- Email: [email protected]
Clinical Characteristics and Outcomes of Patients With Heart Failure and Methamphetamine Abuse
Methamphetamine and Heart Failure
Journal Pre-proof
Clinical Characteristics and Outcomes of Patients with Heart Failure and Methamphetamine Abuse Isac C. Thomas MD, MPH , Marin Nishimura MD , Janet Ma MD , Stephen D. Dickson MD , Laith Alshawabkeh MD, MSCI , Eric Adler MD , Alan Maisel MD , Michael H. Criqui MD, MPH , Barry Greenberg MD PII: DOI: Reference:
S1071-9164(19)30484-1 https://doi.org/10.1016/j.cardfail.2019.10.002 YJCAF 4428
To appear in:
Journal of Cardiac Failure
Received date: Revised date: Accepted date:
1 May 2019 26 August 2019 7 October 2019
Please cite this article as: Isac C. Thomas MD, MPH , Marin Nishimura MD , Janet Ma MD , Stephen D. Dickson MD , Laith Alshawabkeh MD, MSCI , Eric Adler MD , Alan Maisel MD , Michael H. Criqui MD, MPH , Barry Greenberg MD , Clinical Characteristics and Outcomes of Patients with Heart Failure and Methamphetamine Abuse, Journal of Cardiac Failure (2019), doi: https://doi.org/10.1016/j.cardfail.2019.10.002
This is a PDF file of an article that has undergone enhancements after acceptance, such as the addition of a cover page and metadata, and formatting for readability, but it is not yet the definitive version of record. This version will undergo additional copyediting, typesetting and review before it is published in its final form, but we are providing this version to give early visibility of the article. 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. © 2019 Published by Elsevier Inc.
Highlights
Methamphetamine abuse is associated with the development of the heart failure syndrome. Methamphetamine abuse is a global epidemic that has resulted in an increasing number of patients with MethHF.
Rates of readmission for heart failure and total mortality observed in this young, predominantly male patient population were poor, and comparable to older patients with heart failure due to other causes.
Co-morbid substance abuse and psychiatric disorders may have contributed to these poor outcomes.
Efforts directed toward substance abuse prevention paired with comprehensive clinical efforts directed toward the treatment of MethHF and associated comorbidities may improve outcomes.
1
Title: Clinical Characteristics and Outcomes of Patients with Heart Failure and Methamphetamine Abuse Short Title: Methamphetamine and Heart Failure Authors: Isac C. Thomas MD, MPH* Marin Nishimura MD* Janet Ma MD* Stephen D. Dickson MD* Laith Alshawabkeh MD, MSCI* Eric Adler MD* Alan Maisel MD* Michael H. Criqui MD, MPH*Ɨ Barry Greenberg MD* Affiliations: * Division of Cardiovascular Medicine, Department of Medicine, University of California, San Diego Ɨ Division of Preventive Medicine, Department of Family Medicine and Public Health, University of California, San Diego Address for Correspondence: Isac C. Thomas, MD, MPH Department of Medicine Division of Cardiovascular Medicine University of California, San Diego 9452 Medical Center Drive La Jolla, CA 92037-7411 Tel: 858-246-2989, Fax: 858-246-2958 Email: [email protected] Disclosures: All of the authors report no conflicts of interest related to this manuscript. Source of Funding: The Project described was partially supported by the National Institutes of Health, Grant UL1TR001442 of CTSA. The content is solely the responsibility of the authors and does not necessarily represent the official views of the NIH.
2
Abstract Background: Despite a global epidemic of methamphetamine abuse, methamphetamine-associated heart failure (MethHF) remains poorly understood. We sought to evaluate characteristics and outcomes for patients with MethHF. Methods: We reviewed the electronic health records of the University of California, San Diego from 2005-16. We compared characteristics and outcomes between 896 patients with MethHF and 20,576 patients with heart failure (HF) identified using diagnosis codes, urine toxicology, and natriuretic peptides. Results: Compared to HF, patients with MethHF were younger (50±10 v. 67±16 years), predominantly male (72% v. 54%), and had more psychiatric and substance use co-morbidities, including mood/anxiety disorders (29% v. 16%) and opioid use (44% v. 7%). MethHF had a higher 5-year HF readmission rate [64±4% v. 45±1%; hazard ratio (HR) 1.53, p<0.001] and a lower 10-year total mortality rate (25±3% v. 28±1%; HR 0.85, p=0.09). Predictors of poor outcomes included mood/anxiety disorders (HF readmission HR 1.41, p=0.04) and opioid abuse (mortality HR 1.52, p=0.04). Conclusions: Patients with MethHF are frequently encumbered by psychiatric and substance abuse comorbidities, and carry a substantial risk of heart failure readmission and mortality. Comprehensive efforts are needed to stem this emerging epidemic.
Key Words: Heart failure, methamphetamine, substance abuse, cardiomyopathy, readmission, mortality
1
Introduction Methamphetamine and other amphetamine-type stimulant drugs rank among the most commonly abused illicit substances worldwide, second only to marijuana in prevalence of use and opioids in associated harm.1 While its abuse is widespread across the United States, in Western and Southeastern states methamphetamine has been identified as the single greatest drug threat by the Drug Enforcement Agency.2 In San Diego, California, for instance, methamphetamine is the most common primary drug of abuse among patients admitted to drug treatment programs, and rates of emergency department visits associated with methamphetamine abuse have been rising precipitously.3, 4
Methamphetamine abuse is associated with myocardial injury, leading to marked reductions in left ventricular systolic function and a non-ischemic, dilated cardiomyopathy.5, 6 Methamphetamine abuse has also been associated with pulmonary hypertension in the absence of left ventricular dysfunction, and can cause acute elevations in blood pressure and heart rate.7, 8 With a growing global epidemic of methamphetamine abuse, a commensurate increase in methamphetamine-associated heart failure (MethHF) is expected. Yet characteristics of patients with MethHF are poorly described in the medical literature, hindering efforts toward delivering effective care to this population. In order to better characterize and understand this disease, we evaluated patients in a region with a high prevalence of methamphetamine abuse and present the clinical characteristics and outcomes of patients with MethHF.
Methods Study Sample The Institutional Review Board of the University of California, San Diego (UCSD) approved the study protocol. We retrospectively evaluated the electronic health record from the UCSD healthcare system. We abstracted International Classification of Diseases, 9th Revision (ICD-9) diagnosis codes as
2
well as results of urine drug toxicology and serum natriuretic peptides to identify patients with methamphetamine abuse, heart failure, or MethHF. We defined methamphetamine abuse as the presence of the ICD-9 code for amphetamine dependence (304.4) or amphetamine abuse (305.7) (ICD-9 codes specific for methamphetamine are not available), and/or a urine drug toxicology screen positive for the amphetamines drug class and/or methamphetamine in particular (methamphetamine-specific toxicology reporting became available in 2013). We defined heart failure as the presence of ICD-9 code for heart failure (428) and/or an elevated serum brain natriuretic peptide (BNP) (>99 pg/ml) or Nterminal proBNP (>449 pg/ml). Although BNP and/or pro-BNP elevations can be nonspecific for heart failure, we included them in the selection criteria to minimize potential misclassification that may occur due to the under-diagnosis of heart failure particularly among patients who abuse methamphetamine. Using the above selection criteria, we identified 25,942 unique patients who were divided into three mutually-exclusive cohorts: 4,470 patients with methamphetamine abuse and no evidence of heart failure (methamphetamine abuse cohort), 20,576 patients with heart failure and no evidence of methamphetamine abuse (heart failure cohort), and 896 with both methamphetamine abuse and heart failure (MethHF cohort). For patients in the MethHF cohort, if the methamphetamine abuse diagnosis preceded the heart failure diagnosis or vice versa, study entry was at the time of the second diagnosis being fulfilled, with baseline characteristics gathered then and follow up proceeding thereafter. We included patients in the MethHF cohort whether the diagnoses of methamphetamine abuse and heart failure occurred simultaneously or at different points over the study period in an effort to minimize misclassification of these patients into the methamphetamine abuse cohort or heart failure cohort. ICD-9 codes were abstracted from both inpatient and outpatient encounters. We abstracted data between the dates of January 1st, 2005 and June 30th, 2016, prior to the transition to ICD-10 coding which occurred in the latter half of 2016. We also assessed the accuracy of ICD-9 coding for heart failure and methamphetamine abuse in addition to the accuracy of the BNP/pro-BNP criteria for heart failure
3
by manually reviewing of the electronic health record for three non-overlapping randomly selected samples: 50 patients with an ICD-9 diagnosis of heart failure, 50 patients with elevations in proBNP (>449 pg/ml) or BNP (>99 pg/ml), and 50 patients with an ICD diagnosis of amphetamine abuse or dependence. For purposes of this manual chart review, we defined clinical heart failure as one or more of the following: inpatient admission for heart failure with intravenous diuretics administered, echocardiogram with LVEF < 40%, or an outpatient encounter with a cardiologist documenting a diagnosis and treatment plan for heart failure. We defined clinical methamphetamine abuse as documentation of methamphetamine abuse or dependence in provider notes.
