors of bone are also rare but can arise in pagetic bone. A cluster of cases was reported in Avellino and other towns of southern Italy.⁴² Again, the lesions occur in older individuals and in different sites than those seen in the benign giant-cell tumors recorded in patients without Paget disease.

Metastases from lymphomas, prostate cancer, and breast cancer certainly occur in bone, but rarely in pagetic sites.⁴³ A recent case study noted that patients with prostate cancer who also had Paget disease had a later onset of metastasis to bone than patients without coincident Paget disease.⁴⁴

A THOUGHTFUL ASSESSMENT

Evaluating a patient with Paget disease requires a thoughtful assessment of its musculoskeletal consequences in an aging skeleton. Pain in Paget disease is often multifactorial. In the elderly, end-stage degenerative disease of the spine, hip, and knees, mechanical instability, compression fractures of the spine, and neuropathies may compound the clinical picture. Therefore, a thorough evaluation is required to plan effective therapy.

Alkaline phosphatase and other markers

A screening serum alkaline phosphatase level is usually sufficient to measure bone turnover. Produced by osteoblasts, alkaline phosphatase is a marker of bone formation, but an imperfect one. Often it is elevated in active Paget disease—but not always.⁴⁵ Many patients have normal serum alkaline phosphatase levels, particularly if they have monostotic (single-bone) disease. It is unclear why, in a disorder marked by accelerated bone remodeling, the biochemical markers are inconsistent measures of bone turnover.

Research into biochemical markers of Paget disease has had two aims: to identify the single best marker for baseline assessment of pagetic bone activity and to find out whether this measurement responds to therapy.^46,47 Measures of bone formation such as bone-specific alkaline phosphatase, osteocalcin, and the procollagen type I peptides, and measures of bone resorption including the pyridinolines, hydroxyproline, and cross-linked collagens, have been analyzed as markers of bone remodeling and show no real advantage over the serum alkaline phosphatase level as reflections of bone turnover. As alkaline phosphatase measurement is inexpensive, available, and reliable, it should be used preferentially, with gamma-glutamyl transpeptidate or 5′ nucleotidase confirming the source as either liver or bone. Readers are directed to a recent review in which the utility of these markers is explored in more detail.⁴⁸

Imaging studies

Bone scans can give us an idea of the extent, location, and general activity of the disease (Figure 3). Uptake is avid in affected bones, beginning in the subchondral region and spreading throughout the bone. Bone scans can be particularly useful in defining sites of active disease when the serum alkaline phosphatase level is normal.

Plain radiography of the affected bones outlines the anatomy of the problem and gives some insight into the cause of pain (Figure 3).

Computed tomography or magnetic resonance imaging may prove useful in cases of spinal stenosis, cauda equina syndrome, compression fractures, or suspected malignancy (Figure  4), but these studies are expensive and generally are not needed.

Radiographic features. Paget disease is presumed to be a disease of the osteoclast, and the earliest lesion is described as lytic. In my own experience, it is unusual to see a purely lytic lesion, although sometimes the disease presents in the skull in this way—osteoporosis circumscripta—or in the femur or tibia with an advancing edge of pure osteolysis.

More often, one sees evidence of both resorption by osteoclasts and formation by osteoblasts, reflecting the coupling of these two processes in this disease. Radiographic findings on plain films are usually definitive, showing enlargement of the affected bone, deformity, coarsened trabeculae, and thickened cortices with tunneling (Figure 5).⁴⁹ In weightbearing bones, pseudofractures may stud the convex surface. These incongruities of bone may persist for years, heralding fracture only when there is focal pain (Figure 6).⁵⁰

Biopsy is infrequently needed

If these diagnostic findings are not present, then biopsy is indicated. In the United Sates and Canada, where Paget disease is fairly common, biopsy is infrequently needed and is usually reserved for situations in which the differential diagnosis includes cancer, as when the cortex cannot be clearly visualized, the lesions are atypical in pattern or location, or there is a single sclerotic vertebral body on imaging.⁵¹

The other indication for biopsy is a “new” pagetic lesion. For reasons unknown, the pattern of skeletal involvement in Paget disease tends to be stable throughout the patient’s lifetime. This is another reason why a baseline bone scan is useful.

TREATMENT WITH BISPHOSPHONATES

Treatment of Paget disease today relies for the most part on the new generation of nitrogen-containing bisphosphonates. As a class, these are antiresorptive agents that inhibit osteoclasts; in this way they slow bone remodeling and enhance the deposition of normal lamellar bone. Their clinical efficacy in Paget disease, coupled with the observation that the earliest lesion in Paget disease is lytic, underscores the principle that Paget disease is a disorder of the osteoclast.

Oral bisphosphonates

Etidronate, approved in 1977, was the first bisphosphonate licensed to treat Paget disease, and it remains available for this indication in the United States. Used in 6-month regimens, it lowers the serum alkaline phosphatase level in some patients, but it has a narrow therapeutic margin. Drug-induced osteomalacia and worsening lytic lesions and fractures in weight-bearing bones are some of the complications.⁵² When the nitrogen-containing bisphosphonates were developed, they proved to be more potent antiresorptive agents that pose less risk of mineralization defects at prescribed doses.

Alendronate, approved in 1995, is an oral nitrogen-containing bisphosphonate that is effective in treating Paget disease.⁵³ Alendronate is now available in the United States only through special programs (eg, the CVS ProCare Program); the paperwork required to secure this drug is onerous, so the drug is used infrequently. Studies in Paget disease showed that it normalizes the serum alkaline phosphatase level, improves the radiographic appearance, and eases pain in many patients.⁵⁴ The dosage is 40 mg daily for 6 months.

Risedronate, approved in 1998, is another oral nitrogen-containing bisphosphonate and is comparable to alendronate in efficacy.⁵⁵ The dosage is 30 mg daily for 2 months.

Tiludronate is another oral bisphosphonate with a different mechanism of action from the nitrogen-containing bisphosphonates.⁵⁶ It is safe, often effective, but less potent than the newer agents.

The oral bisphosphonates are well tolerated, with few side effects other than gastrointestinal distress. As a class, they are poorly absorbed and so must be taken fasting with a full glass of water on rising, after which the patient should remain upright without food or drink for 30 to 60 minutes. This is a nuisance for elderly patients already on multiple medications and thus makes intravenous agents appealing.

Intravenous bisphosphonates

Pamidronate was approved in 1994. It is quite effective in many patients with Paget disease. There is no consensus around the world on dosing, with regimens ranging from 30 mg to 90 mg or more intravenously in divided doses given over 2 to 4 hours from once a day to once a week. In the United States, 30 mg is given over 4 hours on 3 consecutive days. Resistance to pamidronate has been described; the mechanism is unknown.

Zoledronic acid is a nitrogen-containing bisphosphonate. It is given as a single infusion over 15 minutes, and re-treatment may not be necessary for years. A randomized clinical trial in 2005 demonstrated the efficacy of zoledronic acid 5 mg by infusion compared with oral risedronate in the treatment of Paget disease.⁵⁷ In observational extension studies lasting as long as 6.5 years, zoledronic acid has been shown to be superior to risedronate in terms of the proportion of patients experiencing a sustained clinical remission.⁵⁸

While there are many bisphosphonates on the market, an infusion of 5 mg of zoledronic acid seems optimal in most patients who do not have a contraindication or an aversion to intravenous therapy. It tends to normalize the serum alkaline phosphatase level quickly and to leave more patients in sustained biochemical remission than do older bisphosphonates, as noted above. It also tends to be more effective in normalizing the serum alkaline phosphatase level when a patient has used other bisphosphonates in the past or has become resistant to them.

Bisphosphonates reduce bone turnover but do not correct deformities

In randomized clinical trials, bisphosphonates have been shown to restore bone remodeling to more normal levels, to ease pain from pagetic bone, to lower the serum alkaline phosphatase level, and to heal radiographic lesions, but these drugs have not been proven to prevent progression of deformity or to restore the structural integrity of bone (Figure 6).

The Paget’s Disease: Randomized Trial of Intensive Versus Symptomatic Management (PRISM), in 1,324 people with Paget disease in the United Kingdom, showed no difference in the incidence of fracture, orthopedic surgery, quality of life, or hearing thresholds over 2 to 5 years in patients treated with bisphosphonates vs those treated symptomatically, despite a significant difference in serum alkaline phosphatase in the two groups (P < .001).⁵⁹

In the observational extension study of zoledronic acid described above,⁵⁸ three of four fractures occurred in the group treated with zoledronic acid, echoing the findings of the PRISM study.

Adverse effects of bisphosphonates

The more potent the bisphosphonate is as an antiresorptive agent, the more it suppresses normal bone remodeling, which can lead to osteonecrosis of the jaw and to atypical femoral fractures.^60,61 These complications are unusual in patients with Paget disease because the treatment is intermittent. Sometimes a single dose of zoledronic acid or one course of risedronate or alendronate will last for years.

All the nitrogen-containing bisphosphonates, particularly zoledronic acid, may provoke flulike symptoms of fever, arthralgias, and bone pain. This effect is self-limited, resolves in days, and does not tend to recur. Bone pain may be more sustained, but this also passes, and within weeks the antiresorptive process has abated and pagetic bone pain will ease. Atrial fibrillation is not an anticipated complication of treatment with a bisphosphonate.⁶² The risk of esophageal cancer is not confirmed at this time.⁶³ Other rare complications of the bisphosphonates include iritis, acute renal failure, and allergy.

Bisphosphonates are not approved for use in patients with creatinine clearance less than 30 mL/min, or in pregnancy.

Other treatments

Calcitonin, an older agent, can still be useful in easing the pain of Paget disease, healing bone lesions, and reducing the metabolic activity of pagetic bone in patients who cannot receive bisphosphonates. It is given by injection in doses of 50 to 100 IU daily or every other day. Although unlikely to effect a sustained clinical remission, calcitonin remains a safe, well-tolerated, and well-studied medication in Paget disease and is approved for this indication.^64,65

Denosumab has not been formally studied in Paget disease, but a recent case report indicated it was effective.⁶⁶

A conservative strategy

Guidelines for treating Paget disease have been written at various times in many countries, including Italy (2007),⁶⁷ the United Kingdom (2004),⁶⁸ Japan (2006),⁶⁹ and Canada (2007).⁷⁰ Recommendations differ, in part because it is hard to ascertain whether long-term outcomes are improved by treatment, and in part because the prevalence of Paget disease is decreasing and its severity is lessening.^11,12 Some guidelines are outdated, since they do not include the newer bisphosphonates.

If the natural history of untreated Paget disease involves the gradual evolution over more than 20 years of bowing deformities in the lower limbs, rigidity and overgrowth of the spine, and softening and enlargement of the skull, as described by Sir James Paget, then treatment should be initiated in hopes that it will modify the outcome. We have no lens to better focus this question on the effect of treatment on the natural history of the disease. We have the PRISM study, designed before zoledronic acid was approved and only 2 to 5 years in duration. And we have the epidemiologic data demonstrating that most patients have no symptoms during their lifetime.

We see the crippling bone disease described by Sir James Paget so infrequently today in the United States that we forget the profound morbidity that may attend the skeletal changes of Paget disease that were common in the early 20th century. Once the bones of the skull are overgrown, the limbs are bowed, and the degenerative joint disease is present, no medication can reverse these changes. Then, the integrity of the bone is lost, and the vulnerability to fracture, early osteoarthritis, nerve compression syndromes, and hearing loss persist. Understanding these consequences prompts the recommendation of early treatment in patients with Paget disease, in hopes of mitigating disease progression.

Patients with active Paget disease, documented either by an elevated serum alkaline phosphatase or by a bone scan, should be treated with a bisphosphonate if the disease is found in sites where remodeling of bone may lead to complications. Such sites include the skull, spine, and long bones of the lower extremity. Paget disease of bone in the pelvis tends to give little trouble (Figure 2) unless it is proximal to a joint, when pain and early arthritis may result. Treatment is safe and, I think, prudent to undertake in any person over age 55 with active disease. To prevent hypocalcemia during treatment, all patients should be repleted with vitamin D and maintained on calcium 1,200 mg daily through diet or supplements with meals.

Throughout the evaluation and treatment, it is important to remember that pain may not emanate from pagetic bone. If medication for Paget disease proves ineffective in the first few months, analgesics, bracing, walking aids, and operative management⁷¹ are adjunctive therapies to improve the functional status of these patients.

It is a remarkable clinical observation that treatment of Paget disease may rapidly reverse neurologic syndromes, resolve the erythema or warmth overlying active pagetic bone, and diminish the risk of bleeding with surgery. This response to therapy suggests that there is prompt inhibition and apoptosis of the osteoclasts, accompanied by diminished vascularity of bone. Whatever the mechanism, it is worth treating patients who have spinal stenosis, arthritis, and nerve compression syndromes with calcitonin or bisphosphonates before surgical intervention, whenever possible.^34,72

Paget disease of bone is a focal disorder of the aging skeleton that can be asymptomatic or can present with pain, bowing deformities, fractures, or nonspecific rheumatic complaints. Physicians often discover it in asymptomatic patients when serum alkaline phosphatase levels are elevated or as an incidental finding on radiography. Despite evidence of germline mutations and polymorphisms that predispose to Paget disease, the environmental determinants that permit disease expression in older people remain unknown.

A STRIKING GEOGRAPHIC DISTRIBUTION

Researchers have been studying the determinants and distribution of Paget disease ever since Sir James Paget first described it in 1877.¹

Paget disease has a predilection for the axial skeleton, particularly the lumbosacral spine and pelvis, as well as the skull, femur, and tibia.² Knowing this, investigators have used screening plain films of the abdomen (kidney-ureter-bladder views) to estimate its prevalence in different populations, as these images capture the lumbosacral spine, pelvis, and proximal femurs. Other means of assessing prevalence have included autopsy series, questionnaires, and screens for biochemical markers of bone turnover, such as elevated serum alkaline phosphatase from bone.^3–6

Using these methods, Paget disease has been estimated to occur in 1% to 3% of people over age 55, and in as many as 8% of people over age 80 in certain countries.⁷

This disease has a striking geographic distribution, being frequent in Europe, Canada, the United States, Australia, New Zealand, and cities of South America, but rare in Scandinavia and Japan. It seems to be equally rare in other countries of the Far East and in India, Russia, and Africa, although its prevalence in these areas has not been thoroughly investigated.⁸

That it is an ancient disease has been corroborated by excavations in churchyards in Great Britain.^9,10 It may be familial or sporadic, but its expression is delayed until late middle age in most persons, and it does not occur in children. For reasons unclear, the prevalence seems to be decreasing in many countries.^11–13

GENETICS IS NOT THE WHOLE STORY

The variable prevalence of Paget disease in different geographic regions and its sometimes-familial expression suggest a genetic predisposition, environmental factor, or both.

Mutations in SQSTM1

In 2002, scientists investigating a cohort of French Canadian families found a mutation in the SQSTM1 gene that was present in almost 50% of people with familial Paget disease and in 16% of those with sporadic Paget disease.¹⁴ Hocking and his colleagues in the United Kingdom subsequently found the same mutation in 19% of cases of familial Paget disease and in 9% of sporadic cases.¹⁵

Further, investigators noted that the mutation was often present on a conserved haplotype, consistent with a stable genetic change occurring in the affected population.¹⁶ This observation of a “founder effect” dovetailed with the epidemiology of Paget disease,¹⁷ but only with this SQSTM1 mutation.

Throughout Europe, Australia, and the United States, comparable rates of the SQSTM1 mutation were reported in or around the ubiquitin-associated domain. Several specific mutations exist, the most common one being P392L, ie, a prolineto-leucine substitution at amino acid 392. Scientists have tried to correlate severity of disease with genotype, but the findings have been inconsistent.^18–21

Investigations into the mechanism of disease have pointed to the role of p62, the product of SQSTM1, in signaling osteoclast activation via nuclear factor kappa B. Since this initial discovery, polymorphisms in the genes affecting osteoclast maturation, activation, and fusion pathways have been shown to predispose to Paget disease. Examples:

• TNFRSF11A, which codes for receptor activator of nuclear factor kappa B, or RANK
• TNFRSF11B, which codes for osteoprotegerin, or OPG
• CSF1, which codes for macrophage colony-stimulating factor 1, and
• OPTN, which codes for optineurin, a member of the nuclear factor kappa B-modulating protein family.

