Accepted Manuscript Update in Outpatient General Internal Medicine: Practice Changing Evidence Published in 2014 Karna K. Sundsted, M.D, Mark L. Wieland, M.D. MPH, Jason H. Szostek, M.D, Jason A. Post, M.D, Karen F. Mauck, M.D. MSc PII:
S0002-9343(15)00444-1
DOI:
10.1016/j.amjmed.2015.04.033
Reference:
AJM 12997
To appear in:
The American Journal of Medicine
Received Date: 1 April 2015 Revised Date:
13 April 2015
Accepted Date: 13 April 2015
Please cite this article as: Sundsted KK, Wieland ML, Szostek JH, Post JA, Mauck KF, Update in Outpatient General Internal Medicine: Practice Changing Evidence Published in 2014, The American Journal of Medicine (2015), doi: 10.1016/j.amjmed.2015.04.033. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
ACCEPTED MANUSCRIPT Update in Outpatient General Internal Medicine: Practice Changing Evidence Published in 2014
RI PT
Karna K. Sundsted, M.D.1 Mark L. Wieland, M.D. MPH2 Jason H. Szostek, M.D.1 Jason A. Post, M.D.2
M AN U
1
SC
Karen F. Mauck, M.D. MSc.1
From the Division of General Internal Medicine and the 2 Division of Primary Care Internal
Medicine, Department of Medicine, Mayo Clinic and Mayo Clinic College of Medicine, Rochester, MN
TE D
Running head: General Internal Medicine Update
Key words: outpatient, literature update, general internal medicine, cancer screening, hepatitis C, Alzheimer disease, ACE-inhibitor, alcohol use disorder, statin, novel
Funding source: None
EP
anticoagulant, lung cancer, stool DNA test
All authors had access to the data and a role in manuscript writing.
AC C
Conflict of interest: Multitarget Stool DNA Testing for Colorectal Cancer Screening was developed in part at Mayo Clinic Rochester, our affiliated institution. Karen F. Mauck owns shares in Exact Science as part of a diverse portfolio of stocks. The remaining authors have no individual financial relationships to disclose.
Word Count: 2910
ACCEPTED MANUSCRIPT Abstract The practice of outpatient general internal medicine requires a diverse and evolving knowledge base. General internists must identify practice changing shifts in the literature and reflect on their impact. Accordingly, we conducted a review of practice changing articles published in outpatient general internal medicine in 2014. To identify high quality,
RI PT
clinically relevant publications, we reviewed all titles and abstracts published in the
following primary data sources in 2014: New England Journal of Medicine, JAMA, Annals of Internal Medicine, JAMA-Internal Medicine and the Cochrane Database of Systematic
Reviews. All 2014 primary data summaries from Journal Watch-General Internal Medicine
SC
and ACP JournalWise were also reviewed. The authors used a modified Delphi method to
reach consensus on inclusion of 8 articles using the following criteria: clinical relevance to
M AN U
outpatient internal medicine, potential for practice change, strength of evidence. Clustering of important articles around one clinical question were considered as a single candidate series. The article merits were debated until consensus was reached on the final 8, spanning
AC C
EP
TE D
a variety of topics commonly encountered in outpatient general internal medicine.
ACCEPTED MANUSCRIPT ACE-Inhibitor use in diabetic patients is associated with reduction in all-cause mortality, CV mortality and major CV events while use of angiotensin receptor blockers are not.(1) Background: The American Diabetes Association recommends that patients with diabetes mellitus and hypertension be treated with either an angiotensin converting enzyme inhibitor
RI PT
(ACE-I) or an angiotensin receptor blocker (ARB) as first line therapy(2). The authors
present meta-analytic comparisons (network meta-analysis) of randomized controlled trials which compare ACE-Is and ARBs separately versus placebo or other anti-hypertensive medications on all-cause mortality, cardiovascular deaths, and cardiovascular events in
SC
patients with DM.
Results: Twenty three randomized controlled trials including 32,827 patients compared
M AN U
ACE-Is with controls (placebo or another antihypertensive drug). ACE-Is were associated with a 13% reduction in all-cause mortality (Relative Risk [RR] 0.87; 95% Confidence Interval [CI] 0.78-0.98), a 17% reduction in CV deaths (RR 0.83; 95% CI 0.70-0.99), a 14% reduction in major cardiovascular events (RR 0.86; 95% CI 0.77-0.95), including a 21% reduction in myocardial infarction (MI) and a 19% lower risk of heart failure, but no apparent risk reduction for stroke.