Clinical Characteristics and Outcomes Clinical characteristics were abstracted from the electronic health record through a combination of ICD-9 codes and echocardiogram data. We evaluated ICD-9 codes relating to several common cardiovascular, psychiatric, and substance abuse co-morbidities. A full list of ICD-9 codes is included in Supplemental Table A. Substance use/abuse was defined by the presence of a positive urine drug toxicology screen or an ICD-9 code for the substance of interest. Transthoracic echocardiogram data were available for 623 patients (14%) in the methamphetamine abuse cohort, 14,443 patients (70%) in the HF cohort, and 731 patients (82%) in the MethHF cohort. When multiple echocardiograms were available for a given patient, the lowest left ventricular ejection fraction (LVEF) and highest pulmonary artery systolic pressure (PASP) were recorded. We also abstracted the date and total number of hospital encounters and mortality over the follow-up period for each patient. The follow-up period extended from the time of first diagnosis (of methamphetamine abuse, heart failure, or MethHF) to patient death or June 30th, 2016. Hospital encounters were defined as any emergency department visit, observation admission, or inpatient admission. These encounters were further characterized by whether heart failure was the primary diagnosis for the encounter. We defined heart failure readmission as the first hospital admission
4
(observation or inpatient) with a primary diagnosis of heart failure that occurred after an index admission with a primary diagnosis of heart failure. We defined mortality as death due to any cause, with vital status abstracted from the electronic health record.
Statistical Analysis We plotted the biennial tally of unique patients diagnosed with methamphetamine abuse and MethHF starting in 2005. We compared characteristics of the methamphetamine abuse, heart failure, and MethHF cohorts using z-tests of proportions with a Bonferroni correction for categorical variables and analysis of variance (ANOVA) with a Bonferroni correction for continuous variables. In cross-sectional analyses, we combined the methamphetamine abuse and MethHF cohorts and constructed a multivariable logistic regression model to identify factors associated with the presence of a heart failure diagnosis among patients who abuse methamphetamine. We also combined the heart failure and MethHF cohorts and constructed a multivariable logistic regression model to identify factors associated with methamphetamine abuse among patients with a diagnosis of heart failure. For both models, we included multiple potential exposure variables in a saturated model based on observed differences in baseline patient characteristics between the cohorts and characteristics that we hypothesized may differentiate the cohorts. We then utilized backward selection with a p-value of 0.10 for exclusion to create final models. We present adjusted odds ratios with 95% confidence intervals (CI) for all variables in the final models. In order to evaluate the association of methamphetamine abuse with adverse outcomes in patients with heart failure, we performed longitudinal analyses comparing the heart failure and MethHF cohort. We present Kaplan Meier curves evaluating the cumulative incidence of heart failure readmission as well as total mortality, comparing the heart failure and MethHF cohorts using log-rank testing. We constructed a Cox regression model for time to death and present the hazard ratio (HR) for methamphetamine abuse among patients with heart failure adjusted for demographics (age, sex,
5
race/ethnicity, and health insurance status). As death can be a competing risk for heart failure readmission, we performed a competing risk regression according to the method described by Fine and Gray,9 and report the demographic-adjusted HR for methamphetamine abuse. We also tallied the total number of hospital encounters over follow up (emergency department visits, and observation and inpatient admissions) and plotted the results, with comparisons between strata performed using posthoc z-tests of proportions with Bonferroni corrections. Finally, we constructed Cox regression models to identify independent predictors of heart failure readmission and total mortality among patients in the MethHF cohort, adjusting for demographics, LVEF, PASP, and co-morbid diagnoses, and sequentially excluded terms using backward selection with a p-value of 0.10 to produce final models. We present adjusted hazard ratios with 95% confidence intervals (CI) for all variables in the final models. As BNP and proBNP elevations can be non-specific for heart failure, we performed two sensitivity analyses to investigate potential bias that may have resulted from their use in the selection criteria. In the first analysis, we doubled the threshold for elevation for BNP (>99 to >199 pg/ml) and proBNP (>449 to >899 pg/ml). In the second analysis, we removed BNP and proBNP from the selection criteria for heart failure. We performed statistical analyses using R Statistical Software (version 3.2.2; R Foundation for Statistical Computing, Vienna, Austria) and SPSS Statistics version 22.0 (IBM Corporation, Armonk, NY, USA). We considered a p-value of <0.05 statistically significant, with Bonferroni-correction when multiple testing was performed.
6
2060
Number of Unique Patients
Methamphetamine Abuse MethHF 1499
446 62 92 '05-'06
178
121
'07-'08
225
165
114 '09-'10
'11-'12
212
'13-'14
192
'15-June '16
Figure 1: Trends in the diagnosis of methamphetamine abuse and MethHF. Biennial tally of unique patients diagnosed with methamphetamine abuse alone or methamphetamine abuse with heart failure (MethHF). The methamphetamine abuse cohort and the MethHF cohort are mutually-exclusive.
Results Trends in Methamphetamine Abuse and MethHF The incidence of methamphetamine abuse and MethHF increased substantially over the study period. As shown in Figure 1, 62 patients were diagnosed with methamphetamine abuse and 92 were diagnosed with MethHF in 2005-06; in 2015-16, 2,060 patients were diagnosed with methamphetamine abuse and 192 were diagnosed with MethHF. Pronounced increases in methamphetamine abuse occurred during the most recent four years of the study period. In manual review of medical records, a diagnosis of heart failure via ICD-9 code corresponded with documented heart failure in 82% of charts reviewed. A diagnosis of heart failure via BNP/proBNP elevation corresponded with documented heart failure in 90% of charts reviewed. A diagnosis of
7
methamphetamine abuse via ICD-9 code and/or urine toxicology corresponded with documented methamphetamine abuse in 84% of charts reviewed.
Clinical Characteristics Characteristics of patients in the methamphetamine abuse, heart failure, and MethHF cohorts are presented in Table 1. Patients in the MethHF cohort had a mean age of 50 ±10 years, were predominantly male (72%), majority white (57%), and frequently uninsured (52%). Co-morbid diagnoses such as hypertension (54%), ischemic heart disease (21%), mood/anxiety disorders (29%), and opioid use (44%) were common. In contrast, patients in the methamphetamine abuse cohort were younger (40 ±13 years) and generally with fewer co-morbid medical diagnoses but comparable levels of psychiatric and substance use diagnoses compared to the MethHF cohort. Patients in the heart failure cohort were, on average, 17 years older (67 ±16 years) than the MethHF cohort and with a lower percentage of men (54%) and African American race (9% v. 19%), similar or higher rates of co-morbid medical diagnoses, and lower rates of psychiatric and substance use diagnoses.
8
Demographics Age, years (standard deviation) Male Sex Race/Ethnicity White Hispanic/Latino African American Asian Other/Mixed No medical insurance Medical and Psychiatric Comorbidities Hypertension Diabetes Mellitus Ischemic Heart Disease Atrial Fibrillation/Flutter Chronic Kidney Disease Stroke Endocarditis Cardiogenic Shock Cardiac Arrest Ventricular Tachycardia/Fibrillation Mood/Anxiety Disorders Schizophrenia/Delusional Disorder Post-Traumatic Stress Disorder Substance Use Opioids Marijuana Alcohol Cocaine Phencyclidine (PCP)
Methamphetamine Abuse (n=4,470)
Heart Failure (n=20,576)
MethHF (n=896)
40.0 (12.7)* 71.6%
67.1 (15.7)* 54.4%*
49.6 (10.0) 72.4%
56%* 22.2%* 13.3%* 2.4%* 5.9% 28%*
59.3% 17.4% 9.4%* 6.7%* 6.8%* 24.4%*
57.7% 15.5% 18.5% 3.9% 4.2% 51.7%
16.7%* 5.8%* 2.1%* 0.4%* 1.7%* 2.3%* 1.1%* 0.0%* 0.4%* 0.1%* 36.0%* 12.1%* 3.2%
64.1%* 23.7% 17.7% 27.4%* 20.4%* 8.4% 1.8%* 1.5% 1.5% 0.8%* 16.8%* 1.4%* 0.5%*
54.4% 23.1% 21.4% 16.0% 15.5% 9.7% 3.1% 2.6% 1.8% 1.4% 28.6% 8.1% 2.1%
24.2%* 17.3% 20.8%* 3.0% 0.6%*
7.1%* 1.4%* 3.9%* 0.3%* 0.1%*
44.2% 20.0% 16.2% 3.9% 1.6%
Table 1: Characteristics of patients in the methamphetamine abuse, heart failure, and MethHF cohorts. Each cohort is mutually exclusive of the others. *indicates a Bonferroni corrected p-value (p< 0.05/3) for comparisons to MethHF.