Clinicians interested in these details can read an excellent review of the pathogenesis of Paget disease.²²

Other possible factors

Although there is good evidence that measles and canine distemper virus can infect osteoclasts and modify their phenotype, there is no good evidence that these infections by themselves cause Paget disease.^23–25 It is, however, tempting to think of these RNA paramyxoviruses as precipitating factors; conceivably, an infectious agent might seed the ends of long bones, accounting for the fixed distribution of Paget disease and its late expression.

Epidemiologic studies from around the world have failed to identify conclusively any environmental exposure that predisposes to Paget disease, although a rural setting, trauma, infection, and milk ingestion have all been proposed.^26–28 It is also possible that as bone ages and the marrow becomes less cellular and more fatty, these changes may permit the disease to develop.

The greatest risk factor for Paget disease is perhaps aging, followed by ancestry and a known family history of it. That genetics is not the whole story is evident by reports of people with SQSTM1 mutations who show no clinical evidence of Paget disease in their old age, and patients with Paget disease who have no SQSTM1 mutation.^20,29

CLINICAL PRESENTATION

Most patients with Paget disease have no symptoms and come to medical attention because of an elevated serum alkaline phosphatase level or characteristic findings on radiographs ordered for other indications.¹¹ Paget disease is the second most common disorder of aging bone after osteoporosis. Yet unlike osteoporosis, which presents as a systemic fragility of bone, the clinical manifestations of Paget disease depend on which bones are affected and how enlarged or misshapen they have become.

Common complications

As a consequence of this abnormal bone remodeling and overgrowth, many patients present with bone pain. Bone deformity, headache, and hearing loss may also occur (Figure 1), as well as fractures and nerve compression syndromes (eg, spinal stenosis, sciatica, cauda equina syndrome).

It is important to remember that “pagetic” bone may not be the source of pain, and that functional impairment caused by degenerative changes at affected sites is common (Figure  2).^30,31

In a study from the New England Registry for Paget’s Disease,³² most patients knew fairly well which bones were affected and what complications resulted from this when deformity, fracture, or total joint replacement had occurred.³² Although Paget disease did affect their quality of life as measured by physical functioning on the Short Form-12 assessment, these impairments did not seem to affect their outlook, which was as good as or better than that in other people their age.

Metabolic complications

Metabolic complications of Paget disease are rare today but can occur in an elderly patient who has active, polyostotic (multibone) disease.³³ The accelerated rate of bone remodeling and the increased vascularity of pagetic bone have been reported to lead to high-output heart failure. In theory, treatment should ease this by diminishing blood flow to pagetic bone and restoring bone turnover to more normal levels.³⁴

Hypercalcemia can occur when patients with Paget disease are immobilized for any reason, and there is probably a higher incidence of renal stones in patients with Paget disease.^35,36

Malignant complications

Osteosarcoma rarely arises in pagetic bone. Yet Paget disease may account for a significant number of cases of this cancer in the elderly.³⁷ In these cases, osteosarcoma is presumed to be driven by a second genetic mutation, has a genetic signature distinct from that in osteosarcomas occurring in youth, and is quite resistant to treatment.³⁸ In Scandinavia and Japan, where Paget disease is rare, the second peak of osteosarcoma that occurs with aging seems muted as well.^39,40 These cancers present with pain, soft-tissue swelling, and variable elevations in serum alkaline phosphatase. Investigations to date suggest that pagetic lesions and osteosarcomas arising in pagetic bone are probably both driven to some extent by stromal cells overexpressing RANK ligand and may not represent defects intrinsic to the osteoclast.⁴¹

Giant-cell tumors of bone are also rare but can arise in pagetic bone. A cluster of cases was reported in Avellino and other towns of southern Italy.⁴² Again, the lesions occur in older individuals and in different sites than those seen in the benign giant-cell tumors recorded in patients without Paget disease.

Metastases from lymphomas, prostate cancer, and breast cancer certainly occur in bone, but rarely in pagetic sites.⁴³ A recent case study noted that patients with prostate cancer who also had Paget disease had a later onset of metastasis to bone than patients without coincident Paget disease.⁴⁴

A THOUGHTFUL ASSESSMENT

Evaluating a patient with Paget disease requires a thoughtful assessment of its musculoskeletal consequences in an aging skeleton. Pain in Paget disease is often multifactorial. In the elderly, end-stage degenerative disease of the spine, hip, and knees, mechanical instability, compression fractures of the spine, and neuropathies may compound the clinical picture. Therefore, a thorough evaluation is required to plan effective therapy.

Alkaline phosphatase and other markers

A screening serum alkaline phosphatase level is usually sufficient to measure bone turnover. Produced by osteoblasts, alkaline phosphatase is a marker of bone formation, but an imperfect one. Often it is elevated in active Paget disease—but not always.⁴⁵ Many patients have normal serum alkaline phosphatase levels, particularly if they have monostotic (single-bone) disease. It is unclear why, in a disorder marked by accelerated bone remodeling, the biochemical markers are inconsistent measures of bone turnover.

Research into biochemical markers of Paget disease has had two aims: to identify the single best marker for baseline assessment of pagetic bone activity and to find out whether this measurement responds to therapy.^46,47 Measures of bone formation such as bone-specific alkaline phosphatase, osteocalcin, and the procollagen type I peptides, and measures of bone resorption including the pyridinolines, hydroxyproline, and cross-linked collagens, have been analyzed as markers of bone remodeling and show no real advantage over the serum alkaline phosphatase level as reflections of bone turnover. As alkaline phosphatase measurement is inexpensive, available, and reliable, it should be used preferentially, with gamma-glutamyl transpeptidate or 5′ nucleotidase confirming the source as either liver or bone. Readers are directed to a recent review in which the utility of these markers is explored in more detail.⁴⁸

Imaging studies

Bone scans can give us an idea of the extent, location, and general activity of the disease (Figure 3). Uptake is avid in affected bones, beginning in the subchondral region and spreading throughout the bone. Bone scans can be particularly useful in defining sites of active disease when the serum alkaline phosphatase level is normal.

Plain radiography of the affected bones outlines the anatomy of the problem and gives some insight into the cause of pain (Figure 3).

Computed tomography or magnetic resonance imaging may prove useful in cases of spinal stenosis, cauda equina syndrome, compression fractures, or suspected malignancy (Figure  4), but these studies are expensive and generally are not needed.

Radiographic features. Paget disease is presumed to be a disease of the osteoclast, and the earliest lesion is described as lytic. In my own experience, it is unusual to see a purely lytic lesion, although sometimes the disease presents in the skull in this way—osteoporosis circumscripta—or in the femur or tibia with an advancing edge of pure osteolysis.

More often, one sees evidence of both resorption by osteoclasts and formation by osteoblasts, reflecting the coupling of these two processes in this disease. Radiographic findings on plain films are usually definitive, showing enlargement of the affected bone, deformity, coarsened trabeculae, and thickened cortices with tunneling (Figure 5).⁴⁹ In weightbearing bones, pseudofractures may stud the convex surface. These incongruities of bone may persist for years, heralding fracture only when there is focal pain (Figure 6).⁵⁰

Biopsy is infrequently needed

If these diagnostic findings are not present, then biopsy is indicated. In the United Sates and Canada, where Paget disease is fairly common, biopsy is infrequently needed and is usually reserved for situations in which the differential diagnosis includes cancer, as when the cortex cannot be clearly visualized, the lesions are atypical in pattern or location, or there is a single sclerotic vertebral body on imaging.⁵¹

The other indication for biopsy is a “new” pagetic lesion. For reasons unknown, the pattern of skeletal involvement in Paget disease tends to be stable throughout the patient’s lifetime. This is another reason why a baseline bone scan is useful.

TREATMENT WITH BISPHOSPHONATES

Treatment of Paget disease today relies for the most part on the new generation of nitrogen-containing bisphosphonates. As a class, these are antiresorptive agents that inhibit osteoclasts; in this way they slow bone remodeling and enhance the deposition of normal lamellar bone. Their clinical efficacy in Paget disease, coupled with the observation that the earliest lesion in Paget disease is lytic, underscores the principle that Paget disease is a disorder of the osteoclast.

Oral bisphosphonates

Etidronate, approved in 1977, was the first bisphosphonate licensed to treat Paget disease, and it remains available for this indication in the United States. Used in 6-month regimens, it lowers the serum alkaline phosphatase level in some patients, but it has a narrow therapeutic margin. Drug-induced osteomalacia and worsening lytic lesions and fractures in weight-bearing bones are some of the complications.⁵² When the nitrogen-containing bisphosphonates were developed, they proved to be more potent antiresorptive agents that pose less risk of mineralization defects at prescribed doses.

Alendronate, approved in 1995, is an oral nitrogen-containing bisphosphonate that is effective in treating Paget disease.⁵³ Alendronate is now available in the United States only through special programs (eg, the CVS ProCare Program); the paperwork required to secure this drug is onerous, so the drug is used infrequently. Studies in Paget disease showed that it normalizes the serum alkaline phosphatase level, improves the radiographic appearance, and eases pain in many patients.⁵⁴ The dosage is 40 mg daily for 6 months.

Risedronate, approved in 1998, is another oral nitrogen-containing bisphosphonate and is comparable to alendronate in efficacy.⁵⁵ The dosage is 30 mg daily for 2 months.

Tiludronate is another oral bisphosphonate with a different mechanism of action from the nitrogen-containing bisphosphonates.⁵⁶ It is safe, often effective, but less potent than the newer agents.

The oral bisphosphonates are well tolerated, with few side effects other than gastrointestinal distress. As a class, they are poorly absorbed and so must be taken fasting with a full glass of water on rising, after which the patient should remain upright without food or drink for 30 to 60 minutes. This is a nuisance for elderly patients already on multiple medications and thus makes intravenous agents appealing.

Intravenous bisphosphonates

Pamidronate was approved in 1994. It is quite effective in many patients with Paget disease. There is no consensus around the world on dosing, with regimens ranging from 30 mg to 90 mg or more intravenously in divided doses given over 2 to 4 hours from once a day to once a week. In the United States, 30 mg is given over 4 hours on 3 consecutive days. Resistance to pamidronate has been described; the mechanism is unknown.

Zoledronic acid is a nitrogen-containing bisphosphonate. It is given as a single infusion over 15 minutes, and re-treatment may not be necessary for years. A randomized clinical trial in 2005 demonstrated the efficacy of zoledronic acid 5 mg by infusion compared with oral risedronate in the treatment of Paget disease.⁵⁷ In observational extension studies lasting as long as 6.5 years, zoledronic acid has been shown to be superior to risedronate in terms of the proportion of patients experiencing a sustained clinical remission.⁵⁸

While there are many bisphosphonates on the market, an infusion of 5 mg of zoledronic acid seems optimal in most patients who do not have a contraindication or an aversion to intravenous therapy. It tends to normalize the serum alkaline phosphatase level quickly and to leave more patients in sustained biochemical remission than do older bisphosphonates, as noted above. It also tends to be more effective in normalizing the serum alkaline phosphatase level when a patient has used other bisphosphonates in the past or has become resistant to them.

Bisphosphonates reduce bone turnover but do not correct deformities

In randomized clinical trials, bisphosphonates have been shown to restore bone remodeling to more normal levels, to ease pain from pagetic bone, to lower the serum alkaline phosphatase level, and to heal radiographic lesions, but these drugs have not been proven to prevent progression of deformity or to restore the structural integrity of bone (Figure 6).

The Paget’s Disease: Randomized Trial of Intensive Versus Symptomatic Management (PRISM), in 1,324 people with Paget disease in the United Kingdom, showed no difference in the incidence of fracture, orthopedic surgery, quality of life, or hearing thresholds over 2 to 5 years in patients treated with bisphosphonates vs those treated symptomatically, despite a significant difference in serum alkaline phosphatase in the two groups (P < .001).⁵⁹

In the observational extension study of zoledronic acid described above,⁵⁸ three of four fractures occurred in the group treated with zoledronic acid, echoing the findings of the PRISM study.

Adverse effects of bisphosphonates

The more potent the bisphosphonate is as an antiresorptive agent, the more it suppresses normal bone remodeling, which can lead to osteonecrosis of the jaw and to atypical femoral fractures.^60,61 These complications are unusual in patients with Paget disease because the treatment is intermittent. Sometimes a single dose of zoledronic acid or one course of risedronate or alendronate will last for years.

All the nitrogen-containing bisphosphonates, particularly zoledronic acid, may provoke flulike symptoms of fever, arthralgias, and bone pain. This effect is self-limited, resolves in days, and does not tend to recur. Bone pain may be more sustained, but this also passes, and within weeks the antiresorptive process has abated and pagetic bone pain will ease. Atrial fibrillation is not an anticipated complication of treatment with a bisphosphonate.⁶² The risk of esophageal cancer is not confirmed at this time.⁶³ Other rare complications of the bisphosphonates include iritis, acute renal failure, and allergy.

Bisphosphonates are not approved for use in patients with creatinine clearance less than 30 mL/min, or in pregnancy.

Other treatments

Calcitonin, an older agent, can still be useful in easing the pain of Paget disease, healing bone lesions, and reducing the metabolic activity of pagetic bone in patients who cannot receive bisphosphonates. It is given by injection in doses of 50 to 100 IU daily or every other day. Although unlikely to effect a sustained clinical remission, calcitonin remains a safe, well-tolerated, and well-studied medication in Paget disease and is approved for this indication.^64,65

Denosumab has not been formally studied in Paget disease, but a recent case report indicated it was effective.⁶⁶

A conservative strategy

Guidelines for treating Paget disease have been written at various times in many countries, including Italy (2007),⁶⁷ the United Kingdom (2004),⁶⁸ Japan (2006),⁶⁹ and Canada (2007).⁷⁰ Recommendations differ, in part because it is hard to ascertain whether long-term outcomes are improved by treatment, and in part because the prevalence of Paget disease is decreasing and its severity is lessening.^11,12 Some guidelines are outdated, since they do not include the newer bisphosphonates.

If the natural history of untreated Paget disease involves the gradual evolution over more than 20 years of bowing deformities in the lower limbs, rigidity and overgrowth of the spine, and softening and enlargement of the skull, as described by Sir James Paget, then treatment should be initiated in hopes that it will modify the outcome. We have no lens to better focus this question on the effect of treatment on the natural history of the disease. We have the PRISM study, designed before zoledronic acid was approved and only 2 to 5 years in duration. And we have the epidemiologic data demonstrating that most patients have no symptoms during their lifetime.

We see the crippling bone disease described by Sir James Paget so infrequently today in the United States that we forget the profound morbidity that may attend the skeletal changes of Paget disease that were common in the early 20th century. Once the bones of the skull are overgrown, the limbs are bowed, and the degenerative joint disease is present, no medication can reverse these changes. Then, the integrity of the bone is lost, and the vulnerability to fracture, early osteoarthritis, nerve compression syndromes, and hearing loss persist. Understanding these consequences prompts the recommendation of early treatment in patients with Paget disease, in hopes of mitigating disease progression.

Patients with active Paget disease, documented either by an elevated serum alkaline phosphatase or by a bone scan, should be treated with a bisphosphonate if the disease is found in sites where remodeling of bone may lead to complications. Such sites include the skull, spine, and long bones of the lower extremity. Paget disease of bone in the pelvis tends to give little trouble (Figure 2) unless it is proximal to a joint, when pain and early arthritis may result. Treatment is safe and, I think, prudent to undertake in any person over age 55 with active disease. To prevent hypocalcemia during treatment, all patients should be repleted with vitamin D and maintained on calcium 1,200 mg daily through diet or supplements with meals.

Throughout the evaluation and treatment, it is important to remember that pain may not emanate from pagetic bone. If medication for Paget disease proves ineffective in the first few months, analgesics, bracing, walking aids, and operative management⁷¹ are adjunctive therapies to improve the functional status of these patients.