TE D
Thirteen randomized controlled trials including 23,867 patients compared ARBs with controls (placebo or another antihypertensive drug). ARBs were not associated with a significant decrease in all-cause mortality (RR 0.94; 95% CI 0.82-1.08), cardiovascular deaths (RR 1.21; 95% CI 0.81-1.80) or a decrease in major cardiovascular events (RR 0.94;
EP
95% CI 0.85-1.01) but were associated with a 30% reduction in heart failure. Limitations: Only indirect comparisons of ACE-I and ARBs on these outcomes could be
AC C
made due to lack of head-to-head comparisons. There was much heterogeneity between the trials with respect to ACE-I or ARB used, target blood pressures, dose, follow-up time and other background therapies. Because of important variations between the trials, this metaanalysis cannot confirm the superiority of ACE-Is compared to ARBs with respect to study outcomes.
Implications for Practice: This evidence suggests the use of ACE-Is for antihypertensive therapy in diabetic patients is associated with reductions in all-cause mortality, cardiovascular mortality and major cardiovascular events; use of ARBs does not appear to be associated with reductions in these outcomes. Given the available evidence, ACE-Is should be preferred over ARBs as first line therapy in hypertensive, diabetic patients.
ACCEPTED MANUSCRIPT Stool DNA testing may be a reasonable alternative to colonoscopy in patients at average risk for colon cancer.(3) Background: A non-invasive test with high sensitivity for colon cancer and advanced precancerous lesions would likely enhance colon cancer screening adherence and may increase cancer detection rates.
RI PT
Results: This multisite, cross-sectional study enrolled 12,776 average-risk patients between the ages of 50 and 84 years who were scheduled to undergo screening colonoscopy. Patients submitted a stool sample for multi-target stool DNA testing and fecal immunochemical test
(FIT) for human hemoglobin. Colonoscopies were then performed in standard fashion and all
SC
biopsy specimens were reviewed locally and confirmed centrally if abnormal. Outcomes
included the sensitivity and specificity of the DNA test to detect colon cancer (primary) or
M AN U
advanced precancerous lesions (secondary). Of the 9989 participants evaluated, 0.7% (n=65) had cancer, 7.6% (n=757) had advanced precancerous lesions, 28.9% (n=2893) had nonadvanced adenomas on colonoscopy and 62.8% (n=6274) had a negative colonoscopy. Using colonoscopy as the reference standard, stool DNA testing had 92.3% sensitivity for detecting colon cancer (versus 73.8% with FIT) and a 69.2% sensitivity for detecting polyps with highgrade dysplasia (versus 23.8% with FIT). Stool DNA testing was less specific than FIT
TE D
testing, 89.8% versus 96.4%, respectively.
Limitations: A large number of enrolled patients were not included in the analysis. Most patients in the stool DNA group were excluded due to assay application failure or sample collection issues suggesting that test complexity may be a potential limitation. Nearly 10% of
EP
the patients with a positive stool DNA test had a negative colonoscopy. This may lead to additional workup for extracolonic cancers or prompt a shorter interval for repeat
AC C
colonoscopy, both of which are associated with significant cost and unknown benefit. Implications for Practice: Stool DNA testing appears to be a reasonable option for average risk patients who need colon cancer screening. There are several issues that still need further clarification to determine the role of stool DNA testing in programatic screening recommendations including cost effectiveness, mortality data, comparative effectiveness, recommended screening interval, and diagnostic evaluation of positive results.
Side effects from statin drugs are not as common as we think.(4) Background: Multiple studies have demonstrated the benefits of statin drugs for both primary and secondary prevention of cardiovascular disease and cerebrovascular disease. However, many side effects
ACCEPTED MANUSCRIPT have been attributed to statin therapy, increasing risk of patient nonadherence. Information on the causal association between statin use and side effects would aid in clinical decision making and may promote patient compliance. Results: This systematic review included 29 randomized controlled trials and 83,880 participants taking statins for primary and secondary prevention of cardiovascular events. While statins were associated
RI PT
with a small increase in the risk of diabetes in the primary prevention trials by 0.5% (95% CI 0.1-1%; p=0.012), no such increase was seen in the single secondary prevention trial that reported on diabetes. In both the primary and secondary prevention trials, statin over placebo increased the rate of an asymptomatic rise in liver transaminases (>3 upper limit normal) by 0.4%. No other symptom, such as
rhabdomyolysis or fatigue was increased in the statin group.