9
Figure 2: Left ventricular ejection fraction and pulmonary artery systolic pressure for patients in the methamphetamine abuse, heart failure, and MethHF cohorts
Left Ventricular Ejection Fraction 3%
12%
33%
14%
18%
18% 17%
Severely Elevated (60+ mm Hg)
16%
Moderately Elevated 13%
Moderately Reduced (30-39%) Mildly Reduced (4049%)
5%
12% 11%
Severely Reduced (<30%)
Pulmonary Artery Systolic Pressure
(45-59 mm Hg)
11% Mildly Elevated (30-
95%
44 mm Hg)
64%
Normal (50%+)
Heart Failure (n=14,443)
33%
33%
33%
Heart Failure (n=14,443)
MethHF (n=731)
75%
Normal (<30mm Hg)
44%
Methamphetamine Abuse (n=623)
36%
MethHF (n=731)
Methamphetamine Abuse (n=623)
1
Figure 2 shows the distribution of left ventricular ejection fraction and pulmonary artery systolic pressure estimated by transthoracic echocardiogram for the three cohorts. Patients in the MethHF cohort had a higher proportion with LVEF<50% (56% v. 36%) and a higher proportion with an LVEF<30% (33% v. 12%) compared with patients in the heart failure cohort. Pulmonary artery systolic pressure (PASP) levels were elevated in 67% of the MethHF cohort, with similar percentages in the heart failure cohort. 25% of patients in the methamphetamine abuse cohort with available echocardiograms had evidence of elevated PASP.
Factors Associated with Heart Failure in Patients with Methamphetamine Abuse Table 2 lists adjusted odds ratios for variables associated with the presence of heart failure among patients diagnosed with methamphetamine abuse. The odds of heart failure were 25% higher with each 5-year increase in age (OR 1.25, 95% CI 1.20-1.31), and were nearly 2.5-fold higher for patients lacking medical insurance (OR 2.49, 95% CI 2.06-3.02). Compared to white patients, patients of African American (OR 1.48, 95% CI 1.14-1.92) and Asian/Pacific Islander (OR 2.22, 95% CI 1.33-3.71) race/ethnicity had higher odds of heart failure. Atrial fibrillation, ischemic heart disease, and chronic kidney disease were among several co-morbid medical diagnoses associated with heart failure, whereas mood/anxiety disorders and schizophrenia were inversely associated with heart failure. Opioid use was associated with heart failure (OR 2.12, 95% CI 1.74-2.58), while alcohol abuse was inversely associated with HF (OR 0.74, 95% CI 0.58-0.94).
1
Adjusted Odds Ratio (95% confidence interval) Older Age (per 5 year increase)
1.25 (1.20, 1.31)
Male Sex
1.37 (1.11, 1.68)
Race/Ethnicity White
Reference
Hispanic/Latino
1.06 (0.83, 1.37)
African American
1.48 (1.14, 1.92)
Asian/Pacific Islander
2.22 (1.33, 3.71)
Other/Mixed
1.04 (0.66, 1.64)
Atrial Fibrillation/Flutter
27.40 (15.80, 47.50)
Ischemic Heart Disease
6.51 (4.75, 8.94)
Chronic Kidney Disease
3.57 (2.46, 5.17)
No Medical Insurance
2.49 (2.06, 3.02)
Hypertension
2.38 (1.93, 2.94)
Opioid Use
2.12 (1.74, 2.58)
Diabetes Mellitus
1.57 (1.20, 2.06)
Cerebrovascular Accident
1.45 (0.99, 2.12)
Alcohol Abuse
0.74 (0.58, 0.94)
Schizophrenia/Delusional Disorder
0.66 (0.48, 0.90)
Mood/Anxiety Disorder
0.63 (0.51, 0.77)
Table 2: Factors associated with heart failure among patients diagnosed with methamphetamine abuse. Saturated model includes the above variables and endocarditis, post-traumatic stress disorder, cocaine, phencyclidine, and marijuana use. The variables listed above were selected for the final multivariable model using backward selection with a p-value for exclusion of 0.10.
Factors Associated with Methamphetamine Abuse in Patients with Heart Failure Table 3 displays adjusted odds ratios for variables associated with the diagnosis of methamphetamine abuse among patients with heart failure. Each 5-year decrement in age was associated with a 32% higher odds of methamphetamine abuse (OR 1.32, 95% CI 1.28-1.36). Male sex (OR 1.57, 95% CI 1.30-1.91) and white race where also associated with a higher odds of methamphetamine abuse. Schizophrenia and all of the substances abuse disorders were associated with
2
higher odds of methamphetamine abuse. Among patients with heart failure, each 5% decrement in LVEF (OR 1.13, 1.10-1.16) and 10 mm Hg increase in PASP (OR 1.05, 1.01-1.08) were associated with a higher likelihood of methamphetamine abuse.
Adjusted Odds Ratio (95% confidence interval) Younger Age (per 5 year decrease)
1.32 (1.28, 1.36)
Lower LVEF (per 5%)
1.13 (1.10, 1.16)
Higher PASP (per 10 mm Hg)
1.05 (1.01, 1.08)
Male Sex 1.57 (1.30, 1.91) Race/Ethnicity White Reference Hispanic/Latino 0.56 (0.43, 0.72) African American 0.72 (0.56, 0.93) Asian/Pacific Islander 0.67 (0.44, 1.01) Other/Mixed 0.74 (0.48, 1.14) Phencyclidine Abuse 4.25 (1.79, 10.12) Opioid Abuse 4.19 (3.46, 5.09) Cocaine Abuse 4.03 (2.31, 7.03) Marijuana Abuse 3.33 (2.52, 4.42) Schizophrenia/Delusional Disorder 2.45 (1.68, 3.58) Alcohol Abuse 1.59 (1.22, 2.07) Chronic Kidney Disease 0.59 (0.47, 0.73) Atrial Fibrillation/Flutter 0.58 (0.46, 0.72) Table 3: Factors associated with methamphetamine abuse among patients diagnosed with heart failure. Saturated model includes the above variables and medical insurance, hypertension, diabetes mellitus, ischemic heart disease, cerebrovascular accident, endocarditis, mood/anxiety disorders, and post-traumatic stress disorder. The variables listed above comprise the final model.
3
60%
p<.01
Heart Failure MethHF
Percentage of Patients
50%
40% p>.01
30% p<.01
p<.01
20% p<.01
10%
0%
0 or 1
2 to 5 6 to 10 11 to 15 No. of Hospital Encounters
16+
Figure 3: Distribution of hospital encounters for patients in the heart failure and MethHF cohorts. Comparisons between the heart failure and MethHF cohorts were made at a Bonferroni-corrected significance level of 0.01.
Clinical Outcomes and Predictors of Readmission and Mortality in MethHF Figure 3 displays the distribution of emergency department, observation, and inpatient hospital encounters for patients in the heart failure and MethHF cohorts. Median follow up was 4.82 [interquartile range 1.98 to 8.36] years for the heart failure cohort and 3.95 [1.71 to 7.30] years for the MethHF cohort. Over follow up, 75,589 hospital encounters occurred between these two cohorts with a median of 1 [0 to 4] encounters for the heart failure cohort and 4 [1 to 11] encounters for the MethHF cohort (p<0.001).
4
Figure 4: Cumulative incidence of heart failure readmission and total mortality for the heart failure and MethHF cohorts. Hazard ratios are adjusted for age, sex, race/ethnicity, and health insurance status.
Over follow-up, 2,857 heart failure readmissions and 4,145 deaths occurred. Figure 4 displays cumulative incidence curves for heart failure readmission and total mortality for the heart failure and MethHF cohorts. Patients in the MethHF cohort had a higher risk of heart failure readmission (log-rank p <0.001) with a 5-year rate of 64 ±4% v. 45 ±1% in the heart failure cohort. Among patients with heart failure, methamphetamine abuse was associated with a HR of 1.53 (95% CI 1.31-1.78, p<0.001) for HF readmission after adjustment for demographics and the competing risk of death. Patients in the MethHF cohort had a lower risk of total mortality (log-rank p<0.001) with a 10-year mortality of 25 ±3% v. 28 ±1%. Among patient with heart failure, methamphetamine abuse was associated with a HR of 0.85 (95% CI 0.71-1.03, p=0.09) for total mortality after adjustment for demographics.
5
Adjusted Hazard Ratio (95% Confidence Interval) Heart Failure Readmission Lower LVEF (per 5%)
1.15 (1.10, 1.20)
Higher PA pressure (per 10 mm Hg)
1.07 (1.03, 1.11)
Stroke
1.49 (1.01, 2.22)
Mood/Anxiety Disorders
1.41 (1.01, 1.96)
Chronic Kidney Disease
1.36 (0.98, 1.88)
Total Mortality Older Age (per 5 years)
1.18 (1.07, 1.30)
African American (v. White)
2.09 (1.36, 3.20)
Opioid Abuse
1.52 (1.07, 2.27)
Mood/Anxiety Disorders
1.47 (0.99, 2.19)
Hypertension
0.62 (0.42, 0.93)
Table 4: Predictors of heart failure readmission and total mortality for patients in the MethHF cohort. The saturated models for each outcome include age, sex, race/ethnicity, medical insurance, hypertension, diabetes, ischemic heart disease, endocarditis, atrial fibrillation, cerebrovascular accident, chronic kidney disease, mood/anxiety disorder, schizophrenia, post-traumatic stress disorder, opioid use, alcohol abuse, marijuana abuse, cocaine abuse, phencyclidine abuse, left ventricular ejection fraction, and pulmonary artery systolic pressure. The variables listed above comprise the final models for each outcome.