It is a remarkable clinical observation that treatment of Paget disease may rapidly reverse neurologic syndromes, resolve the erythema or warmth overlying active pagetic bone, and diminish the risk of bleeding with surgery. This response to therapy suggests that there is prompt inhibition and apoptosis of the osteoclasts, accompanied by diminished vascularity of bone. Whatever the mechanism, it is worth treating patients who have spinal stenosis, arthritis, and nerve compression syndromes with calcitonin or bisphosphonates before surgical intervention, whenever possible.^34,72

Paget disease of bone is a focal disorder of the aging skeleton that can be asymptomatic or can present with pain, bowing deformities, fractures, or nonspecific rheumatic complaints. Physicians often discover it in asymptomatic patients when serum alkaline phosphatase levels are elevated or as an incidental finding on radiography. Despite evidence of germline mutations and polymorphisms that predispose to Paget disease, the environmental determinants that permit disease expression in older people remain unknown.

A STRIKING GEOGRAPHIC DISTRIBUTION

Researchers have been studying the determinants and distribution of Paget disease ever since Sir James Paget first described it in 1877.¹

Paget disease has a predilection for the axial skeleton, particularly the lumbosacral spine and pelvis, as well as the skull, femur, and tibia.² Knowing this, investigators have used screening plain films of the abdomen (kidney-ureter-bladder views) to estimate its prevalence in different populations, as these images capture the lumbosacral spine, pelvis, and proximal femurs. Other means of assessing prevalence have included autopsy series, questionnaires, and screens for biochemical markers of bone turnover, such as elevated serum alkaline phosphatase from bone.^3–6

Using these methods, Paget disease has been estimated to occur in 1% to 3% of people over age 55, and in as many as 8% of people over age 80 in certain countries.⁷

This disease has a striking geographic distribution, being frequent in Europe, Canada, the United States, Australia, New Zealand, and cities of South America, but rare in Scandinavia and Japan. It seems to be equally rare in other countries of the Far East and in India, Russia, and Africa, although its prevalence in these areas has not been thoroughly investigated.⁸

That it is an ancient disease has been corroborated by excavations in churchyards in Great Britain.^9,10 It may be familial or sporadic, but its expression is delayed until late middle age in most persons, and it does not occur in children. For reasons unclear, the prevalence seems to be decreasing in many countries.^11–13

GENETICS IS NOT THE WHOLE STORY

The variable prevalence of Paget disease in different geographic regions and its sometimes-familial expression suggest a genetic predisposition, environmental factor, or both.

Mutations in SQSTM1

In 2002, scientists investigating a cohort of French Canadian families found a mutation in the SQSTM1 gene that was present in almost 50% of people with familial Paget disease and in 16% of those with sporadic Paget disease.¹⁴ Hocking and his colleagues in the United Kingdom subsequently found the same mutation in 19% of cases of familial Paget disease and in 9% of sporadic cases.¹⁵

Further, investigators noted that the mutation was often present on a conserved haplotype, consistent with a stable genetic change occurring in the affected population.¹⁶ This observation of a “founder effect” dovetailed with the epidemiology of Paget disease,¹⁷ but only with this SQSTM1 mutation.

Throughout Europe, Australia, and the United States, comparable rates of the SQSTM1 mutation were reported in or around the ubiquitin-associated domain. Several specific mutations exist, the most common one being P392L, ie, a prolineto-leucine substitution at amino acid 392. Scientists have tried to correlate severity of disease with genotype, but the findings have been inconsistent.^18–21

Investigations into the mechanism of disease have pointed to the role of p62, the product of SQSTM1, in signaling osteoclast activation via nuclear factor kappa B. Since this initial discovery, polymorphisms in the genes affecting osteoclast maturation, activation, and fusion pathways have been shown to predispose to Paget disease. Examples:

• TNFRSF11A, which codes for receptor activator of nuclear factor kappa B, or RANK
• TNFRSF11B, which codes for osteoprotegerin, or OPG
• CSF1, which codes for macrophage colony-stimulating factor 1, and
• OPTN, which codes for optineurin, a member of the nuclear factor kappa B-modulating protein family.

Clinicians interested in these details can read an excellent review of the pathogenesis of Paget disease.²²

Other possible factors

Although there is good evidence that measles and canine distemper virus can infect osteoclasts and modify their phenotype, there is no good evidence that these infections by themselves cause Paget disease.^23–25 It is, however, tempting to think of these RNA paramyxoviruses as precipitating factors; conceivably, an infectious agent might seed the ends of long bones, accounting for the fixed distribution of Paget disease and its late expression.

Epidemiologic studies from around the world have failed to identify conclusively any environmental exposure that predisposes to Paget disease, although a rural setting, trauma, infection, and milk ingestion have all been proposed.^26–28 It is also possible that as bone ages and the marrow becomes less cellular and more fatty, these changes may permit the disease to develop.

The greatest risk factor for Paget disease is perhaps aging, followed by ancestry and a known family history of it. That genetics is not the whole story is evident by reports of people with SQSTM1 mutations who show no clinical evidence of Paget disease in their old age, and patients with Paget disease who have no SQSTM1 mutation.^20,29

CLINICAL PRESENTATION

Most patients with Paget disease have no symptoms and come to medical attention because of an elevated serum alkaline phosphatase level or characteristic findings on radiographs ordered for other indications.¹¹ Paget disease is the second most common disorder of aging bone after osteoporosis. Yet unlike osteoporosis, which presents as a systemic fragility of bone, the clinical manifestations of Paget disease depend on which bones are affected and how enlarged or misshapen they have become.

Common complications

As a consequence of this abnormal bone remodeling and overgrowth, many patients present with bone pain. Bone deformity, headache, and hearing loss may also occur (Figure 1), as well as fractures and nerve compression syndromes (eg, spinal stenosis, sciatica, cauda equina syndrome).

It is important to remember that “pagetic” bone may not be the source of pain, and that functional impairment caused by degenerative changes at affected sites is common (Figure  2).^30,31

In a study from the New England Registry for Paget’s Disease,³² most patients knew fairly well which bones were affected and what complications resulted from this when deformity, fracture, or total joint replacement had occurred.³² Although Paget disease did affect their quality of life as measured by physical functioning on the Short Form-12 assessment, these impairments did not seem to affect their outlook, which was as good as or better than that in other people their age.

Metabolic complications

Metabolic complications of Paget disease are rare today but can occur in an elderly patient who has active, polyostotic (multibone) disease.³³ The accelerated rate of bone remodeling and the increased vascularity of pagetic bone have been reported to lead to high-output heart failure. In theory, treatment should ease this by diminishing blood flow to pagetic bone and restoring bone turnover to more normal levels.³⁴

Hypercalcemia can occur when patients with Paget disease are immobilized for any reason, and there is probably a higher incidence of renal stones in patients with Paget disease.^35,36

Malignant complications

Osteosarcoma rarely arises in pagetic bone. Yet Paget disease may account for a significant number of cases of this cancer in the elderly.³⁷ In these cases, osteosarcoma is presumed to be driven by a second genetic mutation, has a genetic signature distinct from that in osteosarcomas occurring in youth, and is quite resistant to treatment.³⁸ In Scandinavia and Japan, where Paget disease is rare, the second peak of osteosarcoma that occurs with aging seems muted as well.^39,40 These cancers present with pain, soft-tissue swelling, and variable elevations in serum alkaline phosphatase. Investigations to date suggest that pagetic lesions and osteosarcomas arising in pagetic bone are probably both driven to some extent by stromal cells overexpressing RANK ligand and may not represent defects intrinsic to the osteoclast.⁴¹

Giant-cell tumors of bone are also rare but can arise in pagetic bone. A cluster of cases was reported in Avellino and other towns of southern Italy.⁴² Again, the lesions occur in older individuals and in different sites than those seen in the benign giant-cell tumors recorded in patients without Paget disease.

Metastases from lymphomas, prostate cancer, and breast cancer certainly occur in bone, but rarely in pagetic sites.⁴³ A recent case study noted that patients with prostate cancer who also had Paget disease had a later onset of metastasis to bone than patients without coincident Paget disease.⁴⁴

A THOUGHTFUL ASSESSMENT

Evaluating a patient with Paget disease requires a thoughtful assessment of its musculoskeletal consequences in an aging skeleton. Pain in Paget disease is often multifactorial. In the elderly, end-stage degenerative disease of the spine, hip, and knees, mechanical instability, compression fractures of the spine, and neuropathies may compound the clinical picture. Therefore, a thorough evaluation is required to plan effective therapy.

Alkaline phosphatase and other markers

A screening serum alkaline phosphatase level is usually sufficient to measure bone turnover. Produced by osteoblasts, alkaline phosphatase is a marker of bone formation, but an imperfect one. Often it is elevated in active Paget disease—but not always.⁴⁵ Many patients have normal serum alkaline phosphatase levels, particularly if they have monostotic (single-bone) disease. It is unclear why, in a disorder marked by accelerated bone remodeling, the biochemical markers are inconsistent measures of bone turnover.

Research into biochemical markers of Paget disease has had two aims: to identify the single best marker for baseline assessment of pagetic bone activity and to find out whether this measurement responds to therapy.^46,47 Measures of bone formation such as bone-specific alkaline phosphatase, osteocalcin, and the procollagen type I peptides, and measures of bone resorption including the pyridinolines, hydroxyproline, and cross-linked collagens, have been analyzed as markers of bone remodeling and show no real advantage over the serum alkaline phosphatase level as reflections of bone turnover. As alkaline phosphatase measurement is inexpensive, available, and reliable, it should be used preferentially, with gamma-glutamyl transpeptidate or 5′ nucleotidase confirming the source as either liver or bone. Readers are directed to a recent review in which the utility of these markers is explored in more detail.⁴⁸

Imaging studies

Bone scans can give us an idea of the extent, location, and general activity of the disease (Figure 3). Uptake is avid in affected bones, beginning in the subchondral region and spreading throughout the bone. Bone scans can be particularly useful in defining sites of active disease when the serum alkaline phosphatase level is normal.

Plain radiography of the affected bones outlines the anatomy of the problem and gives some insight into the cause of pain (Figure 3).

Computed tomography or magnetic resonance imaging may prove useful in cases of spinal stenosis, cauda equina syndrome, compression fractures, or suspected malignancy (Figure  4), but these studies are expensive and generally are not needed.

Radiographic features. Paget disease is presumed to be a disease of the osteoclast, and the earliest lesion is described as lytic. In my own experience, it is unusual to see a purely lytic lesion, although sometimes the disease presents in the skull in this way—osteoporosis circumscripta—or in the femur or tibia with an advancing edge of pure osteolysis.

More often, one sees evidence of both resorption by osteoclasts and formation by osteoblasts, reflecting the coupling of these two processes in this disease. Radiographic findings on plain films are usually definitive, showing enlargement of the affected bone, deformity, coarsened trabeculae, and thickened cortices with tunneling (Figure 5).⁴⁹ In weightbearing bones, pseudofractures may stud the convex surface. These incongruities of bone may persist for years, heralding fracture only when there is focal pain (Figure 6).⁵⁰

Biopsy is infrequently needed

If these diagnostic findings are not present, then biopsy is indicated. In the United Sates and Canada, where Paget disease is fairly common, biopsy is infrequently needed and is usually reserved for situations in which the differential diagnosis includes cancer, as when the cortex cannot be clearly visualized, the lesions are atypical in pattern or location, or there is a single sclerotic vertebral body on imaging.⁵¹

The other indication for biopsy is a “new” pagetic lesion. For reasons unknown, the pattern of skeletal involvement in Paget disease tends to be stable throughout the patient’s lifetime. This is another reason why a baseline bone scan is useful.

TREATMENT WITH BISPHOSPHONATES

Treatment of Paget disease today relies for the most part on the new generation of nitrogen-containing bisphosphonates. As a class, these are antiresorptive agents that inhibit osteoclasts; in this way they slow bone remodeling and enhance the deposition of normal lamellar bone. Their clinical efficacy in Paget disease, coupled with the observation that the earliest lesion in Paget disease is lytic, underscores the principle that Paget disease is a disorder of the osteoclast.

Oral bisphosphonates

Etidronate, approved in 1977, was the first bisphosphonate licensed to treat Paget disease, and it remains available for this indication in the United States. Used in 6-month regimens, it lowers the serum alkaline phosphatase level in some patients, but it has a narrow therapeutic margin. Drug-induced osteomalacia and worsening lytic lesions and fractures in weight-bearing bones are some of the complications.⁵² When the nitrogen-containing bisphosphonates were developed, they proved to be more potent antiresorptive agents that pose less risk of mineralization defects at prescribed doses.

Alendronate, approved in 1995, is an oral nitrogen-containing bisphosphonate that is effective in treating Paget disease.⁵³ Alendronate is now available in the United States only through special programs (eg, the CVS ProCare Program); the paperwork required to secure this drug is onerous, so the drug is used infrequently. Studies in Paget disease showed that it normalizes the serum alkaline phosphatase level, improves the radiographic appearance, and eases pain in many patients.⁵⁴ The dosage is 40 mg daily for 6 months.

Risedronate, approved in 1998, is another oral nitrogen-containing bisphosphonate and is comparable to alendronate in efficacy.⁵⁵ The dosage is 30 mg daily for 2 months.

Tiludronate is another oral bisphosphonate with a different mechanism of action from the nitrogen-containing bisphosphonates.⁵⁶ It is safe, often effective, but less potent than the newer agents.

The oral bisphosphonates are well tolerated, with few side effects other than gastrointestinal distress. As a class, they are poorly absorbed and so must be taken fasting with a full glass of water on rising, after which the patient should remain upright without food or drink for 30 to 60 minutes. This is a nuisance for elderly patients already on multiple medications and thus makes intravenous agents appealing.

Intravenous bisphosphonates

Pamidronate was approved in 1994. It is quite effective in many patients with Paget disease. There is no consensus around the world on dosing, with regimens ranging from 30 mg to 90 mg or more intravenously in divided doses given over 2 to 4 hours from once a day to once a week. In the United States, 30 mg is given over 4 hours on 3 consecutive days. Resistance to pamidronate has been described; the mechanism is unknown.

Zoledronic acid is a nitrogen-containing bisphosphonate. It is given as a single infusion over 15 minutes, and re-treatment may not be necessary for years. A randomized clinical trial in 2005 demonstrated the efficacy of zoledronic acid 5 mg by infusion compared with oral risedronate in the treatment of Paget disease.⁵⁷ In observational extension studies lasting as long as 6.5 years, zoledronic acid has been shown to be superior to risedronate in terms of the proportion of patients experiencing a sustained clinical remission.⁵⁸

While there are many bisphosphonates on the market, an infusion of 5 mg of zoledronic acid seems optimal in most patients who do not have a contraindication or an aversion to intravenous therapy. It tends to normalize the serum alkaline phosphatase level quickly and to leave more patients in sustained biochemical remission than do older bisphosphonates, as noted above. It also tends to be more effective in normalizing the serum alkaline phosphatase level when a patient has used other bisphosphonates in the past or has become resistant to them.

Bisphosphonates reduce bone turnover but do not correct deformities

In randomized clinical trials, bisphosphonates have been shown to restore bone remodeling to more normal levels, to ease pain from pagetic bone, to lower the serum alkaline phosphatase level, and to heal radiographic lesions, but these drugs have not been proven to prevent progression of deformity or to restore the structural integrity of bone (Figure 6).

The Paget’s Disease: Randomized Trial of Intensive Versus Symptomatic Management (PRISM), in 1,324 people with Paget disease in the United Kingdom, showed no difference in the incidence of fracture, orthopedic surgery, quality of life, or hearing thresholds over 2 to 5 years in patients treated with bisphosphonates vs those treated symptomatically, despite a significant difference in serum alkaline phosphatase in the two groups (P < .001).⁵⁹

In the observational extension study of zoledronic acid described above,⁵⁸ three of four fractures occurred in the group treated with zoledronic acid, echoing the findings of the PRISM study.

Adverse effects of bisphosphonates

The more potent the bisphosphonate is as an antiresorptive agent, the more it suppresses normal bone remodeling, which can lead to osteonecrosis of the jaw and to atypical femoral fractures.^60,61 These complications are unusual in patients with Paget disease because the treatment is intermittent. Sometimes a single dose of zoledronic acid or one course of risedronate or alendronate will last for years.