SC
gastrointestinal disturbance, myalgias, renal disorders, CK elevation, newly diagnosed cancer,
M AN U
Limitations: The primary limitation was inconsistent reporting of adverse events. For example, 24 of the 30 trials reported data for liver enzyme abnormalities, but only three of the trials reported on newly diagnosed diabetes mellitus. Furthermore, many of the trials tested low-strength statin regimens; moderate and high intensity regimens may be associated with more side effects. Implications for Practice: This review provides some reassurance that side effects from statin drugs are not as common as clinical practice might lead us to believe. Often patients report symptoms that are
TE D
attributed to statin side effect which probably are not. Given the established benefit of statins, we should encourage patients to continue statin therapy unless there is convincing evidence that their symptoms are a true side effect.
patients.(5, 6)
EP
The U.S. Preventive Services Task Force now recommends lung cancer screening in high risk
AC C
Background: The U.S. Preventative Services Task Force (USPSTF) recently issued new lung cancer screening guidelines, which include low-dose computed tomography (LDCT) in patients 55-80 years who have a cumulative 30 pack-year history and currently smoke or have quit within the last fifteen years. After guideline release, the Centers for Medicare and Medicaid Services convened a Medicare Evidence Development and Coverage Advisory Committee (MEDCAC) to address age-related screening effectiveness and potential harms, as only 25% of patients in the National Lung Screening Trial (NLST), upon which the USPSTF guideline was largely based, were aged 65 years or older. Results: The NLST enrolled 53,454 patients (aged 55-74 years) and compared LDCT to single view chest radiograph, finding a relative reduction in mortality from lung cancer with low-dose CT screening of 20.0% (95% CI 6.8 to 26.7; P=0.004). The USPSTF then used modeling studies to predict benefits
ACCEPTED MANUSCRIPT and harms and found insufficient evidence for harm related to incidental findings and false positive LDCT scans, noting that screening may not be appropriate in those at the upper end of the age range, particularly when significant comorbid conditions exist. In the analysis of Medicare age patients versus those under age 65, the positive predictive value (PPV) was higher in the 65+ age group (4.9% vs 3.0%, p=
S0002-9343(15)00444-1
DOI:
10.1016/j.amjmed.2015.04.033
Reference:
AJM 12997
To appear in:
The American Journal of Medicine
Received Date: 1 April 2015 Revised Date:
13 April 2015
Accepted Date: 13 April 2015
Please cite this article as: Sundsted KK, Wieland ML, Szostek JH, Post JA, Mauck KF, Update in Outpatient General Internal Medicine: Practice Changing Evidence Published in 2014, The American Journal of Medicine (2015), doi: 10.1016/j.amjmed.2015.04.033. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
ACCEPTED MANUSCRIPT Update in Outpatient General Internal Medicine: Practice Changing Evidence Published in 2014
RI PT
Karna K. Sundsted, M.D.1 Mark L. Wieland, M.D. MPH2 Jason H. Szostek, M.D.1 Jason A. Post, M.D.2
M AN U
1
SC
Karen F. Mauck, M.D. MSc.1
From the Division of General Internal Medicine and the 2 Division of Primary Care Internal
Medicine, Department of Medicine, Mayo Clinic and Mayo Clinic College of Medicine, Rochester, MN
TE D
Running head: General Internal Medicine Update
Key words: outpatient, literature update, general internal medicine, cancer screening, hepatitis C, Alzheimer disease, ACE-inhibitor, alcohol use disorder, statin, novel
Funding source: None
EP
anticoagulant, lung cancer, stool DNA test
All authors had access to the data and a role in manuscript writing.
AC C
Conflict of interest: Multitarget Stool DNA Testing for Colorectal Cancer Screening was developed in part at Mayo Clinic Rochester, our affiliated institution. Karen F. Mauck owns shares in Exact Science as part of a diverse portfolio of stocks. The remaining authors have no individual financial relationships to disclose.