Table 4 displays adjusted hazard ratios for predictors of heart failure readmission and total mortality among patients in the MethHF cohort. Reductions in LVEF (HR 1.15, 95% CI 1.10-1.20 per 5% decrease), elevations in PASP (HR 1.07, 95% CI 1.03-1.11 per 10 mm Hg increase), stroke (HR 1.49, 95% CI 1.01, 2.22), and mood/anxiety disorders (HR 1.41, 95% CI 1.01-1.96) were all independently increased the risk of a HF readmission. While LVEF and PASP were not independently associated with mortality in this model, older age (HR 1.18, 95% CI 1.07-1.30 per 5 years), African American race (HR 2.09, 95% CI 1.36-3.20 v. white race), and opioid use (HR 1.52, 95% CI 1.07-2.27) were associated with a higher risk of mortality.
6
In sensitivity analyses, doubling the threshold for BNP/ proBNP elevation in the heart failure selection criteria did not materially change the results for heart failure readmission and total mortality in the heart failure and MethHF cohorts. Removing BNP/proBNP from the heart failure selection criteria resulted in a reduction in the sample size for both cohorts, but the comparisons between the cohorts for heart failure readmission and total mortality were maintained. Detailed results are presented in Supplemental Table B.
Discussion Drug abuse has increased with the rise in access to illicit substances and physician-prescribed medications with addiction potential.3 Although widespread opioid abuse has resulted in significant harm in recent years,5 the abuse of methamphetamine and other amphetamine-type substances is regarded as the most significant threat to public health in much of the United States and in many countries across the world.2 Methamphetamine abuse has a long-documented association with cardiomyopathy and heart failure via mechanisms that have not been fully elucidated,10-12 but may be similar to those found in cardiomyopathies associated with cocaine and other high catecholamine conditions.13, 14 Chronic abuse appears to lead to myocardial inflammation and fibrosis, coupled with marked reductions in LV systolic function that may be reversible with methamphetamine cessation in some cases.5, 15 Despite this potential for recovery and the relatively young ages of these patients, MethHF is associated with poor heart failure-related clinical outcomes.5, 8 In this study, we found that patients in the MethHF cohort were younger than those in the HF cohort, yet had relatively high rates of several medical co-morbidities including hypertension, diabetes, ischemic heart disease, and chronic kidney disease. These findings suggest that methamphetamine abuse may not be the sole cause of heart failure for a number of patients with MethHF, particularly in areas where methamphetamine abuse is highly prevalent. Rather, methamphetamine abuse may be a
7
contributor to heart failure, particularly among patients at heightened risk due to the presence of traditional heart failure risk factors. Patients in the MethHF cohort also had relatively high rates of the psychiatric and substance abuse disorders that were commonly found in the methamphetamine abuse cohort, highlighting the disproportionate burden of a range of co-morbidities among patients with MethHF. We also found that clinical outcomes for patients in the MethHF cohort were commensurate with or worse than those of patients in the heart failure cohort. Despite their mean age being 17 years lower than the heart failure cohort, patients in the MethHF cohort had only a slightly lower risk of total mortality, and as expected this lower risk was not significant with age adjustment. Perhaps owing to underlying medical, psychiatric, and substance abuse co-morbidities, patients with MethHF had a substantially higher rate of hospital encounters for any cause over the follow up period, and a higher risk of heart failure readmission. We found that while traditional risk factors predict heart failure readmission and mortality in the MethHF cohort, opioid use and mood/anxiety disorders also independently contribute to heightened risk. Opioid use was highly prevalent in the MethHF cohort and, in line with the national opioid abuse epidemic, a source of significant harm. Psychiatric disorders such as depression are associated with poor clinical outcomes among patients with cardiovascular disease including heart failure.16 Clinical trials utilizing selective serotonin reuptake inhibitors to reduce the risk of cardiovascular outcomes in patients with mood disorders and cardiovascular disease have shown mixed results.17, 18 Nevertheless, an emphasis on the detection and treatment of psychiatric comorbidities and concomitant substance abuse is likely necessary to improve outcomes among patients with MethHF. Effective prevention of MethHF will likely center on efforts toward early methamphetamine abuse detection and cessation, with a heightened emphasis on patients at greatest risk of developing heart failure. In patients who abuse methamphetamine, we found that heart failure was strongly
8
associated with older age, African American and Asian/Pacific Islander races, and a lack of medical insurance. While relatively uncommon among these patients, factors such as atrial fibrillation, chronic kidney disease, and ischemic heart disease were strongly associated with heart failure. This latter finding highlights the fact that although MethHF is understood to be due to a dilated cardiomyopathy secondary to a toxic exposure, the contributions of traditional heart failure risk factors should be considered, particularly in older patients. Surprisingly, we found alcohol abuse, mood/anxiety disorders, and schizophrenia to be inversely associated with heart failure among patients who abuse methamphetamine. Although alcohol itself has a known toxic effect on the myocardium, concomitant alcohol abuse may lessen the intake of methamphetamine, thereby exerting a “protective” effect. Alternatively, these co-morbidities may be frequently diagnosed among patients with methamphetamine abuse and under-diagnosed among patients with heart failure, leading to a spurious inverse association. As substance abuse and psychiatric disorders affect treatment strategies and outcomes among patients with MethHF, providers should screen for these disorders and treat accordingly. Substance abuse is often hidden by patients and difficult for providers to diagnose.19 In patients with a diagnosis of heart failure, we found that a reduction in LVEF, as well as younger age, male sex, white race, concomitant substance abuse, and schizophrenia were strongly correlated with methamphetamine abuse. As methamphetamine and its metabolites are only detectable in the urine for approximately 48 hours after exposure,20 urine drug screens alone may not be sufficient to detect methamphetamine abuse, and repeated inquiry and urine testing of patients at high risk of methamphetamine abuse are often necessary. Our findings add to the relatively limited literature on patients with MethHF. In a recent study, Schürer et al. characterized tissue obtained from endomyocardial biopsies on 30 patients with MethHF and found that the duration of methamphetamine abuse correlated with the extent of tissue fibrosis,
9
and that the extent of tissue fibrosis predicted the likelihood of recovery of LV function after methamphetamine cessation.5 These findings demonstrate the importance of early detection and intervention among patients with MethHF. Zhao et al. described clinical characteristics for patients with MethHF as well as patients with methamphetamine associated pulmonary arterial hypertension, and showed that the mortality risk for both groups was high.8 We found that PASP was elevated in 25% of patients who abuse methamphetamine without a diagnosis of HF, as possibly an early manifestation of the effect of methamphetamine on the cardiovascular system. Pulmonary arterial hypertension in the absence of LV systolic dysfunction is an increasingly recognized manifestation of methamphetamine abuse, and the development of cardiomyopathy versus pulmonary arterial hypertension may be linked to the mode of consumption (i.e. snorting, smoking, ingestion or intravenous injection).21
Limitations Our study evaluates a large cohort of patients with MethHF and identifies important predictors of adverse clinical outcomes in these patients. Our study also has limitations. Since our study relies on the diagnosis of methamphetamine abuse via ICD-9 coding and screening with urine toxicology, we have potentially underestimated the true prevalence of methamphetamine abuse in our health care system, and given rising trends in methamphetamine abuse, many patients classified in the heart failure cohort may have also been abusing methamphetamines. Moreover, heart failure may not have been investigated or diagnosed in many young patients in the methamphetamine abuse cohort. Rising rates of methamphetamine abuse that we observed in this study may be in part due to increased awareness and diagnosis of methamphetamine abuse, rather than simply due to increased abuse of methamphetamines. However, these trends have been paralleled by increasing arrests for methamphetamine sales and possession reported by local law enforcement.4 As outcomes were obtained from the UCSD electronic health record, individuals who abuse methamphetamine who do not seek medical attention are not represented, hospitalizations that occurred outside of our health care
10
system were not captured, and deaths that occurred outside of our health care system were likely under-reported. ICD-9 codes and urine toxicology for the majority of the study period were for amphetamines in general (not methamphetamine in particular), and thus we could not exclude the use or abuse of other amphetamine–type stimulants, such as MDMA (ecstasy), or amphetamine containing prescription drugs. Finally, our study population is from a region with a high prevalence of methamphetamine abuse, and thus these results may not be generalizable.