All the nitrogen-containing bisphosphonates, particularly zoledronic acid, may provoke flulike symptoms of fever, arthralgias, and bone pain. This effect is self-limited, resolves in days, and does not tend to recur. Bone pain may be more sustained, but this also passes, and within weeks the antiresorptive process has abated and pagetic bone pain will ease. Atrial fibrillation is not an anticipated complication of treatment with a bisphosphonate.⁶² The risk of esophageal cancer is not confirmed at this time.⁶³ Other rare complications of the bisphosphonates include iritis, acute renal failure, and allergy.

Bisphosphonates are not approved for use in patients with creatinine clearance less than 30 mL/min, or in pregnancy.

Other treatments

Calcitonin, an older agent, can still be useful in easing the pain of Paget disease, healing bone lesions, and reducing the metabolic activity of pagetic bone in patients who cannot receive bisphosphonates. It is given by injection in doses of 50 to 100 IU daily or every other day. Although unlikely to effect a sustained clinical remission, calcitonin remains a safe, well-tolerated, and well-studied medication in Paget disease and is approved for this indication.^64,65

Denosumab has not been formally studied in Paget disease, but a recent case report indicated it was effective.⁶⁶

A conservative strategy

Guidelines for treating Paget disease have been written at various times in many countries, including Italy (2007),⁶⁷ the United Kingdom (2004),⁶⁸ Japan (2006),⁶⁹ and Canada (2007).⁷⁰ Recommendations differ, in part because it is hard to ascertain whether long-term outcomes are improved by treatment, and in part because the prevalence of Paget disease is decreasing and its severity is lessening.^11,12 Some guidelines are outdated, since they do not include the newer bisphosphonates.

If the natural history of untreated Paget disease involves the gradual evolution over more than 20 years of bowing deformities in the lower limbs, rigidity and overgrowth of the spine, and softening and enlargement of the skull, as described by Sir James Paget, then treatment should be initiated in hopes that it will modify the outcome. We have no lens to better focus this question on the effect of treatment on the natural history of the disease. We have the PRISM study, designed before zoledronic acid was approved and only 2 to 5 years in duration. And we have the epidemiologic data demonstrating that most patients have no symptoms during their lifetime.

We see the crippling bone disease described by Sir James Paget so infrequently today in the United States that we forget the profound morbidity that may attend the skeletal changes of Paget disease that were common in the early 20th century. Once the bones of the skull are overgrown, the limbs are bowed, and the degenerative joint disease is present, no medication can reverse these changes. Then, the integrity of the bone is lost, and the vulnerability to fracture, early osteoarthritis, nerve compression syndromes, and hearing loss persist. Understanding these consequences prompts the recommendation of early treatment in patients with Paget disease, in hopes of mitigating disease progression.

Patients with active Paget disease, documented either by an elevated serum alkaline phosphatase or by a bone scan, should be treated with a bisphosphonate if the disease is found in sites where remodeling of bone may lead to complications. Such sites include the skull, spine, and long bones of the lower extremity. Paget disease of bone in the pelvis tends to give little trouble (Figure 2) unless it is proximal to a joint, when pain and early arthritis may result. Treatment is safe and, I think, prudent to undertake in any person over age 55 with active disease. To prevent hypocalcemia during treatment, all patients should be repleted with vitamin D and maintained on calcium 1,200 mg daily through diet or supplements with meals.

Throughout the evaluation and treatment, it is important to remember that pain may not emanate from pagetic bone. If medication for Paget disease proves ineffective in the first few months, analgesics, bracing, walking aids, and operative management⁷¹ are adjunctive therapies to improve the functional status of these patients.

It is a remarkable clinical observation that treatment of Paget disease may rapidly reverse neurologic syndromes, resolve the erythema or warmth overlying active pagetic bone, and diminish the risk of bleeding with surgery. This response to therapy suggests that there is prompt inhibition and apoptosis of the osteoclasts, accompanied by diminished vascularity of bone. Whatever the mechanism, it is worth treating patients who have spinal stenosis, arthritis, and nerve compression syndromes with calcitonin or bisphosphonates before surgical intervention, whenever possible.^34,72

In the past several years, three new oral anticoagulants—dabigatran etexilate (Pradaxa), rivaroxaban (Xarelto), and apixaban (Eliquis)—have been approved for use in the United States. These long-awaited agents are appealing because they are easy to use, do not require laboratory monitoring, and have demonstrated equivalence, or in some cases, superiority to warfarin in preventing stroke or systemic embolism in at-risk populations.^1–4 However, unlike warfarin, they have no specific reversal agents. How then should one manage spontaneous bleeding problems and those due to drug overdose, and how can we quickly reverse anticoagulation if emergency surgery is needed?

For these reasons, physicians and patients have been wary of these agents. However, with a systematic approach based on an understanding of the properties of these drugs, the appropriate use and interpretation of coagulation tests, and awareness of available therapeutic strategies, physicians can more confidently provide care for patients who require urgent reversal of anticoagulant effects.

Here, we review the available literature and suggest practical strategies for management based on an understanding of the pharmacokinetic and pharmacodynamic effects of these drugs and our current knowledge of the coagulation tests.

NEED FOR ANTICOAGULANTS

Anticoagulants are important in preventing systemic embolization in patients with atrial fibrillation and preventing pulmonary embolism in patients with venous thromboembolism.

And the numbers are staggering. The estimated prevalence of atrial fibrillation in the United States was 3.03 million in 2005 and is projected to increase to 7.56 million by 2050.⁵ Ischemic stroke is the most serious complication of atrial fibrillation, which accounts for 23.5% of strokes in patients ages 80 through 89 according to Framingham data.⁶ Venous thromboembolism accounts for 900,000 incident or recurrent fatal and nonfatal events in the United States yearly.⁷

HOW THE NEW AGENTS BLOCK COAGULATION

Thrombin (factor IIa), a serine protease, is central to the process of clot formation during hemostasis. It activates factors V, VIII, and XI (thus generating more thrombin), catalyzes the conversion of fibrinogen to fibrin, and stimulates platelet aggregation. Its role in the final steps of the coagulation cascade has made it a target for new direct thrombin inhibitors such as dabigatran.

Factor Xa is a serine protease that plays a central role in the coagulation cascade. It is a desirable target for anticoagulation because it is the convergence point for the extrinsic and the intrinsic coagulation pathways. It converts prothrombin to thrombin. Rivaroxaban and apixaban are direct factor Xa inhibitors (Figure 1).

Dabigatran, a direct thrombin inhibitor

Dabigatran etexilate is a synthetic, orally available prodrug that is rapidly absorbed and converted by esterases to its active form, dabigatran, a potent direct inhibitor of both free thrombin and clot-bound thrombin.⁸

Plasma levels of dabigatran peak within 2 hours of administration, and its half-life is 14 to 17 hours.⁹ Dabigatran is eliminated mainly via the kidneys, with more that 80% of the drug excreted unchanged in the urine (Table 1).

Rivaroxaban, a factor Xa inhibitor

Rivaroxaban is a potent, selective, direct factor Xa inhibitor.

Plasma levels of rivaroxaban peak 2 to 3 hours after administration, and it is cleared with a terminal half-life of 7 to 11 hours.^10,11

Rivaroxaban is eliminated by the kidneys and in the feces. The kidneys eliminate one-third of the active drug unchanged and another one-third as inactive metabolites. The remaining one-third is metabolized by the liver and then excreted in the feces. Rivaroxaban has a predictable and dose-dependent pharmacodynamic and pharmacokinetic profile that is not affected by age, sex, or body weight (Table 1).¹²

Apixaban, an oral factor Xa inhibitor

Apixaban is a selective, direct oral factor Xa inhibitor.

Plasma levels of apixaban peak about 3 hours after administration, and its terminal half-life is 8 to 14 hours.¹³ Apixaban is eliminated by oxidative metabolism, by the kidney, and in the feces. It has predictable pharmacodynamic and pharmacokinetic profiles and has the least renal dependence of the three agents (Table 1).

THE NEW ORAL ANTICOAGULANTS AND BLOOD COAGULATION ASSAYS

Assessment of the anticoagulant activity of the new oral anticoagulants is not necessary in routine clinical practice, but it may be useful in planning intervention in patients with major bleeding, those with drug overdose, or those who need emergency surgery.

The activated partial thromboplastin time

The activated partial thromboplastin time (aPTT) is a measure of the activity of the intrinsic pathway of the coagulation cascade.

Dabigatran. There is a curvilinear relationship between the aPTT and the plasma concentration of dabigatran and other direct thrombin inhibitors, although the aPTT prolongation appears to vary with different reagents and coagulometers.^9,14,15 However, Stangier et al⁹ found a linear relationship between the aPTT and the square root of the dabigatran plasma concentration.

Rivaroxaban prolongs the aPTT in a dose-dependent manner, but there is no standard for calibration of this assay. Hence, the aPTT is not recommended for monitoring rivaroxaban in clinical practice.

Apixaban may also prolong the aPTT, but there are limited data on its reactivity with different reagents.

The prothrombin time and international normalized ratio

The prothrombin time and international normalized ratio (INR) are measures of the extrinsic pathway of the coagulation cascade.

Dabigatran. The INR has a linear response to the dabigatran concentration, but it is insensitive.⁹ Hence, it is not suitable for monitoring the anticoagulant effects of direct thrombin inhibitors.

Rivaroxaban. The prothrombin time correlates strongly with the plasma concentration of rivaroxaban in healthy trial participants¹¹ and in patients undergoing total hip arthroplasty or total knee arthroplasty.¹⁶ Samama et al¹⁷ noted that, unlike with vitamin K antagonists, the INR cannot be used to monitor patients on rivaroxaban because the prothrombin time results varied with different reagents. They used a standard calibration curve to express the prothrombin time results in plasma concentrations of rivaroxaban rather than in seconds or the INR.

Apixaban increases the INR in a dose-dependent manner.¹⁸ Its effect on different reagents remains unknown.

The thrombin time

The thrombin time reflects the activity of thrombin in the plasma. The amount of thrombin and the concentration of thrombin inhibitors in the plasma sample determine the time to clot formation.

Dabigatran. The thrombin time displays a linear dose-response to dabigatran, but only over the range of therapeutic concentrations. At a dabigatran concentration greater than 600 ng/mL, the test often exceeds the maximum measurement time of coagulometers.⁹ Hence, this test is too sensitive for emergency monitoring, especially in cases of drug overdose. However, it is well suited for determining if any dabigatran is present.

Rivaroxaban and apixaban have no effect on the thrombin time.

The Hemoclot direct thrombin inhibitor assay and dabigatran

The Hemoclot direct thrombin inhibitor assay (Hyphen BioMed, France) is a sensitive diluted thrombin time assay that can be used for quantitative measurement of dabigatran activity in plasma. This test is based on inhibition of a constant amount of highly purified human alpha-thrombin by adding it to diluted test plasma (1:8 to 1:20) mixed with normal pooled human plasma.^19,20

Stangier et al¹⁹ found that the Hemoclot assay was suitable for calculating a wide range of dabigatran concentrations up to 4,000 nmol/L (1,886 ng/mL). Although this finding has not been confirmed in larger studies, this test may provide a rapid and accurate assessment of dabigatran’s anticoagulant activity in cases of emergency surgery or overdose.

The ecarin clotting time and dabigatran

The ecarin clotting time is a measure of the activity of direct thrombin inhibitors, but not the factor Xa inhibitors.

Ecarin is a highly purified metalloprotease isolated from the venom of a snake, Echis carinatus, and it generates meizothrombin from prothrombin.²¹ Meizothrombin facilitates clot formation by converting fibrinogen to fibrin and, like thrombin, it can be inactivated by direct thrombin inhibitors, thereby prolonging the clotting time.

The limitations of the ecarin clotting time include dependence on the plasma levels of fibrinogen and prothrombin.

The ecarin chromogenic assay and dabigatran

The ecarin chromogenic assay is an improvement on the principle of the ecarin clotting time that can be used to measure the activity of direct thrombin inhibitors.²² In this test, ecarin is added to a plasma sample to generate meizothrombin, and the amidolytic activity of meizothrombin towards a chromogenic substrate is then determined.

Results of the ecarin chromogenic assay are not influenced by the levels of fibrinogen or prothrombin. Another advantage is that this assay can be used in automated and manual analyzers, thus enabling its use at the bedside. However, to our knowledge, it is not being regularly used to monitor direct thrombin inhibitors in the clinical setting, and there is no standard calibration of the ecarin clotting time method.

Assays of factor Xa activity

A variety of assays to monitor the anticoagulant activity of factor Xa inhibitors have been proposed.^23–25 All measure inhibition of the activity of factor Xa using methods similar to those used in monitoring heparin levels. All require calibrators with a known concentration of the Xa inhibitor; many are easily adapted for laboratories currently providing measurement of factor Xa inhibition from heparin.²³ These assays have been suggested as a better indicator of plasma concentration of factor Xa inhibitor drugs than the prothrombin time.²⁵

CONTROLLING BLEEDING IN PATIENTS ON THE NEW ORAL ANTICOAGULANTS

Bleeding is an anticipated adverse event in patients taking anticoagulants. It is associated with significant morbidity and risk of death.^26,27

Many physicians still have limited experience with using the new oral anticoagulants and managing the attendant bleeding risks. Hence, we recommend that every health institution have a treatment policy or algorithm to guide all clinical staff in the management of such emergencies.

Prevention of bleeding

Management of bleeding from these agents should begin with preventing bleeding in the first place.

The physician should adhere to the recommended dosages of these medications. Studies have shown that the plasma concentration of these drugs and the risk of bleeding increase with increasing dosage.^1,28,29

In addition, these medications should be used for the shortest time for which anticoagulation is required, especially when used for preventing deep vein thrombosis. Prolonged use increases the risk of bleeding.^30,31

Most patients who need anticoagulation have comorbidities such as heart failure, renal failure, diabetes mellitus, and hypertension. Although the kidneys play a major role in the excretion of dabigatran and, to some extent, rivaroxaban and apixaban, patients with severe renal impairment were excluded from the major trials of all three drugs.^1–3 Hence, to avoid excessive drug accumulation and bleeding, these medications should not be used in such patients pending further studies. Further, patients taking these medications should be closely followed to detect new clinical situations, such as acute renal failure, that will necessitate their discontinuation or dose adjustment.

If a patient taking a new oral anticoagulant needs to undergo elective surgery, it is important to temporarily discontinue the drug, assess the risk of bleeding, and test for renal impairment.

Renal impairment is particularly relevant in the case of dabigatran, since more than 80% of the unchanged drug is cleared by the kidneys. Decreasing the dose, prolonging the dosing interval, or both have been suggested as means to reduce the risk of bleeding in patients with renal impairment who are taking dabigatran.^32,33 Patients with normal renal function undergoing low-risk surgery should discontinue dabigatran at least 24 hours before the surgery. If the creatinine clearance is 31 to 50 mL/min, inclusively, the last dose should be at least 48 hours before the procedure for low-risk surgery, and 4 days before a procedure that poses a high risk of bleeding.^32–34 Some experts have given the same recommendations for rivaroxaban and apixaban (Table 2).³⁴

The aPTT and prothrombin time are readily available tests, but they cannot determine the residual anticoagulant effects of dabigatran, rivaroxaban, or apixaban. However, in many (but not all) cases, a normal aPTT suggests that the hemostatic function is not impaired by dabigatran, and a normal prothrombin time or an absence of anti-factor Xa activity would similarly exclude hemostatic dysfunction caused by rivaroxaban or apixaban. These tests are potentially useful as adjuncts before surgical procedures that require complete hemostasis.

Furthermore, a normal thrombin time rules out the presence of a significant amount of dabigatran. Therefore, a normal thrombin time might be particularly useful in a patient undergoing a high-risk intervention such as epidural cannulation or neurosurgery and who is normally receiving dabigatran.

Managing overdose and bleeding complications

Assessing the severity of bleeding is the key to managing bleeding complications (Table 3).