Word Count: 2910
ACCEPTED MANUSCRIPT Abstract The practice of outpatient general internal medicine requires a diverse and evolving knowledge base. General internists must identify practice changing shifts in the literature and reflect on their impact. Accordingly, we conducted a review of practice changing articles published in outpatient general internal medicine in 2014. To identify high quality,
RI PT
clinically relevant publications, we reviewed all titles and abstracts published in the
following primary data sources in 2014: New England Journal of Medicine, JAMA, Annals of Internal Medicine, JAMA-Internal Medicine and the Cochrane Database of Systematic
Reviews. All 2014 primary data summaries from Journal Watch-General Internal Medicine
SC
and ACP JournalWise were also reviewed. The authors used a modified Delphi method to
reach consensus on inclusion of 8 articles using the following criteria: clinical relevance to
M AN U
outpatient internal medicine, potential for practice change, strength of evidence. Clustering of important articles around one clinical question were considered as a single candidate series. The article merits were debated until consensus was reached on the final 8, spanning
AC C
EP
TE D
a variety of topics commonly encountered in outpatient general internal medicine.
ACCEPTED MANUSCRIPT ACE-Inhibitor use in diabetic patients is associated with reduction in all-cause mortality, CV mortality and major CV events while use of angiotensin receptor blockers are not.(1) Background: The American Diabetes Association recommends that patients with diabetes mellitus and hypertension be treated with either an angiotensin converting enzyme inhibitor
RI PT
(ACE-I) or an angiotensin receptor blocker (ARB) as first line therapy(2). The authors
present meta-analytic comparisons (network meta-analysis) of randomized controlled trials which compare ACE-Is and ARBs separately versus placebo or other anti-hypertensive medications on all-cause mortality, cardiovascular deaths, and cardiovascular events in
SC
patients with DM.
Results: Twenty three randomized controlled trials including 32,827 patients compared
M AN U
ACE-Is with controls (placebo or another antihypertensive drug). ACE-Is were associated with a 13% reduction in all-cause mortality (Relative Risk [RR] 0.87; 95% Confidence Interval [CI] 0.78-0.98), a 17% reduction in CV deaths (RR 0.83; 95% CI 0.70-0.99), a 14% reduction in major cardiovascular events (RR 0.86; 95% CI 0.77-0.95), including a 21% reduction in myocardial infarction (MI) and a 19% lower risk of heart failure, but no apparent risk reduction for stroke.
TE D
Thirteen randomized controlled trials including 23,867 patients compared ARBs with controls (placebo or another antihypertensive drug). ARBs were not associated with a significant decrease in all-cause mortality (RR 0.94; 95% CI 0.82-1.08), cardiovascular deaths (RR 1.21; 95% CI 0.81-1.80) or a decrease in major cardiovascular events (RR 0.94;
EP
95% CI 0.85-1.01) but were associated with a 30% reduction in heart failure. Limitations: Only indirect comparisons of ACE-I and ARBs on these outcomes could be
AC C
made due to lack of head-to-head comparisons. There was much heterogeneity between the trials with respect to ACE-I or ARB used, target blood pressures, dose, follow-up time and other background therapies. Because of important variations between the trials, this metaanalysis cannot confirm the superiority of ACE-Is compared to ARBs with respect to study outcomes.
Implications for Practice: This evidence suggests the use of ACE-Is for antihypertensive therapy in diabetic patients is associated with reductions in all-cause mortality, cardiovascular mortality and major cardiovascular events; use of ARBs does not appear to be associated with reductions in these outcomes. Given the available evidence, ACE-Is should be preferred over ARBs as first line therapy in hypertensive, diabetic patients.
ACCEPTED MANUSCRIPT Stool DNA testing may be a reasonable alternative to colonoscopy in patients at average risk for colon cancer.(3) Background: A non-invasive test with high sensitivity for colon cancer and advanced precancerous lesions would likely enhance colon cancer screening adherence and may increase cancer detection rates.
RI PT
Results: This multisite, cross-sectional study enrolled 12,776 average-risk patients between the ages of 50 and 84 years who were scheduled to undergo screening colonoscopy. Patients submitted a stool sample for multi-target stool DNA testing and fecal immunochemical test
(FIT) for human hemoglobin. Colonoscopies were then performed in standard fashion and all
SC
biopsy specimens were reviewed locally and confirmed centrally if abnormal. Outcomes
included the sensitivity and specificity of the DNA test to detect colon cancer (primary) or
M AN U
advanced precancerous lesions (secondary). Of the 9989 participants evaluated, 0.7% (n=65) had cancer, 7.6% (n=757) had advanced precancerous lesions, 28.9% (n=2893) had nonadvanced adenomas on colonoscopy and 62.8% (n=6274) had a negative colonoscopy. Using colonoscopy as the reference standard, stool DNA testing had 92.3% sensitivity for detecting colon cancer (versus 73.8% with FIT) and a 69.2% sensitivity for detecting polyps with highgrade dysplasia (versus 23.8% with FIT). Stool DNA testing was less specific than FIT
TE D
testing, 89.8% versus 96.4%, respectively.