Conclusions As methamphetamine abuse grows increasingly common, health care providers are likely to encounter an epidemic of MethHF in the coming years. Increasing amounts and duration of exposure to methamphetamine are associated with the development and severity of heart failure,13 suggesting that early intervention can reduce the burden of this disease. Innovative therapies directed at treating methamphetamine addiction are needed and an understanding of the underlying mechanisms of MethHF development are necessary in order to develop novel therapeutic interventions. Given the complex underpinnings of MethHF, comprehensive multidisciplinary efforts are needed to stem this emerging epidemic.
11
References 1.
United Nations Office on Drugs and Crime. World Drug Report 2017. ISBN: 978-92-1-148291-1,
eISBN: 978-92-1-060623-3, United Nations publication, Sales No. E.17.XI.6. 2.
United States Department of Justice Drug Enforcement Administration. 2017 National Drug
Threat Assessment. DEA Strategic Intelligence Section. October 2017. 3.
National Institute on Drug Abuse. Nationwide Trends. https://www.drugabuse.gov/
publications/drugfacts/nationwide-trends. Accessed July 30, 2018. 4.
Methamphetamine Strike Force. Local Statistics. San Diego County Health and Human Services
Administration, Behavioral Health Services. https://www.no2meth.org/local-statistics. Accessed November 11, 2018. 5.
Schurer S, Klingel K, Sandri M, Majunke N, Besler C, Kandolf R, et al. Clinical Characteristics,
Histopathological Features, and Clinical Outcome of Methamphetamine-Associated Cardiomyopathy. JACC Heart failure. 2017;5:435-445. 6.
Neeki MM, Kulczycki M, Toy J, Dong F, Lee C, Borger R, et al. Frequency of Methamphetamine
Use as a Major Contributor Toward the Severity of Cardiomyopathy in Adults ≤50 Years. American Journal of Cardiology. 2016;118:585-589. 7.
Frishman WH, Del Vecchio A, Sanal S and Ismail A. Cardiovascular manifestations of substance
abuse: part 2: alcohol, amphetamines, heroin, cannabis, and caffeine. Heart disease (Hagerstown, Md). 2003;5:253-71. 8.
Zhao SX, Kwong C, Swaminathan A, Gohil A and Crawford MH. Clinical Characteristics and
Outcome of Methamphetamine-Associated Pulmonary Arterial Hypertension and Dilated Cardiomyopathy. JACC: Heart Failure. 2018;6:209-218. 9.
Fine JP and Gray RJ. A Proportional Hazards Model for the Subdistribution of a Competing Risk.
Journal of the American Statistical Association. 1999;94:496-509.
12
10.
Smith HJ, Roche AH, Jausch MF and Herdson PB. Cardiomyopathy associated with amphetamine
administration. American heart journal. 1976;91:792-7. 11.
Hong R, Matsuyama E and Nur K. Cardiomyopathy associated with the smoking of crystal
methamphetamine. Jama. 1991;265:1152-4. 12.
Islam MN, Kuroki H, Hongcheng B, Ogura Y, Kawaguchi N, Onishi S, et al. Cardiac lesions and
their reversibility after long term administration of methamphetamine. Forensic science international. 1995;75:29-43. 13.
Sliman S, Waalen J and Shaw D. Methamphetamine-Associated Congestive Heart Failure:
Increasing Prevalence and Relationship of Clinical Outcomes to Continued Use or Abstinence. Cardiovascular toxicology. 2015. 14.
Won S, Hong RA, Shohet RV, Seto TB and Parikh NI. Methamphetamine-Associated
Cardiomyopathy. Clinical cardiology. 2013;36:737-742. 15.
Voskoboinik A, Ihle JF, Bloom JE and Kaye DM. Methamphetamine-associated cardiomyopathy:
patterns and predictors of recovery. Internal medicine journal. 2016;46:723-7. 16.
Rutledge T, Reis VA, Linke SE, Greenberg BH and Mills PJ. Depression in Heart Failure: A Meta-
Analytic Review of Prevalence, Intervention Effects, and Associations With Clinical Outcomes. Journal of the American College of Cardiology. 2006;48:1527-1537. 17.
Kim JM, Stewart R, Lee YS, Lee HJ, Kim MC, Kim JW, et al. Effect of Escitalopram vs Placebo
Treatment for Depression on Long-term Cardiac Outcomes in Patients With Acute Coronary Syndrome: A Randomized Clinical Trial. Jama. 2018;320:350-358. 18.
O'Connor CM, Jiang W, Kuchibhatla M, Silva SG, Cuffe MS, Callwood DD, et al. Safety and
Efficacy of Sertraline for Depression in Patients With Heart Failure: Results of the SADHART-CHF Trial. Journal of the American College of Cardiology. 2010;56:692-699.
13
19.
Diercks DB, Fonarow GC, Kirk JD, Jois-Bilowich P, Hollander JE, Weber JE, et al. Illicit stimulant
use in a United States heart failure population presenting to the emergency department (from the Acute Decompensated Heart Failure National Registry Emergency Module). The American journal of cardiology. 2008;102:1216-9. 20.
Oyler JM, Cone EJ, Joseph RE, Moolchan ET and Huestis MA. Duration of Detectable
Methamphetamine and Amphetamine Excretion in Urine after Controlled Oral Administration of Methamphetamine to Humans. Clinical Chemistry. 2002;48:1703. 21.
Ben-Yehuda O and Siecke N. Crystal Methamphetamine: A Drug and Cardiovascular Epidemic.
JACC Heart failure. 2018;6:219-221.
14
Journal Pre-proof
Clinical Characteristics and Outcomes of Patients with Heart Failure and Methamphetamine Abuse Isac C. Thomas MD, MPH , Marin Nishimura MD , Janet Ma MD , Stephen D. Dickson MD , Laith Alshawabkeh MD, MSCI , Eric Adler MD , Alan Maisel MD , Michael H. Criqui MD, MPH , Barry Greenberg MD PII: DOI: Reference:
S1071-9164(19)30484-1 https://doi.org/10.1016/j.cardfail.2019.10.002 YJCAF 4428
To appear in:
Journal of Cardiac Failure
Received date: Revised date: Accepted date:
1 May 2019 26 August 2019 7 October 2019
Please cite this article as: Isac C. Thomas MD, MPH , Marin Nishimura MD , Janet Ma MD , Stephen D. Dickson MD , Laith Alshawabkeh MD, MSCI , Eric Adler MD , Alan Maisel MD , Michael H. Criqui MD, MPH , Barry Greenberg MD , Clinical Characteristics and Outcomes of Patients with Heart Failure and Methamphetamine Abuse, Journal of Cardiac Failure (2019), doi: https://doi.org/10.1016/j.cardfail.2019.10.002
This is a PDF file of an article that has undergone enhancements after acceptance, such as the addition of a cover page and metadata, and formatting for readability, but it is not yet the definitive version of record. This version will undergo additional copyediting, typesetting and review before it is published in its final form, but we are providing this version to give early visibility of the article. 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. © 2019 Published by Elsevier Inc.
Highlights
Methamphetamine abuse is associated with the development of the heart failure syndrome. Methamphetamine abuse is a global epidemic that has resulted in an increasing number of patients with MethHF.
Rates of readmission for heart failure and total mortality observed in this young, predominantly male patient population were poor, and comparable to older patients with heart failure due to other causes.
Co-morbid substance abuse and psychiatric disorders may have contributed to these poor outcomes.
Efforts directed toward substance abuse prevention paired with comprehensive clinical efforts directed toward the treatment of MethHF and associated comorbidities may improve outcomes.
1
Title: Clinical Characteristics and Outcomes of Patients with Heart Failure and Methamphetamine Abuse Short Title: Methamphetamine and Heart Failure Authors: Isac C. Thomas MD, MPH* Marin Nishimura MD* Janet Ma MD* Stephen D. Dickson MD* Laith Alshawabkeh MD, MSCI* Eric Adler MD* Alan Maisel MD* Michael H. Criqui MD, MPH*Ɨ Barry Greenberg MD* Affiliations: * Division of Cardiovascular Medicine, Department of Medicine, University of California, San Diego Ɨ Division of Preventive Medicine, Department of Family Medicine and Public Health, University of California, San Diego Address for Correspondence: Isac C. Thomas, MD, MPH Department of Medicine Division of Cardiovascular Medicine University of California, San Diego 9452 Medical Center Drive La Jolla, CA 92037-7411 Tel: 858-246-2989, Fax: 858-246-2958 Email: [email protected] Disclosures: All of the authors report no conflicts of interest related to this manuscript. Source of Funding: The Project described was partially supported by the National Institutes of Health, Grant UL1TR001442 of CTSA. The content is solely the responsibility of the authors and does not necessarily represent the official views of the NIH.