Minor bleeding such as epistaxis and ecchymosis can be managed symptomatically (eg, with nasal packing), perhaps with short-term withdrawal of the anticoagulant. Moderate bleeding such as upper or lower gastrointestinal bleeding can be managed by withdrawal of the anticoagulant, clinical monitoring, blood transfusion if needed, and treatment directed at the etiology.

Major and life-threatening bleeding (eg, intracerebral hemorrhage) requires aggressive treatment in the intensive care unit, withdrawal of the anticoagulant, mechanical compression of the bleeding site if accessible, fluid replacement and blood transfusion as appropriate, and interventional procedures. Nonspecific reversal agents might be considered in patients with major or life-threatening bleeding.

The half-life of dabigatran after multiple doses is approximately 14 to 17 hours and is not dose-dependent.⁹ Hence, if there is no active bleeding after a dabigatran overdose, stopping the drug may be sufficient. Since the pharmacodynamic effect of dabigatran declines in parallel to its plasma concentration, urgent but not emergency surgery may need to be delayed for only about 12 hours from the last dose of dabigatran.

The 2011 American College of Cardiology Foundation/American Heart Association guidelines recommend that patients with severe hemorrhage resulting from dabigatran should receive supportive therapy, including transfusion of fresh-frozen plasma, transfusion of packed red blood cells, or surgical intervention if appropriate.³⁵ However, transfusion of fresh-frozen plasma is debatable because there is no evidence to support its use in this situation. While fresh-frozen plasma may be useful in cases of coagulation factor depletion, it does not effectively reverse inhibition of coagulation factors.³⁶

Off-label use of nonspecific hemostatic agents

To date, no specific agent has been demonstrated to reverse excessive bleeding in patients taking the new oral anticoagulants. However, in view of their procoagulant capabilities, nonspecific hemostatic agents have been suggested for use in reversal of major bleeding resulting from these drugs.^37–39 Examples are:

Recombinant factor VIIa (NovoSeven) initiates thrombin generation by activating factor X.

Four-factor prothrombin complex concentrate (Beriplex, recently approved in the United States) contains relatively large amounts of four nonactive vitamin K-dependent procoagulant factors (factors II, VII, IX, and X) that stimulate thrombin formation.

Three-factor prothrombin complex concentrate (Bebulin VH and Profilnine SD) contains low amounts of nonactive factor VII relative to factors II, IX, and X. In some centers a four-factor equivalent is produced by transfusion of a three-factor product with the addition of small amounts of recombinant factor VIIa or fresh-frozen plasma to replace the missing factor VII.⁴⁰

Activated prothrombin complex concentrate (FEIBA NF) contains activated factor VII and factors II, IX, and X, mainly in nonactivated form.³⁶ Therefore, it combines the effect of both recombinant factor VIIa and four-factor prothrombin complex concentrate.³⁷

Studies of nonspecific hemostatic agents

In a study of rats infused with high doses of dabigatran, van Ryn et al³⁸ observed that activated prothrombin complex concentrate at a dose of 50 or 100 U/kg and recombinant factor VIIa at a dose of 0.1 or 0.5 mg/kg reduced the rat-tail bleeding time in a dose-dependent manner but not the blood loss, compared with controls, even with a higher dose of recombinant factor VIIa (1 mg/kg). Recombinant factor VIIa also reversed the prolonged aPTT induced by dabigatran, whereas activated prothrombin complex concentrate did not. They suggested that recombinant factor VIIa and activated prothrombin complex concentrate may be potential antidotes for dabigatran-induced severe bleeding in humans.

In an ex vivo study of healthy people who took a single dose of dabigatran 150 mg or rivaroxaban 20 mg, Marlu et al³⁷ found that activated prothrombin complex concentrate and four-factor prothrombin complex concentrate could be reasonable antidotes to these drugs.

Dabigatran-associated bleeding after cardiac surgery in humans has been successfully managed with hemodialysis and recombinant factor VIIa, although the efficacy of the latter cannot be individually assessed in the study.⁴¹

In a randomized placebo-controlled trial aimed at reversing rivaroxaban and dabigatran in healthy participants, Eerenberg et al³⁹ showed that four-factor prothrombin complex concentrate at a dose of 50 IU/kg reversed prolongation of the prothrombin time and decreased the endogenous thrombin potential in those who received rivaroxaban, but it failed to reverse the aPTT, the endogenous thrombin potential, and thrombin time in those who received dabigatran.

However, Marlu et al reported that four-factor prothrombin complex concentrate at three doses (12.5 U/kg, 25 U/kg, and 50 U/kg)—or better still, activated prothrombin complex concentrate (40–80 U/kg)—could be a useful antidote to dabigatran.³⁷

It is important to note that the healthy participants in the Eerenberg et al study³⁹ took dabigatran 150 mg twice daily and rivaroxaban 20 mg daily for 2.5 days, whereas those in the Marlu et al study³⁷ took the same dose of each medication, but only once.

The three-factor prothrombin complex concentrate products have been shown to be less effective than four-factor ones in reversing supratherapeutic INRs in patients with warfarin overdose, but whether this will be true with the new oral anticoagulants remains unknown. Furthermore, the four-factor concentrates effectively reversed warfarin-induced coagulopathy and bleeding in patients,⁴² but to our knowledge, the same is yet to be demonstrated in bleeding related to the newer agents.

Other measures

Gastric lavage or the administration of activated charcoal (or in some cases both) may reduce drug absorption if done within 2 or 3 hours of drug ingestion (Table 1). Because it is lipophilic, more than 99.9% of dabigatran etexilate was adsorbed by activated charcoal from water prepared to simulate gastric fluid in an in vitro experiment by van Ryn et al.⁴³ This has not been tested in patients, and no similar study has been done for rivaroxaban or apixaban. However, use of charcoal in cases of recent ingestion, particularly with intentional overdose of these agents, seems reasonable.

Hemodialysis may reverse the anticoagulant effects of dabigatran overdose or severe bleeding because only about 35% of dabigatran is bound to plasma proteins (Table 1). In a single-center study, 50 mg of dabigatran etexilate was given orally to six patients with end-stage renal disease before dialysis, and the mean fraction of the drug removed by the dialyzer was 62% at 2 hours and 68% at 4 hours.³² This study suggests that hemodialysis may be useful to accelerate the removal of the drug in cases of life-threatening bleeding.

Rivaroxaban and apixaban are not dialyzable: the plasma protein binding of rivaroxaban is 95% and that of apixaban is 87%.

FUTURE DIRECTIONS

Because the new oral anticoagulants, unlike warfarin, have a wide therapeutic window, routine anticoagulant monitoring is not needed and might be misleading. However, there are times when monitoring might be useful; at such times, a validated, widely available, easily understood test would be good to have—but we don’t have it—at least not yet.

Therapeutic ranges for the aPTT have been established empirically for heparin in various indications.⁴⁴ Additional study is needed to determine if an appropriate aPTT range can be determined for the new oral anticoagulants, particularly dabigatran.

Similarly, as with low-molecular-weight heparins, anti-factor Xa activity monitoring may become a more available validated means of testing for exposure to rivaroxaban and apixaban. More promising, using concepts derived from the development of the INR for warfarin monitoring,⁴⁵ Tripodi et al⁴⁶ have derived normalized INR-like assays to report rivaroxaban levels. A standardized schema for reporting results is being developed.⁴⁶ Studies are required to determine if and how this assay may be useful. Initial trials in this regard are encouraging.⁴⁷

Finally, the thrombotic risk associated with the use of nonspecific prohemostatic agents is unknown.^37,48 Additional studies are required to standardize their dosages, frequency of administration, and duration of action, as well as to quantify their complications in bleeding patients.

In the past several years, three new oral anticoagulants—dabigatran etexilate (Pradaxa), rivaroxaban (Xarelto), and apixaban (Eliquis)—have been approved for use in the United States. These long-awaited agents are appealing because they are easy to use, do not require laboratory monitoring, and have demonstrated equivalence, or in some cases, superiority to warfarin in preventing stroke or systemic embolism in at-risk populations.^1–4 However, unlike warfarin, they have no specific reversal agents. How then should one manage spontaneous bleeding problems and those due to drug overdose, and how can we quickly reverse anticoagulation if emergency surgery is needed?

For these reasons, physicians and patients have been wary of these agents. However, with a systematic approach based on an understanding of the properties of these drugs, the appropriate use and interpretation of coagulation tests, and awareness of available therapeutic strategies, physicians can more confidently provide care for patients who require urgent reversal of anticoagulant effects.

Here, we review the available literature and suggest practical strategies for management based on an understanding of the pharmacokinetic and pharmacodynamic effects of these drugs and our current knowledge of the coagulation tests.

NEED FOR ANTICOAGULANTS

Anticoagulants are important in preventing systemic embolization in patients with atrial fibrillation and preventing pulmonary embolism in patients with venous thromboembolism.

And the numbers are staggering. The estimated prevalence of atrial fibrillation in the United States was 3.03 million in 2005 and is projected to increase to 7.56 million by 2050.⁵ Ischemic stroke is the most serious complication of atrial fibrillation, which accounts for 23.5% of strokes in patients ages 80 through 89 according to Framingham data.⁶ Venous thromboembolism accounts for 900,000 incident or recurrent fatal and nonfatal events in the United States yearly.⁷

HOW THE NEW AGENTS BLOCK COAGULATION

Thrombin (factor IIa), a serine protease, is central to the process of clot formation during hemostasis. It activates factors V, VIII, and XI (thus generating more thrombin), catalyzes the conversion of fibrinogen to fibrin, and stimulates platelet aggregation. Its role in the final steps of the coagulation cascade has made it a target for new direct thrombin inhibitors such as dabigatran.

Factor Xa is a serine protease that plays a central role in the coagulation cascade. It is a desirable target for anticoagulation because it is the convergence point for the extrinsic and the intrinsic coagulation pathways. It converts prothrombin to thrombin. Rivaroxaban and apixaban are direct factor Xa inhibitors (Figure 1).

Dabigatran, a direct thrombin inhibitor

Dabigatran etexilate is a synthetic, orally available prodrug that is rapidly absorbed and converted by esterases to its active form, dabigatran, a potent direct inhibitor of both free thrombin and clot-bound thrombin.⁸

Plasma levels of dabigatran peak within 2 hours of administration, and its half-life is 14 to 17 hours.⁹ Dabigatran is eliminated mainly via the kidneys, with more that 80% of the drug excreted unchanged in the urine (Table 1).

Rivaroxaban, a factor Xa inhibitor

Rivaroxaban is a potent, selective, direct factor Xa inhibitor.

Plasma levels of rivaroxaban peak 2 to 3 hours after administration, and it is cleared with a terminal half-life of 7 to 11 hours.^10,11

Rivaroxaban is eliminated by the kidneys and in the feces. The kidneys eliminate one-third of the active drug unchanged and another one-third as inactive metabolites. The remaining one-third is metabolized by the liver and then excreted in the feces. Rivaroxaban has a predictable and dose-dependent pharmacodynamic and pharmacokinetic profile that is not affected by age, sex, or body weight (Table 1).¹²

Apixaban, an oral factor Xa inhibitor

Apixaban is a selective, direct oral factor Xa inhibitor.

Plasma levels of apixaban peak about 3 hours after administration, and its terminal half-life is 8 to 14 hours.¹³ Apixaban is eliminated by oxidative metabolism, by the kidney, and in the feces. It has predictable pharmacodynamic and pharmacokinetic profiles and has the least renal dependence of the three agents (Table 1).

THE NEW ORAL ANTICOAGULANTS AND BLOOD COAGULATION ASSAYS

Assessment of the anticoagulant activity of the new oral anticoagulants is not necessary in routine clinical practice, but it may be useful in planning intervention in patients with major bleeding, those with drug overdose, or those who need emergency surgery.

The activated partial thromboplastin time

The activated partial thromboplastin time (aPTT) is a measure of the activity of the intrinsic pathway of the coagulation cascade.

Dabigatran. There is a curvilinear relationship between the aPTT and the plasma concentration of dabigatran and other direct thrombin inhibitors, although the aPTT prolongation appears to vary with different reagents and coagulometers.^9,14,15 However, Stangier et al⁹ found a linear relationship between the aPTT and the square root of the dabigatran plasma concentration.

Rivaroxaban prolongs the aPTT in a dose-dependent manner, but there is no standard for calibration of this assay. Hence, the aPTT is not recommended for monitoring rivaroxaban in clinical practice.

Apixaban may also prolong the aPTT, but there are limited data on its reactivity with different reagents.

The prothrombin time and international normalized ratio

The prothrombin time and international normalized ratio (INR) are measures of the extrinsic pathway of the coagulation cascade.

Dabigatran. The INR has a linear response to the dabigatran concentration, but it is insensitive.⁹ Hence, it is not suitable for monitoring the anticoagulant effects of direct thrombin inhibitors.

Rivaroxaban. The prothrombin time correlates strongly with the plasma concentration of rivaroxaban in healthy trial participants¹¹ and in patients undergoing total hip arthroplasty or total knee arthroplasty.¹⁶ Samama et al¹⁷ noted that, unlike with vitamin K antagonists, the INR cannot be used to monitor patients on rivaroxaban because the prothrombin time results varied with different reagents. They used a standard calibration curve to express the prothrombin time results in plasma concentrations of rivaroxaban rather than in seconds or the INR.

Apixaban increases the INR in a dose-dependent manner.¹⁸ Its effect on different reagents remains unknown.

The thrombin time

The thrombin time reflects the activity of thrombin in the plasma. The amount of thrombin and the concentration of thrombin inhibitors in the plasma sample determine the time to clot formation.

Dabigatran. The thrombin time displays a linear dose-response to dabigatran, but only over the range of therapeutic concentrations. At a dabigatran concentration greater than 600 ng/mL, the test often exceeds the maximum measurement time of coagulometers.⁹ Hence, this test is too sensitive for emergency monitoring, especially in cases of drug overdose. However, it is well suited for determining if any dabigatran is present.

Rivaroxaban and apixaban have no effect on the thrombin time.

The Hemoclot direct thrombin inhibitor assay and dabigatran

The Hemoclot direct thrombin inhibitor assay (Hyphen BioMed, France) is a sensitive diluted thrombin time assay that can be used for quantitative measurement of dabigatran activity in plasma. This test is based on inhibition of a constant amount of highly purified human alpha-thrombin by adding it to diluted test plasma (1:8 to 1:20) mixed with normal pooled human plasma.^19,20

Stangier et al¹⁹ found that the Hemoclot assay was suitable for calculating a wide range of dabigatran concentrations up to 4,000 nmol/L (1,886 ng/mL). Although this finding has not been confirmed in larger studies, this test may provide a rapid and accurate assessment of dabigatran’s anticoagulant activity in cases of emergency surgery or overdose.

The ecarin clotting time and dabigatran

The ecarin clotting time is a measure of the activity of direct thrombin inhibitors, but not the factor Xa inhibitors.

Ecarin is a highly purified metalloprotease isolated from the venom of a snake, Echis carinatus, and it generates meizothrombin from prothrombin.²¹ Meizothrombin facilitates clot formation by converting fibrinogen to fibrin and, like thrombin, it can be inactivated by direct thrombin inhibitors, thereby prolonging the clotting time.

The limitations of the ecarin clotting time include dependence on the plasma levels of fibrinogen and prothrombin.

The ecarin chromogenic assay and dabigatran

The ecarin chromogenic assay is an improvement on the principle of the ecarin clotting time that can be used to measure the activity of direct thrombin inhibitors.²² In this test, ecarin is added to a plasma sample to generate meizothrombin, and the amidolytic activity of meizothrombin towards a chromogenic substrate is then determined.

Results of the ecarin chromogenic assay are not influenced by the levels of fibrinogen or prothrombin. Another advantage is that this assay can be used in automated and manual analyzers, thus enabling its use at the bedside. However, to our knowledge, it is not being regularly used to monitor direct thrombin inhibitors in the clinical setting, and there is no standard calibration of the ecarin clotting time method.