Limitations: A large number of enrolled patients were not included in the analysis. Most patients in the stool DNA group were excluded due to assay application failure or sample collection issues suggesting that test complexity may be a potential limitation. Nearly 10% of
EP
the patients with a positive stool DNA test had a negative colonoscopy. This may lead to additional workup for extracolonic cancers or prompt a shorter interval for repeat
AC C
colonoscopy, both of which are associated with significant cost and unknown benefit. Implications for Practice: Stool DNA testing appears to be a reasonable option for average risk patients who need colon cancer screening. There are several issues that still need further clarification to determine the role of stool DNA testing in programatic screening recommendations including cost effectiveness, mortality data, comparative effectiveness, recommended screening interval, and diagnostic evaluation of positive results.
Side effects from statin drugs are not as common as we think.(4) Background: Multiple studies have demonstrated the benefits of statin drugs for both primary and secondary prevention of cardiovascular disease and cerebrovascular disease. However, many side effects
ACCEPTED MANUSCRIPT have been attributed to statin therapy, increasing risk of patient nonadherence. Information on the causal association between statin use and side effects would aid in clinical decision making and may promote patient compliance. Results: This systematic review included 29 randomized controlled trials and 83,880 participants taking statins for primary and secondary prevention of cardiovascular events. While statins were associated
RI PT
with a small increase in the risk of diabetes in the primary prevention trials by 0.5% (95% CI 0.1-1%; p=0.012), no such increase was seen in the single secondary prevention trial that reported on diabetes. In both the primary and secondary prevention trials, statin over placebo increased the rate of an asymptomatic rise in liver transaminases (>3 upper limit normal) by 0.4%. No other symptom, such as
rhabdomyolysis or fatigue was increased in the statin group.
SC
gastrointestinal disturbance, myalgias, renal disorders, CK elevation, newly diagnosed cancer,
M AN U
Limitations: The primary limitation was inconsistent reporting of adverse events. For example, 24 of the 30 trials reported data for liver enzyme abnormalities, but only three of the trials reported on newly diagnosed diabetes mellitus. Furthermore, many of the trials tested low-strength statin regimens; moderate and high intensity regimens may be associated with more side effects. Implications for Practice: This review provides some reassurance that side effects from statin drugs are not as common as clinical practice might lead us to believe. Often patients report symptoms that are
TE D
attributed to statin side effect which probably are not. Given the established benefit of statins, we should encourage patients to continue statin therapy unless there is convincing evidence that their symptoms are a true side effect.
patients.(5, 6)
EP
The U.S. Preventive Services Task Force now recommends lung cancer screening in high risk
AC C
Background: The U.S. Preventative Services Task Force (USPSTF) recently issued new lung cancer screening guidelines, which include low-dose computed tomography (LDCT) in patients 55-80 years who have a cumulative 30 pack-year history and currently smoke or have quit within the last fifteen years. After guideline release, the Centers for Medicare and Medicaid Services convened a Medicare Evidence Development and Coverage Advisory Committee (MEDCAC) to address age-related screening effectiveness and potential harms, as only 25% of patients in the National Lung Screening Trial (NLST), upon which the USPSTF guideline was largely based, were aged 65 years or older. Results: The NLST enrolled 53,454 patients (aged 55-74 years) and compared LDCT to single view chest radiograph, finding a relative reduction in mortality from lung cancer with low-dose CT screening of 20.0% (95% CI 6.8 to 26.7; P=0.004). The USPSTF then used modeling studies to predict benefits
ACCEPTED MANUSCRIPT and harms and found insufficient evidence for harm related to incidental findings and false positive LDCT scans, noting that screening may not be appropriate in those at the upper end of the age range, particularly when significant comorbid conditions exist. In the analysis of Medicare age patients versus those under age 65, the positive predictive value (PPV) was higher in the 65+ age group (4.9% vs 3.0%, p=