2
Abstract Background: Despite a global epidemic of methamphetamine abuse, methamphetamine-associated heart failure (MethHF) remains poorly understood. We sought to evaluate characteristics and outcomes for patients with MethHF. Methods: We reviewed the electronic health records of the University of California, San Diego from 2005-16. We compared characteristics and outcomes between 896 patients with MethHF and 20,576 patients with heart failure (HF) identified using diagnosis codes, urine toxicology, and natriuretic peptides. Results: Compared to HF, patients with MethHF were younger (50±10 v. 67±16 years), predominantly male (72% v. 54%), and had more psychiatric and substance use co-morbidities, including mood/anxiety disorders (29% v. 16%) and opioid use (44% v. 7%). MethHF had a higher 5-year HF readmission rate [64±4% v. 45±1%; hazard ratio (HR) 1.53, p<0.001] and a lower 10-year total mortality rate (25±3% v. 28±1%; HR 0.85, p=0.09). Predictors of poor outcomes included mood/anxiety disorders (HF readmission HR 1.41, p=0.04) and opioid abuse (mortality HR 1.52, p=0.04). Conclusions: Patients with MethHF are frequently encumbered by psychiatric and substance abuse comorbidities, and carry a substantial risk of heart failure readmission and mortality. Comprehensive efforts are needed to stem this emerging epidemic.
Key Words: Heart failure, methamphetamine, substance abuse, cardiomyopathy, readmission, mortality
1
Introduction Methamphetamine and other amphetamine-type stimulant drugs rank among the most commonly abused illicit substances worldwide, second only to marijuana in prevalence of use and opioids in associated harm.1 While its abuse is widespread across the United States, in Western and Southeastern states methamphetamine has been identified as the single greatest drug threat by the Drug Enforcement Agency.2 In San Diego, California, for instance, methamphetamine is the most common primary drug of abuse among patients admitted to drug treatment programs, and rates of emergency department visits associated with methamphetamine abuse have been rising precipitously.3, 4
Methamphetamine abuse is associated with myocardial injury, leading to marked reductions in left ventricular systolic function and a non-ischemic, dilated cardiomyopathy.5, 6 Methamphetamine abuse has also been associated with pulmonary hypertension in the absence of left ventricular dysfunction, and can cause acute elevations in blood pressure and heart rate.7, 8 With a growing global epidemic of methamphetamine abuse, a commensurate increase in methamphetamine-associated heart failure (MethHF) is expected. Yet characteristics of patients with MethHF are poorly described in the medical literature, hindering efforts toward delivering effective care to this population. In order to better characterize and understand this disease, we evaluated patients in a region with a high prevalence of methamphetamine abuse and present the clinical characteristics and outcomes of patients with MethHF.
Methods Study Sample The Institutional Review Board of the University of California, San Diego (UCSD) approved the study protocol. We retrospectively evaluated the electronic health record from the UCSD healthcare system. We abstracted International Classification of Diseases, 9th Revision (ICD-9) diagnosis codes as
2
well as results of urine drug toxicology and serum natriuretic peptides to identify patients with methamphetamine abuse, heart failure, or MethHF. We defined methamphetamine abuse as the presence of the ICD-9 code for amphetamine dependence (304.4) or amphetamine abuse (305.7) (ICD-9 codes specific for methamphetamine are not available), and/or a urine drug toxicology screen positive for the amphetamines drug class and/or methamphetamine in particular (methamphetamine-specific toxicology reporting became available in 2013). We defined heart failure as the presence of ICD-9 code for heart failure (428) and/or an elevated serum brain natriuretic peptide (BNP) (>99 pg/ml) or Nterminal proBNP (>449 pg/ml). Although BNP and/or pro-BNP elevations can be nonspecific for heart failure, we included them in the selection criteria to minimize potential misclassification that may occur due to the under-diagnosis of heart failure particularly among patients who abuse methamphetamine. Using the above selection criteria, we identified 25,942 unique patients who were divided into three mutually-exclusive cohorts: 4,470 patients with methamphetamine abuse and no evidence of heart failure (methamphetamine abuse cohort), 20,576 patients with heart failure and no evidence of methamphetamine abuse (heart failure cohort), and 896 with both methamphetamine abuse and heart failure (MethHF cohort). For patients in the MethHF cohort, if the methamphetamine abuse diagnosis preceded the heart failure diagnosis or vice versa, study entry was at the time of the second diagnosis being fulfilled, with baseline characteristics gathered then and follow up proceeding thereafter. We included patients in the MethHF cohort whether the diagnoses of methamphetamine abuse and heart failure occurred simultaneously or at different points over the study period in an effort to minimize misclassification of these patients into the methamphetamine abuse cohort or heart failure cohort. ICD-9 codes were abstracted from both inpatient and outpatient encounters. We abstracted data between the dates of January 1st, 2005 and June 30th, 2016, prior to the transition to ICD-10 coding which occurred in the latter half of 2016. We also assessed the accuracy of ICD-9 coding for heart failure and methamphetamine abuse in addition to the accuracy of the BNP/pro-BNP criteria for heart failure
3
by manually reviewing of the electronic health record for three non-overlapping randomly selected samples: 50 patients with an ICD-9 diagnosis of heart failure, 50 patients with elevations in proBNP (>449 pg/ml) or BNP (>99 pg/ml), and 50 patients with an ICD diagnosis of amphetamine abuse or dependence. For purposes of this manual chart review, we defined clinical heart failure as one or more of the following: inpatient admission for heart failure with intravenous diuretics administered, echocardiogram with LVEF < 40%, or an outpatient encounter with a cardiologist documenting a diagnosis and treatment plan for heart failure. We defined clinical methamphetamine abuse as documentation of methamphetamine abuse or dependence in provider notes.
Clinical Characteristics and Outcomes Clinical characteristics were abstracted from the electronic health record through a combination of ICD-9 codes and echocardiogram data. We evaluated ICD-9 codes relating to several common cardiovascular, psychiatric, and substance abuse co-morbidities. A full list of ICD-9 codes is included in Supplemental Table A. Substance use/abuse was defined by the presence of a positive urine drug toxicology screen or an ICD-9 code for the substance of interest. Transthoracic echocardiogram data were available for 623 patients (14%) in the methamphetamine abuse cohort, 14,443 patients (70%) in the HF cohort, and 731 patients (82%) in the MethHF cohort. When multiple echocardiograms were available for a given patient, the lowest left ventricular ejection fraction (LVEF) and highest pulmonary artery systolic pressure (PASP) were recorded. We also abstracted the date and total number of hospital encounters and mortality over the follow-up period for each patient. The follow-up period extended from the time of first diagnosis (of methamphetamine abuse, heart failure, or MethHF) to patient death or June 30th, 2016. Hospital encounters were defined as any emergency department visit, observation admission, or inpatient admission. These encounters were further characterized by whether heart failure was the primary diagnosis for the encounter. We defined heart failure readmission as the first hospital admission
4
(observation or inpatient) with a primary diagnosis of heart failure that occurred after an index admission with a primary diagnosis of heart failure. We defined mortality as death due to any cause, with vital status abstracted from the electronic health record.
Statistical Analysis We plotted the biennial tally of unique patients diagnosed with methamphetamine abuse and MethHF starting in 2005. We compared characteristics of the methamphetamine abuse, heart failure, and MethHF cohorts using z-tests of proportions with a Bonferroni correction for categorical variables and analysis of variance (ANOVA) with a Bonferroni correction for continuous variables. In cross-sectional analyses, we combined the methamphetamine abuse and MethHF cohorts and constructed a multivariable logistic regression model to identify factors associated with the presence of a heart failure diagnosis among patients who abuse methamphetamine. We also combined the heart failure and MethHF cohorts and constructed a multivariable logistic regression model to identify factors associated with methamphetamine abuse among patients with a diagnosis of heart failure. For both models, we included multiple potential exposure variables in a saturated model based on observed differences in baseline patient characteristics between the cohorts and characteristics that we hypothesized may differentiate the cohorts. We then utilized backward selection with a p-value of 0.10 for exclusion to create final models. We present adjusted odds ratios with 95% confidence intervals (CI) for all variables in the final models. In order to evaluate the association of methamphetamine abuse with adverse outcomes in patients with heart failure, we performed longitudinal analyses comparing the heart failure and MethHF cohort. We present Kaplan Meier curves evaluating the cumulative incidence of heart failure readmission as well as total mortality, comparing the heart failure and MethHF cohorts using log-rank testing. We constructed a Cox regression model for time to death and present the hazard ratio (HR) for methamphetamine abuse among patients with heart failure adjusted for demographics (age, sex,
5
race/ethnicity, and health insurance status). As death can be a competing risk for heart failure readmission, we performed a competing risk regression according to the method described by Fine and Gray,9 and report the demographic-adjusted HR for methamphetamine abuse. We also tallied the total number of hospital encounters over follow up (emergency department visits, and observation and inpatient admissions) and plotted the results, with comparisons between strata performed using posthoc z-tests of proportions with Bonferroni corrections. Finally, we constructed Cox regression models to identify independent predictors of heart failure readmission and total mortality among patients in the MethHF cohort, adjusting for demographics, LVEF, PASP, and co-morbid diagnoses, and sequentially excluded terms using backward selection with a p-value of 0.10 to produce final models. We present adjusted hazard ratios with 95% confidence intervals (CI) for all variables in the final models. As BNP and proBNP elevations can be non-specific for heart failure, we performed two sensitivity analyses to investigate potential bias that may have resulted from their use in the selection criteria. In the first analysis, we doubled the threshold for elevation for BNP (>99 to >199 pg/ml) and proBNP (>449 to >899 pg/ml). In the second analysis, we removed BNP and proBNP from the selection criteria for heart failure. We performed statistical analyses using R Statistical Software (version 3.2.2; R Foundation for Statistical Computing, Vienna, Austria) and SPSS Statistics version 22.0 (IBM Corporation, Armonk, NY, USA). We considered a p-value of <0.05 statistically significant, with Bonferroni-correction when multiple testing was performed.