Assays of factor Xa activity

A variety of assays to monitor the anticoagulant activity of factor Xa inhibitors have been proposed.^23–25 All measure inhibition of the activity of factor Xa using methods similar to those used in monitoring heparin levels. All require calibrators with a known concentration of the Xa inhibitor; many are easily adapted for laboratories currently providing measurement of factor Xa inhibition from heparin.²³ These assays have been suggested as a better indicator of plasma concentration of factor Xa inhibitor drugs than the prothrombin time.²⁵

CONTROLLING BLEEDING IN PATIENTS ON THE NEW ORAL ANTICOAGULANTS

Bleeding is an anticipated adverse event in patients taking anticoagulants. It is associated with significant morbidity and risk of death.^26,27

Many physicians still have limited experience with using the new oral anticoagulants and managing the attendant bleeding risks. Hence, we recommend that every health institution have a treatment policy or algorithm to guide all clinical staff in the management of such emergencies.

Prevention of bleeding

Management of bleeding from these agents should begin with preventing bleeding in the first place.

The physician should adhere to the recommended dosages of these medications. Studies have shown that the plasma concentration of these drugs and the risk of bleeding increase with increasing dosage.^1,28,29

In addition, these medications should be used for the shortest time for which anticoagulation is required, especially when used for preventing deep vein thrombosis. Prolonged use increases the risk of bleeding.^30,31

Most patients who need anticoagulation have comorbidities such as heart failure, renal failure, diabetes mellitus, and hypertension. Although the kidneys play a major role in the excretion of dabigatran and, to some extent, rivaroxaban and apixaban, patients with severe renal impairment were excluded from the major trials of all three drugs.^1–3 Hence, to avoid excessive drug accumulation and bleeding, these medications should not be used in such patients pending further studies. Further, patients taking these medications should be closely followed to detect new clinical situations, such as acute renal failure, that will necessitate their discontinuation or dose adjustment.

If a patient taking a new oral anticoagulant needs to undergo elective surgery, it is important to temporarily discontinue the drug, assess the risk of bleeding, and test for renal impairment.

Renal impairment is particularly relevant in the case of dabigatran, since more than 80% of the unchanged drug is cleared by the kidneys. Decreasing the dose, prolonging the dosing interval, or both have been suggested as means to reduce the risk of bleeding in patients with renal impairment who are taking dabigatran.^32,33 Patients with normal renal function undergoing low-risk surgery should discontinue dabigatran at least 24 hours before the surgery. If the creatinine clearance is 31 to 50 mL/min, inclusively, the last dose should be at least 48 hours before the procedure for low-risk surgery, and 4 days before a procedure that poses a high risk of bleeding.^32–34 Some experts have given the same recommendations for rivaroxaban and apixaban (Table 2).³⁴

The aPTT and prothrombin time are readily available tests, but they cannot determine the residual anticoagulant effects of dabigatran, rivaroxaban, or apixaban. However, in many (but not all) cases, a normal aPTT suggests that the hemostatic function is not impaired by dabigatran, and a normal prothrombin time or an absence of anti-factor Xa activity would similarly exclude hemostatic dysfunction caused by rivaroxaban or apixaban. These tests are potentially useful as adjuncts before surgical procedures that require complete hemostasis.

Furthermore, a normal thrombin time rules out the presence of a significant amount of dabigatran. Therefore, a normal thrombin time might be particularly useful in a patient undergoing a high-risk intervention such as epidural cannulation or neurosurgery and who is normally receiving dabigatran.

Managing overdose and bleeding complications

Assessing the severity of bleeding is the key to managing bleeding complications (Table 3).

Minor bleeding such as epistaxis and ecchymosis can be managed symptomatically (eg, with nasal packing), perhaps with short-term withdrawal of the anticoagulant. Moderate bleeding such as upper or lower gastrointestinal bleeding can be managed by withdrawal of the anticoagulant, clinical monitoring, blood transfusion if needed, and treatment directed at the etiology.

Major and life-threatening bleeding (eg, intracerebral hemorrhage) requires aggressive treatment in the intensive care unit, withdrawal of the anticoagulant, mechanical compression of the bleeding site if accessible, fluid replacement and blood transfusion as appropriate, and interventional procedures. Nonspecific reversal agents might be considered in patients with major or life-threatening bleeding.

The half-life of dabigatran after multiple doses is approximately 14 to 17 hours and is not dose-dependent.⁹ Hence, if there is no active bleeding after a dabigatran overdose, stopping the drug may be sufficient. Since the pharmacodynamic effect of dabigatran declines in parallel to its plasma concentration, urgent but not emergency surgery may need to be delayed for only about 12 hours from the last dose of dabigatran.

The 2011 American College of Cardiology Foundation/American Heart Association guidelines recommend that patients with severe hemorrhage resulting from dabigatran should receive supportive therapy, including transfusion of fresh-frozen plasma, transfusion of packed red blood cells, or surgical intervention if appropriate.³⁵ However, transfusion of fresh-frozen plasma is debatable because there is no evidence to support its use in this situation. While fresh-frozen plasma may be useful in cases of coagulation factor depletion, it does not effectively reverse inhibition of coagulation factors.³⁶

Off-label use of nonspecific hemostatic agents

To date, no specific agent has been demonstrated to reverse excessive bleeding in patients taking the new oral anticoagulants. However, in view of their procoagulant capabilities, nonspecific hemostatic agents have been suggested for use in reversal of major bleeding resulting from these drugs.^37–39 Examples are:

Recombinant factor VIIa (NovoSeven) initiates thrombin generation by activating factor X.

Four-factor prothrombin complex concentrate (Beriplex, recently approved in the United States) contains relatively large amounts of four nonactive vitamin K-dependent procoagulant factors (factors II, VII, IX, and X) that stimulate thrombin formation.

Three-factor prothrombin complex concentrate (Bebulin VH and Profilnine SD) contains low amounts of nonactive factor VII relative to factors II, IX, and X. In some centers a four-factor equivalent is produced by transfusion of a three-factor product with the addition of small amounts of recombinant factor VIIa or fresh-frozen plasma to replace the missing factor VII.⁴⁰

Activated prothrombin complex concentrate (FEIBA NF) contains activated factor VII and factors II, IX, and X, mainly in nonactivated form.³⁶ Therefore, it combines the effect of both recombinant factor VIIa and four-factor prothrombin complex concentrate.³⁷

Studies of nonspecific hemostatic agents

In a study of rats infused with high doses of dabigatran, van Ryn et al³⁸ observed that activated prothrombin complex concentrate at a dose of 50 or 100 U/kg and recombinant factor VIIa at a dose of 0.1 or 0.5 mg/kg reduced the rat-tail bleeding time in a dose-dependent manner but not the blood loss, compared with controls, even with a higher dose of recombinant factor VIIa (1 mg/kg). Recombinant factor VIIa also reversed the prolonged aPTT induced by dabigatran, whereas activated prothrombin complex concentrate did not. They suggested that recombinant factor VIIa and activated prothrombin complex concentrate may be potential antidotes for dabigatran-induced severe bleeding in humans.

In an ex vivo study of healthy people who took a single dose of dabigatran 150 mg or rivaroxaban 20 mg, Marlu et al³⁷ found that activated prothrombin complex concentrate and four-factor prothrombin complex concentrate could be reasonable antidotes to these drugs.

Dabigatran-associated bleeding after cardiac surgery in humans has been successfully managed with hemodialysis and recombinant factor VIIa, although the efficacy of the latter cannot be individually assessed in the study.⁴¹

In a randomized placebo-controlled trial aimed at reversing rivaroxaban and dabigatran in healthy participants, Eerenberg et al³⁹ showed that four-factor prothrombin complex concentrate at a dose of 50 IU/kg reversed prolongation of the prothrombin time and decreased the endogenous thrombin potential in those who received rivaroxaban, but it failed to reverse the aPTT, the endogenous thrombin potential, and thrombin time in those who received dabigatran.

However, Marlu et al reported that four-factor prothrombin complex concentrate at three doses (12.5 U/kg, 25 U/kg, and 50 U/kg)—or better still, activated prothrombin complex concentrate (40–80 U/kg)—could be a useful antidote to dabigatran.³⁷

It is important to note that the healthy participants in the Eerenberg et al study³⁹ took dabigatran 150 mg twice daily and rivaroxaban 20 mg daily for 2.5 days, whereas those in the Marlu et al study³⁷ took the same dose of each medication, but only once.

The three-factor prothrombin complex concentrate products have been shown to be less effective than four-factor ones in reversing supratherapeutic INRs in patients with warfarin overdose, but whether this will be true with the new oral anticoagulants remains unknown. Furthermore, the four-factor concentrates effectively reversed warfarin-induced coagulopathy and bleeding in patients,⁴² but to our knowledge, the same is yet to be demonstrated in bleeding related to the newer agents.

Other measures

Gastric lavage or the administration of activated charcoal (or in some cases both) may reduce drug absorption if done within 2 or 3 hours of drug ingestion (Table 1). Because it is lipophilic, more than 99.9% of dabigatran etexilate was adsorbed by activated charcoal from water prepared to simulate gastric fluid in an in vitro experiment by van Ryn et al.⁴³ This has not been tested in patients, and no similar study has been done for rivaroxaban or apixaban. However, use of charcoal in cases of recent ingestion, particularly with intentional overdose of these agents, seems reasonable.

Hemodialysis may reverse the anticoagulant effects of dabigatran overdose or severe bleeding because only about 35% of dabigatran is bound to plasma proteins (Table 1). In a single-center study, 50 mg of dabigatran etexilate was given orally to six patients with end-stage renal disease before dialysis, and the mean fraction of the drug removed by the dialyzer was 62% at 2 hours and 68% at 4 hours.³² This study suggests that hemodialysis may be useful to accelerate the removal of the drug in cases of life-threatening bleeding.

Rivaroxaban and apixaban are not dialyzable: the plasma protein binding of rivaroxaban is 95% and that of apixaban is 87%.

FUTURE DIRECTIONS

Because the new oral anticoagulants, unlike warfarin, have a wide therapeutic window, routine anticoagulant monitoring is not needed and might be misleading. However, there are times when monitoring might be useful; at such times, a validated, widely available, easily understood test would be good to have—but we don’t have it—at least not yet.

Therapeutic ranges for the aPTT have been established empirically for heparin in various indications.⁴⁴ Additional study is needed to determine if an appropriate aPTT range can be determined for the new oral anticoagulants, particularly dabigatran.

Similarly, as with low-molecular-weight heparins, anti-factor Xa activity monitoring may become a more available validated means of testing for exposure to rivaroxaban and apixaban. More promising, using concepts derived from the development of the INR for warfarin monitoring,⁴⁵ Tripodi et al⁴⁶ have derived normalized INR-like assays to report rivaroxaban levels. A standardized schema for reporting results is being developed.⁴⁶ Studies are required to determine if and how this assay may be useful. Initial trials in this regard are encouraging.⁴⁷

Finally, the thrombotic risk associated with the use of nonspecific prohemostatic agents is unknown.^37,48 Additional studies are required to standardize their dosages, frequency of administration, and duration of action, as well as to quantify their complications in bleeding patients.

In the past several years, three new oral anticoagulants—dabigatran etexilate (Pradaxa), rivaroxaban (Xarelto), and apixaban (Eliquis)—have been approved for use in the United States. These long-awaited agents are appealing because they are easy to use, do not require laboratory monitoring, and have demonstrated equivalence, or in some cases, superiority to warfarin in preventing stroke or systemic embolism in at-risk populations.^1–4 However, unlike warfarin, they have no specific reversal agents. How then should one manage spontaneous bleeding problems and those due to drug overdose, and how can we quickly reverse anticoagulation if emergency surgery is needed?

For these reasons, physicians and patients have been wary of these agents. However, with a systematic approach based on an understanding of the properties of these drugs, the appropriate use and interpretation of coagulation tests, and awareness of available therapeutic strategies, physicians can more confidently provide care for patients who require urgent reversal of anticoagulant effects.

Here, we review the available literature and suggest practical strategies for management based on an understanding of the pharmacokinetic and pharmacodynamic effects of these drugs and our current knowledge of the coagulation tests.

NEED FOR ANTICOAGULANTS

Anticoagulants are important in preventing systemic embolization in patients with atrial fibrillation and preventing pulmonary embolism in patients with venous thromboembolism.

And the numbers are staggering. The estimated prevalence of atrial fibrillation in the United States was 3.03 million in 2005 and is projected to increase to 7.56 million by 2050.⁵ Ischemic stroke is the most serious complication of atrial fibrillation, which accounts for 23.5% of strokes in patients ages 80 through 89 according to Framingham data.⁶ Venous thromboembolism accounts for 900,000 incident or recurrent fatal and nonfatal events in the United States yearly.⁷

HOW THE NEW AGENTS BLOCK COAGULATION

Thrombin (factor IIa), a serine protease, is central to the process of clot formation during hemostasis. It activates factors V, VIII, and XI (thus generating more thrombin), catalyzes the conversion of fibrinogen to fibrin, and stimulates platelet aggregation. Its role in the final steps of the coagulation cascade has made it a target for new direct thrombin inhibitors such as dabigatran.

Factor Xa is a serine protease that plays a central role in the coagulation cascade. It is a desirable target for anticoagulation because it is the convergence point for the extrinsic and the intrinsic coagulation pathways. It converts prothrombin to thrombin. Rivaroxaban and apixaban are direct factor Xa inhibitors (Figure 1).

Dabigatran, a direct thrombin inhibitor

Dabigatran etexilate is a synthetic, orally available prodrug that is rapidly absorbed and converted by esterases to its active form, dabigatran, a potent direct inhibitor of both free thrombin and clot-bound thrombin.⁸

Plasma levels of dabigatran peak within 2 hours of administration, and its half-life is 14 to 17 hours.⁹ Dabigatran is eliminated mainly via the kidneys, with more that 80% of the drug excreted unchanged in the urine (Table 1).

Rivaroxaban, a factor Xa inhibitor

Rivaroxaban is a potent, selective, direct factor Xa inhibitor.

Plasma levels of rivaroxaban peak 2 to 3 hours after administration, and it is cleared with a terminal half-life of 7 to 11 hours.^10,11

Rivaroxaban is eliminated by the kidneys and in the feces. The kidneys eliminate one-third of the active drug unchanged and another one-third as inactive metabolites. The remaining one-third is metabolized by the liver and then excreted in the feces. Rivaroxaban has a predictable and dose-dependent pharmacodynamic and pharmacokinetic profile that is not affected by age, sex, or body weight (Table 1).¹²

Apixaban, an oral factor Xa inhibitor

Apixaban is a selective, direct oral factor Xa inhibitor.

Plasma levels of apixaban peak about 3 hours after administration, and its terminal half-life is 8 to 14 hours.¹³ Apixaban is eliminated by oxidative metabolism, by the kidney, and in the feces. It has predictable pharmacodynamic and pharmacokinetic profiles and has the least renal dependence of the three agents (Table 1).

THE NEW ORAL ANTICOAGULANTS AND BLOOD COAGULATION ASSAYS

Assessment of the anticoagulant activity of the new oral anticoagulants is not necessary in routine clinical practice, but it may be useful in planning intervention in patients with major bleeding, those with drug overdose, or those who need emergency surgery.

The activated partial thromboplastin time

The activated partial thromboplastin time (aPTT) is a measure of the activity of the intrinsic pathway of the coagulation cascade.

Dabigatran. There is a curvilinear relationship between the aPTT and the plasma concentration of dabigatran and other direct thrombin inhibitors, although the aPTT prolongation appears to vary with different reagents and coagulometers.^9,14,15 However, Stangier et al⁹ found a linear relationship between the aPTT and the square root of the dabigatran plasma concentration.

Rivaroxaban prolongs the aPTT in a dose-dependent manner, but there is no standard for calibration of this assay. Hence, the aPTT is not recommended for monitoring rivaroxaban in clinical practice.

Apixaban may also prolong the aPTT, but there are limited data on its reactivity with different reagents.

The prothrombin time and international normalized ratio

The prothrombin time and international normalized ratio (INR) are measures of the extrinsic pathway of the coagulation cascade.

Dabigatran. The INR has a linear response to the dabigatran concentration, but it is insensitive.⁹ Hence, it is not suitable for monitoring the anticoagulant effects of direct thrombin inhibitors.

Rivaroxaban. The prothrombin time correlates strongly with the plasma concentration of rivaroxaban in healthy trial participants¹¹ and in patients undergoing total hip arthroplasty or total knee arthroplasty.¹⁶ Samama et al¹⁷ noted that, unlike with vitamin K antagonists, the INR cannot be used to monitor patients on rivaroxaban because the prothrombin time results varied with different reagents. They used a standard calibration curve to express the prothrombin time results in plasma concentrations of rivaroxaban rather than in seconds or the INR.

Apixaban increases the INR in a dose-dependent manner.¹⁸ Its effect on different reagents remains unknown.

The thrombin time

The thrombin time reflects the activity of thrombin in the plasma. The amount of thrombin and the concentration of thrombin inhibitors in the plasma sample determine the time to clot formation.

Dabigatran. The thrombin time displays a linear dose-response to dabigatran, but only over the range of therapeutic concentrations. At a dabigatran concentration greater than 600 ng/mL, the test often exceeds the maximum measurement time of coagulometers.⁹ Hence, this test is too sensitive for emergency monitoring, especially in cases of drug overdose. However, it is well suited for determining if any dabigatran is present.

Rivaroxaban and apixaban have no effect on the thrombin time.

The Hemoclot direct thrombin inhibitor assay and dabigatran

The Hemoclot direct thrombin inhibitor assay (Hyphen BioMed, France) is a sensitive diluted thrombin time assay that can be used for quantitative measurement of dabigatran activity in plasma. This test is based on inhibition of a constant amount of highly purified human alpha-thrombin by adding it to diluted test plasma (1:8 to 1:20) mixed with normal pooled human plasma.^19,20

Stangier et al¹⁹ found that the Hemoclot assay was suitable for calculating a wide range of dabigatran concentrations up to 4,000 nmol/L (1,886 ng/mL). Although this finding has not been confirmed in larger studies, this test may provide a rapid and accurate assessment of dabigatran’s anticoagulant activity in cases of emergency surgery or overdose.

The ecarin clotting time and dabigatran

The ecarin clotting time is a measure of the activity of direct thrombin inhibitors, but not the factor Xa inhibitors.

Ecarin is a highly purified metalloprotease isolated from the venom of a snake, Echis carinatus, and it generates meizothrombin from prothrombin.²¹ Meizothrombin facilitates clot formation by converting fibrinogen to fibrin and, like thrombin, it can be inactivated by direct thrombin inhibitors, thereby prolonging the clotting time.

The limitations of the ecarin clotting time include dependence on the plasma levels of fibrinogen and prothrombin.

The ecarin chromogenic assay and dabigatran

The ecarin chromogenic assay is an improvement on the principle of the ecarin clotting time that can be used to measure the activity of direct thrombin inhibitors.²² In this test, ecarin is added to a plasma sample to generate meizothrombin, and the amidolytic activity of meizothrombin towards a chromogenic substrate is then determined.

Results of the ecarin chromogenic assay are not influenced by the levels of fibrinogen or prothrombin. Another advantage is that this assay can be used in automated and manual analyzers, thus enabling its use at the bedside. However, to our knowledge, it is not being regularly used to monitor direct thrombin inhibitors in the clinical setting, and there is no standard calibration of the ecarin clotting time method.

Assays of factor Xa activity

A variety of assays to monitor the anticoagulant activity of factor Xa inhibitors have been proposed.^23–25 All measure inhibition of the activity of factor Xa using methods similar to those used in monitoring heparin levels. All require calibrators with a known concentration of the Xa inhibitor; many are easily adapted for laboratories currently providing measurement of factor Xa inhibition from heparin.²³ These assays have been suggested as a better indicator of plasma concentration of factor Xa inhibitor drugs than the prothrombin time.²⁵

CONTROLLING BLEEDING IN PATIENTS ON THE NEW ORAL ANTICOAGULANTS

Bleeding is an anticipated adverse event in patients taking anticoagulants. It is associated with significant morbidity and risk of death.^26,27

Many physicians still have limited experience with using the new oral anticoagulants and managing the attendant bleeding risks. Hence, we recommend that every health institution have a treatment policy or algorithm to guide all clinical staff in the management of such emergencies.

Prevention of bleeding

Management of bleeding from these agents should begin with preventing bleeding in the first place.

The physician should adhere to the recommended dosages of these medications. Studies have shown that the plasma concentration of these drugs and the risk of bleeding increase with increasing dosage.^1,28,29

In addition, these medications should be used for the shortest time for which anticoagulation is required, especially when used for preventing deep vein thrombosis. Prolonged use increases the risk of bleeding.^30,31

Most patients who need anticoagulation have comorbidities such as heart failure, renal failure, diabetes mellitus, and hypertension. Although the kidneys play a major role in the excretion of dabigatran and, to some extent, rivaroxaban and apixaban, patients with severe renal impairment were excluded from the major trials of all three drugs.^1–3 Hence, to avoid excessive drug accumulation and bleeding, these medications should not be used in such patients pending further studies. Further, patients taking these medications should be closely followed to detect new clinical situations, such as acute renal failure, that will necessitate their discontinuation or dose adjustment.

If a patient taking a new oral anticoagulant needs to undergo elective surgery, it is important to temporarily discontinue the drug, assess the risk of bleeding, and test for renal impairment.

Renal impairment is particularly relevant in the case of dabigatran, since more than 80% of the unchanged drug is cleared by the kidneys. Decreasing the dose, prolonging the dosing interval, or both have been suggested as means to reduce the risk of bleeding in patients with renal impairment who are taking dabigatran.^32,33 Patients with normal renal function undergoing low-risk surgery should discontinue dabigatran at least 24 hours before the surgery. If the creatinine clearance is 31 to 50 mL/min, inclusively, the last dose should be at least 48 hours before the procedure for low-risk surgery, and 4 days before a procedure that poses a high risk of bleeding.^32–34 Some experts have given the same recommendations for rivaroxaban and apixaban (Table 2).³⁴

The aPTT and prothrombin time are readily available tests, but they cannot determine the residual anticoagulant effects of dabigatran, rivaroxaban, or apixaban. However, in many (but not all) cases, a normal aPTT suggests that the hemostatic function is not impaired by dabigatran, and a normal prothrombin time or an absence of anti-factor Xa activity would similarly exclude hemostatic dysfunction caused by rivaroxaban or apixaban. These tests are potentially useful as adjuncts before surgical procedures that require complete hemostasis.

Furthermore, a normal thrombin time rules out the presence of a significant amount of dabigatran. Therefore, a normal thrombin time might be particularly useful in a patient undergoing a high-risk intervention such as epidural cannulation or neurosurgery and who is normally receiving dabigatran.

Managing overdose and bleeding complications

Assessing the severity of bleeding is the key to managing bleeding complications (Table 3).

Minor bleeding such as epistaxis and ecchymosis can be managed symptomatically (eg, with nasal packing), perhaps with short-term withdrawal of the anticoagulant. Moderate bleeding such as upper or lower gastrointestinal bleeding can be managed by withdrawal of the anticoagulant, clinical monitoring, blood transfusion if needed, and treatment directed at the etiology.

Major and life-threatening bleeding (eg, intracerebral hemorrhage) requires aggressive treatment in the intensive care unit, withdrawal of the anticoagulant, mechanical compression of the bleeding site if accessible, fluid replacement and blood transfusion as appropriate, and interventional procedures. Nonspecific reversal agents might be considered in patients with major or life-threatening bleeding.

The half-life of dabigatran after multiple doses is approximately 14 to 17 hours and is not dose-dependent.⁹ Hence, if there is no active bleeding after a dabigatran overdose, stopping the drug may be sufficient. Since the pharmacodynamic effect of dabigatran declines in parallel to its plasma concentration, urgent but not emergency surgery may need to be delayed for only about 12 hours from the last dose of dabigatran.

The 2011 American College of Cardiology Foundation/American Heart Association guidelines recommend that patients with severe hemorrhage resulting from dabigatran should receive supportive therapy, including transfusion of fresh-frozen plasma, transfusion of packed red blood cells, or surgical intervention if appropriate.³⁵ However, transfusion of fresh-frozen plasma is debatable because there is no evidence to support its use in this situation. While fresh-frozen plasma may be useful in cases of coagulation factor depletion, it does not effectively reverse inhibition of coagulation factors.³⁶

Off-label use of nonspecific hemostatic agents

To date, no specific agent has been demonstrated to reverse excessive bleeding in patients taking the new oral anticoagulants. However, in view of their procoagulant capabilities, nonspecific hemostatic agents have been suggested for use in reversal of major bleeding resulting from these drugs.^37–39 Examples are:

Recombinant factor VIIa (NovoSeven) initiates thrombin generation by activating factor X.

Four-factor prothrombin complex concentrate (Beriplex, recently approved in the United States) contains relatively large amounts of four nonactive vitamin K-dependent procoagulant factors (factors II, VII, IX, and X) that stimulate thrombin formation.

Three-factor prothrombin complex concentrate (Bebulin VH and Profilnine SD) contains low amounts of nonactive factor VII relative to factors II, IX, and X. In some centers a four-factor equivalent is produced by transfusion of a three-factor product with the addition of small amounts of recombinant factor VIIa or fresh-frozen plasma to replace the missing factor VII.⁴⁰

Activated prothrombin complex concentrate (FEIBA NF) contains activated factor VII and factors II, IX, and X, mainly in nonactivated form.³⁶ Therefore, it combines the effect of both recombinant factor VIIa and four-factor prothrombin complex concentrate.³⁷

Studies of nonspecific hemostatic agents

In a study of rats infused with high doses of dabigatran, van Ryn et al³⁸ observed that activated prothrombin complex concentrate at a dose of 50 or 100 U/kg and recombinant factor VIIa at a dose of 0.1 or 0.5 mg/kg reduced the rat-tail bleeding time in a dose-dependent manner but not the blood loss, compared with controls, even with a higher dose of recombinant factor VIIa (1 mg/kg). Recombinant factor VIIa also reversed the prolonged aPTT induced by dabigatran, whereas activated prothrombin complex concentrate did not. They suggested that recombinant factor VIIa and activated prothrombin complex concentrate may be potential antidotes for dabigatran-induced severe bleeding in humans.

In an ex vivo study of healthy people who took a single dose of dabigatran 150 mg or rivaroxaban 20 mg, Marlu et al³⁷ found that activated prothrombin complex concentrate and four-factor prothrombin complex concentrate could be reasonable antidotes to these drugs.

Dabigatran-associated bleeding after cardiac surgery in humans has been successfully managed with hemodialysis and recombinant factor VIIa, although the efficacy of the latter cannot be individually assessed in the study.⁴¹

In a randomized placebo-controlled trial aimed at reversing rivaroxaban and dabigatran in healthy participants, Eerenberg et al³⁹ showed that four-factor prothrombin complex concentrate at a dose of 50 IU/kg reversed prolongation of the prothrombin time and decreased the endogenous thrombin potential in those who received rivaroxaban, but it failed to reverse the aPTT, the endogenous thrombin potential, and thrombin time in those who received dabigatran.

However, Marlu et al reported that four-factor prothrombin complex concentrate at three doses (12.5 U/kg, 25 U/kg, and 50 U/kg)—or better still, activated prothrombin complex concentrate (40–80 U/kg)—could be a useful antidote to dabigatran.³⁷

It is important to note that the healthy participants in the Eerenberg et al study³⁹ took dabigatran 150 mg twice daily and rivaroxaban 20 mg daily for 2.5 days, whereas those in the Marlu et al study³⁷ took the same dose of each medication, but only once.

The three-factor prothrombin complex concentrate products have been shown to be less effective than four-factor ones in reversing supratherapeutic INRs in patients with warfarin overdose, but whether this will be true with the new oral anticoagulants remains unknown. Furthermore, the four-factor concentrates effectively reversed warfarin-induced coagulopathy and bleeding in patients,⁴² but to our knowledge, the same is yet to be demonstrated in bleeding related to the newer agents.

Other measures

Gastric lavage or the administration of activated charcoal (or in some cases both) may reduce drug absorption if done within 2 or 3 hours of drug ingestion (Table 1). Because it is lipophilic, more than 99.9% of dabigatran etexilate was adsorbed by activated charcoal from water prepared to simulate gastric fluid in an in vitro experiment by van Ryn et al.⁴³ This has not been tested in patients, and no similar study has been done for rivaroxaban or apixaban. However, use of charcoal in cases of recent ingestion, particularly with intentional overdose of these agents, seems reasonable.

Hemodialysis may reverse the anticoagulant effects of dabigatran overdose or severe bleeding because only about 35% of dabigatran is bound to plasma proteins (Table 1). In a single-center study, 50 mg of dabigatran etexilate was given orally to six patients with end-stage renal disease before dialysis, and the mean fraction of the drug removed by the dialyzer was 62% at 2 hours and 68% at 4 hours.³² This study suggests that hemodialysis may be useful to accelerate the removal of the drug in cases of life-threatening bleeding.

Rivaroxaban and apixaban are not dialyzable: the plasma protein binding of rivaroxaban is 95% and that of apixaban is 87%.

FUTURE DIRECTIONS

Because the new oral anticoagulants, unlike warfarin, have a wide therapeutic window, routine anticoagulant monitoring is not needed and might be misleading. However, there are times when monitoring might be useful; at such times, a validated, widely available, easily understood test would be good to have—but we don’t have it—at least not yet.

Therapeutic ranges for the aPTT have been established empirically for heparin in various indications.⁴⁴ Additional study is needed to determine if an appropriate aPTT range can be determined for the new oral anticoagulants, particularly dabigatran.

Similarly, as with low-molecular-weight heparins, anti-factor Xa activity monitoring may become a more available validated means of testing for exposure to rivaroxaban and apixaban. More promising, using concepts derived from the development of the INR for warfarin monitoring,⁴⁵ Tripodi et al⁴⁶ have derived normalized INR-like assays to report rivaroxaban levels. A standardized schema for reporting results is being developed.⁴⁶ Studies are required to determine if and how this assay may be useful. Initial trials in this regard are encouraging.⁴⁷

Finally, the thrombotic risk associated with the use of nonspecific prohemostatic agents is unknown.^37,48 Additional studies are required to standardize their dosages, frequency of administration, and duration of action, as well as to quantify their complications in bleeding patients.

READ THE REST OF THE SERIES
Pillar 1: Keep your current patients happy (March 2013)
Dr. Baum describes his number one strategy to retain patients (Audiocast, March 2013)
Pillar 2: Attract new patients (May 2013)
Pillar 4: Motivate your staff (August 2013)

Discussions of medical marketing often begin with the three As: availability, affability, and affordability. But most physicians already think of themselves as available, likeable, and offering appropriately priced services.

How do you differentiate yourself from the competition?

Fancy stationery; a slick, three-color brochure; a catchy logo; and a Web site will not do the trick. In fact, these are the last things you need.

One of the biggest misconceptions about marketing is that, to do it well, you must spend lots of money on peripherals. In truth, there are many other actions that are far more effective and essential to marketing than merely polishing your public relations image. The most essential element of your marketing plan is to make your practice user-friendly.

Nowhere is this need greater than when it comes to working with colleagues who are capable of referring patients to you—or are already doing so. In this article, I describe 10 strategies you can use to enhance your relationships with referring physicians.

1. WRITE AN EFFECTIVE REFERRAL LETTER

To obtain referrals from your colleagues, you need to ensure that your name crosses their mind and desk as frequently as possible—and in a positive fashion.

If you interview referring physicians, you will find that prompt communication is one of the most important reasons they refer a patient to a particular provider. According to the Annals of Family Medicine, more than 50% of physicians state that effective communication is the reason they select a doctor for referral (TABLE).¹

Source: Kinchen et al¹

Keep your referral letter short

The traditional referral letter is far too long, often 2 or 3 pages. It usually arrives 10 to 14 days after the patient was seen and is very expensive, costing a practice $12–$15 for each letter sent. The goal of an effective referral letter: Get it there before the patient returns to the primary care provider.

The key ingredients of an effective referral letter are:

• diagnosis

• medications you have prescribed for the patient

• your treatment plan.