6
2060
Number of Unique Patients
Methamphetamine Abuse MethHF 1499
446 62 92 '05-'06
178
121
'07-'08
225
165
114 '09-'10
'11-'12
212
'13-'14
192
'15-June '16
Figure 1: Trends in the diagnosis of methamphetamine abuse and MethHF. Biennial tally of unique patients diagnosed with methamphetamine abuse alone or methamphetamine abuse with heart failure (MethHF). The methamphetamine abuse cohort and the MethHF cohort are mutually-exclusive.
Results Trends in Methamphetamine Abuse and MethHF The incidence of methamphetamine abuse and MethHF increased substantially over the study period. As shown in Figure 1, 62 patients were diagnosed with methamphetamine abuse and 92 were diagnosed with MethHF in 2005-06; in 2015-16, 2,060 patients were diagnosed with methamphetamine abuse and 192 were diagnosed with MethHF. Pronounced increases in methamphetamine abuse occurred during the most recent four years of the study period. In manual review of medical records, a diagnosis of heart failure via ICD-9 code corresponded with documented heart failure in 82% of charts reviewed. A diagnosis of heart failure via BNP/proBNP elevation corresponded with documented heart failure in 90% of charts reviewed. A diagnosis of
7
methamphetamine abuse via ICD-9 code and/or urine toxicology corresponded with documented methamphetamine abuse in 84% of charts reviewed.
Clinical Characteristics Characteristics of patients in the methamphetamine abuse, heart failure, and MethHF cohorts are presented in Table 1. Patients in the MethHF cohort had a mean age of 50 ±10 years, were predominantly male (72%), majority white (57%), and frequently uninsured (52%). Co-morbid diagnoses such as hypertension (54%), ischemic heart disease (21%), mood/anxiety disorders (29%), and opioid use (44%) were common. In contrast, patients in the methamphetamine abuse cohort were younger (40 ±13 years) and generally with fewer co-morbid medical diagnoses but comparable levels of psychiatric and substance use diagnoses compared to the MethHF cohort. Patients in the heart failure cohort were, on average, 17 years older (67 ±16 years) than the MethHF cohort and with a lower percentage of men (54%) and African American race (9% v. 19%), similar or higher rates of co-morbid medical diagnoses, and lower rates of psychiatric and substance use diagnoses.
8
Demographics Age, years (standard deviation) Male Sex Race/Ethnicity White Hispanic/Latino African American Asian Other/Mixed No medical insurance Medical and Psychiatric Comorbidities Hypertension Diabetes Mellitus Ischemic Heart Disease Atrial Fibrillation/Flutter Chronic Kidney Disease Stroke Endocarditis Cardiogenic Shock Cardiac Arrest Ventricular Tachycardia/Fibrillation Mood/Anxiety Disorders Schizophrenia/Delusional Disorder Post-Traumatic Stress Disorder Substance Use Opioids Marijuana Alcohol Cocaine Phencyclidine (PCP)
Methamphetamine Abuse (n=4,470)
Heart Failure (n=20,576)
MethHF (n=896)
40.0 (12.7)* 71.6%
67.1 (15.7)* 54.4%*
49.6 (10.0) 72.4%
56%* 22.2%* 13.3%* 2.4%* 5.9% 28%*
59.3% 17.4% 9.4%* 6.7%* 6.8%* 24.4%*
57.7% 15.5% 18.5% 3.9% 4.2% 51.7%
16.7%* 5.8%* 2.1%* 0.4%* 1.7%* 2.3%* 1.1%* 0.0%* 0.4%* 0.1%* 36.0%* 12.1%* 3.2%
64.1%* 23.7% 17.7% 27.4%* 20.4%* 8.4% 1.8%* 1.5% 1.5% 0.8%* 16.8%* 1.4%* 0.5%*
54.4% 23.1% 21.4% 16.0% 15.5% 9.7% 3.1% 2.6% 1.8% 1.4% 28.6% 8.1% 2.1%
24.2%* 17.3% 20.8%* 3.0% 0.6%*
7.1%* 1.4%* 3.9%* 0.3%* 0.1%*
44.2% 20.0% 16.2% 3.9% 1.6%
Table 1: Characteristics of patients in the methamphetamine abuse, heart failure, and MethHF cohorts. Each cohort is mutually exclusive of the others. *indicates a Bonferroni corrected p-value (p< 0.05/3) for comparisons to MethHF.
9
Figure 2: Left ventricular ejection fraction and pulmonary artery systolic pressure for patients in the methamphetamine abuse, heart failure, and MethHF cohorts
Left Ventricular Ejection Fraction 3%
12%
33%
14%
18%
18% 17%
Severely Elevated (60+ mm Hg)
16%
Moderately Elevated 13%
Moderately Reduced (30-39%) Mildly Reduced (4049%)
5%
12% 11%
Severely Reduced (<30%)
Pulmonary Artery Systolic Pressure
(45-59 mm Hg)
11% Mildly Elevated (30-
95%
44 mm Hg)
64%
Normal (50%+)
Heart Failure (n=14,443)
33%
33%
33%
Heart Failure (n=14,443)
MethHF (n=731)
75%
Normal (<30mm Hg)
44%
Methamphetamine Abuse (n=623)
36%
MethHF (n=731)
Methamphetamine Abuse (n=623)
1
Figure 2 shows the distribution of left ventricular ejection fraction and pulmonary artery systolic pressure estimated by transthoracic echocardiogram for the three cohorts. Patients in the MethHF cohort had a higher proportion with LVEF<50% (56% v. 36%) and a higher proportion with an LVEF<30% (33% v. 12%) compared with patients in the heart failure cohort. Pulmonary artery systolic pressure (PASP) levels were elevated in 67% of the MethHF cohort, with similar percentages in the heart failure cohort. 25% of patients in the methamphetamine abuse cohort with available echocardiograms had evidence of elevated PASP.
Factors Associated with Heart Failure in Patients with Methamphetamine Abuse Table 2 lists adjusted odds ratios for variables associated with the presence of heart failure among patients diagnosed with methamphetamine abuse. The odds of heart failure were 25% higher with each 5-year increase in age (OR 1.25, 95% CI 1.20-1.31), and were nearly 2.5-fold higher for patients lacking medical insurance (OR 2.49, 95% CI 2.06-3.02). Compared to white patients, patients of African American (OR 1.48, 95% CI 1.14-1.92) and Asian/Pacific Islander (OR 2.22, 95% CI 1.33-3.71) race/ethnicity had higher odds of heart failure. Atrial fibrillation, ischemic heart disease, and chronic kidney disease were among several co-morbid medical diagnoses associated with heart failure, whereas mood/anxiety disorders and schizophrenia were inversely associated with heart failure. Opioid use was associated with heart failure (OR 2.12, 95% CI 1.74-2.58), while alcohol abuse was inversely associated with HF (OR 0.74, 95% CI 0.58-0.94).
1
Adjusted Odds Ratio (95% confidence interval) Older Age (per 5 year increase)
1.25 (1.20, 1.31)
Male Sex
1.37 (1.11, 1.68)
Race/Ethnicity White
Reference
Hispanic/Latino
1.06 (0.83, 1.37)
African American
1.48 (1.14, 1.92)
Asian/Pacific Islander
2.22 (1.33, 3.71)
Other/Mixed
1.04 (0.66, 1.64)
Atrial Fibrillation/Flutter
27.40 (15.80, 47.50)
Ischemic Heart Disease
6.51 (4.75, 8.94)
Chronic Kidney Disease
3.57 (2.46, 5.17)
No Medical Insurance
2.49 (2.06, 3.02)
Hypertension
2.38 (1.93, 2.94)
Opioid Use
2.12 (1.74, 2.58)
Diabetes Mellitus
1.57 (1.20, 2.06)
Cerebrovascular Accident
1.45 (0.99, 2.12)
Alcohol Abuse
0.74 (0.58, 0.94)
Schizophrenia/Delusional Disorder
0.66 (0.48, 0.90)
Mood/Anxiety Disorder
0.63 (0.51, 0.77)
Table 2: Factors associated with heart failure among patients diagnosed with methamphetamine abuse. Saturated model includes the above variables and endocarditis, post-traumatic stress disorder, cocaine, phencyclidine, and marijuana use. The variables listed above were selected for the final multivariable model using backward selection with a p-value for exclusion of 0.10.