The referring doctor is not interested in the nuances of your history or physical exam. They just want the three ingredients listed above.

For example, let’s say that Dr. Bill Smith refers Jane Doe, who has an overactive bladder and cystocele. Her urinalysis is negative, so you prescribe an anticholinergic agent and schedule a follow-up visit in 1 month to check symptoms and to conduct a urodynamic study if she has not improved. Your letter to Dr. Smith would read as follows:

Now the letter can be faxed to the referring doctor, often before the patient leaves the office. That way you can be certain that the letter arrives before the patient calls the physician with questions or concerns.

This is the best way to keep the referring physician informed and to function as the captain of the patient’s health-care ship.

EHRs can smooth the referral process

Most electronic health records (EHRs) have the capability to fax the entire note to the referring physician. However, if you were to ask a referring physician if she would like to read your entire note, the answer would probably be “No.” Most EHRs will allow you to select fields that contain the diagnosis, medications prescribed, and the treatment plan. A sample of this kind of letter appears in the FIGURE.

2. MAKE AN EFFORT TO PERSONALLY MEET EVERY PHYSICIAN WHO REFERS A PATIENT

Not only that, but try to meet all new physicians in your area. It is important to coddle your existing sources of referrals, but don’t forget to reach out to new physicians to let them know about your areas of interest or expertise.

3. REFER YOUR NEW PATIENTS TO REFERRING PHYSICIANS

Don’t refer to the same colleagues time after time. If a doctor starts sending new patients your way, it’s in your best interest to “reverse-refer” when a patient needs a primary care doctor, endocrinologist, or cardiologist.

You can be sure these referring doctors will appreciate your recommendations.

Related Article  Complex atypical endometrial hyperplasia: When to refer

4. CREATE A LUNCH-AND-LEARN PROGRAM

You want other offices and medical staffs to get to know your staff and to be familiar with what you do. There’s no better way than to create a lunch-and-learn program in your office and extend an invitation to other offices in the area. At the program, have all of the staff members introduce themselves. Provide a tour of your office and give a 3- to 5-minute lecture on areas of your gynecologic interest and expertise.

5. ACKNOWLEDGE THE ACCOMPLISHMENTS OF REFERRING PHYSICIANS AND THEIR FAMILIES

If you see that one of your referring physicians has received an honor or award, send him a congratulatory note. If her children have been recognized for academic or athletic achievement, acknowledge this accomplishment with a note. You can be sure it will be one of the only acknowledgments they receive and will be deeply appreciated.

6. SHARE INFORMATION WITH A NO-MEETING JOURNAL CLUB

It’s very difficult to keep up with the medical literature. It’s challenging enough to keep up with the literature in your own specialty, let alone articles appearing in other specialty publications. One of the nicest gestures you can make is to copy any article that may be of interest to your colleagues and send it to them. Include a sticky note indicating where you would like them to look so that they don’t have to read the entire article.

7. SHARE NONMEDICAL INFORMATION, TOO

Your colleagues will appreciate it when you share nonmedical information to let them know you are thinking of them even when you are not discussing patient care. For example, one of my colleagues collects fine pens. When I saw an article about a very expensive pen made with diamonds, I sent the story to my friend, suggesting that he tell his wife what was on his wish list.

8. KEEP THE REFERRING DOCTOR IN THE MEDICAL LOOP

If you are caring for a patient and plan to discharge her from the hospital, make sure that you or someone in your office contacts the referring doctor to inform him that the patient is being discharged so he doesn’t make unnecessary rounds. Other times to notify the referring doctor:

• upon admission of her patient to the hospital

• after surgery or a procedure

• when you receive a significant laboratory or pathology report.

9. BE USER-FRIENDLY

If you perform gynecologic surgery on a referred patient, be sure to dictate a discharge summary. If the patient is to be discharged with gynecologic medications, give the patient their names in writing. Another convenience for the patient: Arrange your follow-up appointment on the same day she is to return to see the referring physician.

10. DON’T FORGET NONPHYSICIAN REFERRAL SOURCES

Nurses, pharmacists, pharmaceutical representatives, social workers, lawyers, beauticians, and manicurists—all of these professionals are likely to refer patients to you if you keep them in the loop.

11. BOTTOM LINE

You can build a practice by word of mouth by doing a great job of caring for patients, hoping that they will tell others about their positive experience. However, there are other opportunities to enhance your practice—notably, by nurturing your relationship with referring physicians. Try a few of these ideas and you will certainly see your referrals increase significantly.

We want to hear from you! Tell us what you think.

READ THE REST OF THE SERIES
Pillar 1: Keep your current patients happy (March 2013)
Dr. Baum describes his number one strategy to retain patients (Audiocast, March 2013)
Pillar 2: Attract new patients (May 2013)
Pillar 4: Motivate your staff (August 2013)

Discussions of medical marketing often begin with the three As: availability, affability, and affordability. But most physicians already think of themselves as available, likeable, and offering appropriately priced services.

How do you differentiate yourself from the competition?

Fancy stationery; a slick, three-color brochure; a catchy logo; and a Web site will not do the trick. In fact, these are the last things you need.

One of the biggest misconceptions about marketing is that, to do it well, you must spend lots of money on peripherals. In truth, there are many other actions that are far more effective and essential to marketing than merely polishing your public relations image. The most essential element of your marketing plan is to make your practice user-friendly.

Nowhere is this need greater than when it comes to working with colleagues who are capable of referring patients to you—or are already doing so. In this article, I describe 10 strategies you can use to enhance your relationships with referring physicians.

1. WRITE AN EFFECTIVE REFERRAL LETTER

To obtain referrals from your colleagues, you need to ensure that your name crosses their mind and desk as frequently as possible—and in a positive fashion.

If you interview referring physicians, you will find that prompt communication is one of the most important reasons they refer a patient to a particular provider. According to the Annals of Family Medicine, more than 50% of physicians state that effective communication is the reason they select a doctor for referral (TABLE).¹

Source: Kinchen et al¹

Keep your referral letter short

The traditional referral letter is far too long, often 2 or 3 pages. It usually arrives 10 to 14 days after the patient was seen and is very expensive, costing a practice $12–$15 for each letter sent. The goal of an effective referral letter: Get it there before the patient returns to the primary care provider.

The key ingredients of an effective referral letter are:

• diagnosis

• medications you have prescribed for the patient

• your treatment plan.

The referring doctor is not interested in the nuances of your history or physical exam. They just want the three ingredients listed above.

For example, let’s say that Dr. Bill Smith refers Jane Doe, who has an overactive bladder and cystocele. Her urinalysis is negative, so you prescribe an anticholinergic agent and schedule a follow-up visit in 1 month to check symptoms and to conduct a urodynamic study if she has not improved. Your letter to Dr. Smith would read as follows:

Now the letter can be faxed to the referring doctor, often before the patient leaves the office. That way you can be certain that the letter arrives before the patient calls the physician with questions or concerns.

This is the best way to keep the referring physician informed and to function as the captain of the patient’s health-care ship.

EHRs can smooth the referral process

Most electronic health records (EHRs) have the capability to fax the entire note to the referring physician. However, if you were to ask a referring physician if she would like to read your entire note, the answer would probably be “No.” Most EHRs will allow you to select fields that contain the diagnosis, medications prescribed, and the treatment plan. A sample of this kind of letter appears in the FIGURE.

2. MAKE AN EFFORT TO PERSONALLY MEET EVERY PHYSICIAN WHO REFERS A PATIENT

Not only that, but try to meet all new physicians in your area. It is important to coddle your existing sources of referrals, but don’t forget to reach out to new physicians to let them know about your areas of interest or expertise.

3. REFER YOUR NEW PATIENTS TO REFERRING PHYSICIANS

Don’t refer to the same colleagues time after time. If a doctor starts sending new patients your way, it’s in your best interest to “reverse-refer” when a patient needs a primary care doctor, endocrinologist, or cardiologist.

You can be sure these referring doctors will appreciate your recommendations.

Related Article  Complex atypical endometrial hyperplasia: When to refer

4. CREATE A LUNCH-AND-LEARN PROGRAM

You want other offices and medical staffs to get to know your staff and to be familiar with what you do. There’s no better way than to create a lunch-and-learn program in your office and extend an invitation to other offices in the area. At the program, have all of the staff members introduce themselves. Provide a tour of your office and give a 3- to 5-minute lecture on areas of your gynecologic interest and expertise.

5. ACKNOWLEDGE THE ACCOMPLISHMENTS OF REFERRING PHYSICIANS AND THEIR FAMILIES

If you see that one of your referring physicians has received an honor or award, send him a congratulatory note. If her children have been recognized for academic or athletic achievement, acknowledge this accomplishment with a note. You can be sure it will be one of the only acknowledgments they receive and will be deeply appreciated.

6. SHARE INFORMATION WITH A NO-MEETING JOURNAL CLUB

It’s very difficult to keep up with the medical literature. It’s challenging enough to keep up with the literature in your own specialty, let alone articles appearing in other specialty publications. One of the nicest gestures you can make is to copy any article that may be of interest to your colleagues and send it to them. Include a sticky note indicating where you would like them to look so that they don’t have to read the entire article.

7. SHARE NONMEDICAL INFORMATION, TOO

Your colleagues will appreciate it when you share nonmedical information to let them know you are thinking of them even when you are not discussing patient care. For example, one of my colleagues collects fine pens. When I saw an article about a very expensive pen made with diamonds, I sent the story to my friend, suggesting that he tell his wife what was on his wish list.

8. KEEP THE REFERRING DOCTOR IN THE MEDICAL LOOP

If you are caring for a patient and plan to discharge her from the hospital, make sure that you or someone in your office contacts the referring doctor to inform him that the patient is being discharged so he doesn’t make unnecessary rounds. Other times to notify the referring doctor:

• upon admission of her patient to the hospital

• after surgery or a procedure

• when you receive a significant laboratory or pathology report.

9. BE USER-FRIENDLY

If you perform gynecologic surgery on a referred patient, be sure to dictate a discharge summary. If the patient is to be discharged with gynecologic medications, give the patient their names in writing. Another convenience for the patient: Arrange your follow-up appointment on the same day she is to return to see the referring physician.

10. DON’T FORGET NONPHYSICIAN REFERRAL SOURCES

Nurses, pharmacists, pharmaceutical representatives, social workers, lawyers, beauticians, and manicurists—all of these professionals are likely to refer patients to you if you keep them in the loop.

11. BOTTOM LINE

You can build a practice by word of mouth by doing a great job of caring for patients, hoping that they will tell others about their positive experience. However, there are other opportunities to enhance your practice—notably, by nurturing your relationship with referring physicians. Try a few of these ideas and you will certainly see your referrals increase significantly.

We want to hear from you! Tell us what you think.

READ THE REST OF THE SERIES
Pillar 1: Keep your current patients happy (March 2013)
Dr. Baum describes his number one strategy to retain patients (Audiocast, March 2013)
Pillar 2: Attract new patients (May 2013)
Pillar 4: Motivate your staff (August 2013)

Discussions of medical marketing often begin with the three As: availability, affability, and affordability. But most physicians already think of themselves as available, likeable, and offering appropriately priced services.

How do you differentiate yourself from the competition?

Fancy stationery; a slick, three-color brochure; a catchy logo; and a Web site will not do the trick. In fact, these are the last things you need.

One of the biggest misconceptions about marketing is that, to do it well, you must spend lots of money on peripherals. In truth, there are many other actions that are far more effective and essential to marketing than merely polishing your public relations image. The most essential element of your marketing plan is to make your practice user-friendly.

Nowhere is this need greater than when it comes to working with colleagues who are capable of referring patients to you—or are already doing so. In this article, I describe 10 strategies you can use to enhance your relationships with referring physicians.

1. WRITE AN EFFECTIVE REFERRAL LETTER

To obtain referrals from your colleagues, you need to ensure that your name crosses their mind and desk as frequently as possible—and in a positive fashion.

If you interview referring physicians, you will find that prompt communication is one of the most important reasons they refer a patient to a particular provider. According to the Annals of Family Medicine, more than 50% of physicians state that effective communication is the reason they select a doctor for referral (TABLE).¹

Source: Kinchen et al¹

Keep your referral letter short

The traditional referral letter is far too long, often 2 or 3 pages. It usually arrives 10 to 14 days after the patient was seen and is very expensive, costing a practice $12–$15 for each letter sent. The goal of an effective referral letter: Get it there before the patient returns to the primary care provider.

The key ingredients of an effective referral letter are:

• diagnosis

• medications you have prescribed for the patient

• your treatment plan.

The referring doctor is not interested in the nuances of your history or physical exam. They just want the three ingredients listed above.

For example, let’s say that Dr. Bill Smith refers Jane Doe, who has an overactive bladder and cystocele. Her urinalysis is negative, so you prescribe an anticholinergic agent and schedule a follow-up visit in 1 month to check symptoms and to conduct a urodynamic study if she has not improved. Your letter to Dr. Smith would read as follows:

Now the letter can be faxed to the referring doctor, often before the patient leaves the office. That way you can be certain that the letter arrives before the patient calls the physician with questions or concerns.

This is the best way to keep the referring physician informed and to function as the captain of the patient’s health-care ship.

EHRs can smooth the referral process

Most electronic health records (EHRs) have the capability to fax the entire note to the referring physician. However, if you were to ask a referring physician if she would like to read your entire note, the answer would probably be “No.” Most EHRs will allow you to select fields that contain the diagnosis, medications prescribed, and the treatment plan. A sample of this kind of letter appears in the FIGURE.

2. MAKE AN EFFORT TO PERSONALLY MEET EVERY PHYSICIAN WHO REFERS A PATIENT

Not only that, but try to meet all new physicians in your area. It is important to coddle your existing sources of referrals, but don’t forget to reach out to new physicians to let them know about your areas of interest or expertise.

3. REFER YOUR NEW PATIENTS TO REFERRING PHYSICIANS

Don’t refer to the same colleagues time after time. If a doctor starts sending new patients your way, it’s in your best interest to “reverse-refer” when a patient needs a primary care doctor, endocrinologist, or cardiologist.

You can be sure these referring doctors will appreciate your recommendations.

Related Article  Complex atypical endometrial hyperplasia: When to refer

4. CREATE A LUNCH-AND-LEARN PROGRAM

You want other offices and medical staffs to get to know your staff and to be familiar with what you do. There’s no better way than to create a lunch-and-learn program in your office and extend an invitation to other offices in the area. At the program, have all of the staff members introduce themselves. Provide a tour of your office and give a 3- to 5-minute lecture on areas of your gynecologic interest and expertise.

5. ACKNOWLEDGE THE ACCOMPLISHMENTS OF REFERRING PHYSICIANS AND THEIR FAMILIES

If you see that one of your referring physicians has received an honor or award, send him a congratulatory note. If her children have been recognized for academic or athletic achievement, acknowledge this accomplishment with a note. You can be sure it will be one of the only acknowledgments they receive and will be deeply appreciated.

6. SHARE INFORMATION WITH A NO-MEETING JOURNAL CLUB

It’s very difficult to keep up with the medical literature. It’s challenging enough to keep up with the literature in your own specialty, let alone articles appearing in other specialty publications. One of the nicest gestures you can make is to copy any article that may be of interest to your colleagues and send it to them. Include a sticky note indicating where you would like them to look so that they don’t have to read the entire article.

7. SHARE NONMEDICAL INFORMATION, TOO

Your colleagues will appreciate it when you share nonmedical information to let them know you are thinking of them even when you are not discussing patient care. For example, one of my colleagues collects fine pens. When I saw an article about a very expensive pen made with diamonds, I sent the story to my friend, suggesting that he tell his wife what was on his wish list.

8. KEEP THE REFERRING DOCTOR IN THE MEDICAL LOOP

If you are caring for a patient and plan to discharge her from the hospital, make sure that you or someone in your office contacts the referring doctor to inform him that the patient is being discharged so he doesn’t make unnecessary rounds. Other times to notify the referring doctor:

• upon admission of her patient to the hospital

• after surgery or a procedure

• when you receive a significant laboratory or pathology report.

9. BE USER-FRIENDLY