Factors Associated with Methamphetamine Abuse in Patients with Heart Failure Table 3 displays adjusted odds ratios for variables associated with the diagnosis of methamphetamine abuse among patients with heart failure. Each 5-year decrement in age was associated with a 32% higher odds of methamphetamine abuse (OR 1.32, 95% CI 1.28-1.36). Male sex (OR 1.57, 95% CI 1.30-1.91) and white race where also associated with a higher odds of methamphetamine abuse. Schizophrenia and all of the substances abuse disorders were associated with
2
higher odds of methamphetamine abuse. Among patients with heart failure, each 5% decrement in LVEF (OR 1.13, 1.10-1.16) and 10 mm Hg increase in PASP (OR 1.05, 1.01-1.08) were associated with a higher likelihood of methamphetamine abuse.
Adjusted Odds Ratio (95% confidence interval) Younger Age (per 5 year decrease)
1.32 (1.28, 1.36)
Lower LVEF (per 5%)
1.13 (1.10, 1.16)
Higher PASP (per 10 mm Hg)
1.05 (1.01, 1.08)
Male Sex 1.57 (1.30, 1.91) Race/Ethnicity White Reference Hispanic/Latino 0.56 (0.43, 0.72) African American 0.72 (0.56, 0.93) Asian/Pacific Islander 0.67 (0.44, 1.01) Other/Mixed 0.74 (0.48, 1.14) Phencyclidine Abuse 4.25 (1.79, 10.12) Opioid Abuse 4.19 (3.46, 5.09) Cocaine Abuse 4.03 (2.31, 7.03) Marijuana Abuse 3.33 (2.52, 4.42) Schizophrenia/Delusional Disorder 2.45 (1.68, 3.58) Alcohol Abuse 1.59 (1.22, 2.07) Chronic Kidney Disease 0.59 (0.47, 0.73) Atrial Fibrillation/Flutter 0.58 (0.46, 0.72) Table 3: Factors associated with methamphetamine abuse among patients diagnosed with heart failure. Saturated model includes the above variables and medical insurance, hypertension, diabetes mellitus, ischemic heart disease, cerebrovascular accident, endocarditis, mood/anxiety disorders, and post-traumatic stress disorder. The variables listed above comprise the final model.
3
60%
p<.01
Heart Failure MethHF
Percentage of Patients
50%
40% p>.01
30% p<.01
p<.01
20% p<.01
10%
0%
0 or 1
2 to 5 6 to 10 11 to 15 No. of Hospital Encounters
16+
Figure 3: Distribution of hospital encounters for patients in the heart failure and MethHF cohorts. Comparisons between the heart failure and MethHF cohorts were made at a Bonferroni-corrected significance level of 0.01.
Clinical Outcomes and Predictors of Readmission and Mortality in MethHF Figure 3 displays the distribution of emergency department, observation, and inpatient hospital encounters for patients in the heart failure and MethHF cohorts. Median follow up was 4.82 [interquartile range 1.98 to 8.36] years for the heart failure cohort and 3.95 [1.71 to 7.30] years for the MethHF cohort. Over follow up, 75,589 hospital encounters occurred between these two cohorts with a median of 1 [0 to 4] encounters for the heart failure cohort and 4 [1 to 11] encounters for the MethHF cohort (p<0.001).
4
Figure 4: Cumulative incidence of heart failure readmission and total mortality for the heart failure and MethHF cohorts. Hazard ratios are adjusted for age, sex, race/ethnicity, and health insurance status.
Over follow-up, 2,857 heart failure readmissions and 4,145 deaths occurred. Figure 4 displays cumulative incidence curves for heart failure readmission and total mortality for the heart failure and MethHF cohorts. Patients in the MethHF cohort had a higher risk of heart failure readmission (log-rank p <0.001) with a 5-year rate of 64 ±4% v. 45 ±1% in the heart failure cohort. Among patients with heart failure, methamphetamine abuse was associated with a HR of 1.53 (95% CI 1.31-1.78, p<0.001) for HF readmission after adjustment for demographics and the competing risk of death. Patients in the MethHF cohort had a lower risk of total mortality (log-rank p<0.001) with a 10-year mortality of 25 ±3% v. 28 ±1%. Among patient with heart failure, methamphetamine abuse was associated with a HR of 0.85 (95% CI 0.71-1.03, p=0.09) for total mortality after adjustment for demographics.
5
Adjusted Hazard Ratio (95% Confidence Interval) Heart Failure Readmission Lower LVEF (per 5%)
1.15 (1.10, 1.20)
Higher PA pressure (per 10 mm Hg)
1.07 (1.03, 1.11)
Stroke
1.49 (1.01, 2.22)
Mood/Anxiety Disorders
1.41 (1.01, 1.96)
Chronic Kidney Disease
1.36 (0.98, 1.88)
Total Mortality Older Age (per 5 years)
1.18 (1.07, 1.30)
African American (v. White)
2.09 (1.36, 3.20)
Opioid Abuse
1.52 (1.07, 2.27)
Mood/Anxiety Disorders
1.47 (0.99, 2.19)
Hypertension
0.62 (0.42, 0.93)
Table 4: Predictors of heart failure readmission and total mortality for patients in the MethHF cohort. The saturated models for each outcome include age, sex, race/ethnicity, medical insurance, hypertension, diabetes, ischemic heart disease, endocarditis, atrial fibrillation, cerebrovascular accident, chronic kidney disease, mood/anxiety disorder, schizophrenia, post-traumatic stress disorder, opioid use, alcohol abuse, marijuana abuse, cocaine abuse, phencyclidine abuse, left ventricular ejection fraction, and pulmonary artery systolic pressure. The variables listed above comprise the final models for each outcome.
Table 4 displays adjusted hazard ratios for predictors of heart failure readmission and total mortality among patients in the MethHF cohort. Reductions in LVEF (HR 1.15, 95% CI 1.10-1.20 per 5% decrease), elevations in PASP (HR 1.07, 95% CI 1.03-1.11 per 10 mm Hg increase), stroke (HR 1.49, 95% CI 1.01, 2.22), and mood/anxiety disorders (HR 1.41, 95% CI 1.01-1.96) were all independently increased the risk of a HF readmission. While LVEF and PASP were not independently associated with mortality in this model, older age (HR 1.18, 95% CI 1.07-1.30 per 5 years), African American race (HR 2.09, 95% CI 1.36-3.20 v. white race), and opioid use (HR 1.52, 95% CI 1.07-2.27) were associated with a higher risk of mortality.
6
In sensitivity analyses, doubling the threshold for BNP/ proBNP elevation in the heart failure selection criteria did not materially change the results for heart failure readmission and total mortality in the heart failure and MethHF cohorts. Removing BNP/proBNP from the heart failure selection criteria resulted in a reduction in the sample size for both cohorts, but the comparisons between the cohorts for heart failure readmission and total mortality were maintained. Detailed results are presented in Supplemental Table B.
Discussion Drug abuse has increased with the rise in access to illicit substances and physician-prescribed medications with addiction potential.3 Although widespread opioid abuse has resulted in significant harm in recent years,5 the abuse of methamphetamine and other amphetamine-type substances is regarded as the most significant threat to public health in much of the United States and in many countries across the world.2 Methamphetamine abuse has a long-documented association with cardiomyopathy and heart failure via mechanisms that have not been fully elucidated,10-12 but may be similar to those found in cardiomyopathies associated with cocaine and other high catecholamine conditions.13, 14 Chronic abuse appears to lead to myocardial inflammation and fibrosis, coupled with marked reductions in LV systolic function that may be reversible with methamphetamine cessation in some cases.5, 15 Despite this potential for recovery and the relatively young ages of these patients, MethHF is associated with poor heart failure-related clinical outcomes.5, 8 In this study, we found that patients in the MethHF cohort were younger than those in the HF cohort, yet had relatively high rates of several medical co-morbidities including hypertension, diabetes, ischemic heart disease, and chronic kidney disease. These findings suggest that methamphetamine abuse may not be the sole cause of heart failure for a number of patients with MethHF, particularly in areas where methamphetamine abuse is highly prevalent. Rather, methamphetamine abuse may be a
7
contributor to heart failure, particularly among patients at heightened risk due to the presence of traditional heart failure risk factors. Patients in the MethHF cohort also had relatively high rates of the psychiatric and substance abuse disorders that were commonly found in the methamphetamine abuse cohort, highlighting the disproportionate burden of a range of co-morbidities among patients with MethHF. We also found that clinical outcomes for patients in the MethHF cohort were commensurate with or worse than those of patients in the heart failure cohort. Despite their mean age being 17 years lower than the heart failure cohort, patients in the MethHF cohort had only a slightly lower risk of total mortality, and as expected this lower risk was not significant with age adjustment. Perhaps owing to underlying medical, psychiatric, and substance abuse co-morbidities, patients with MethHF had a substantially higher rate of hospital encounters for any cause over the follow up period, and a higher risk of heart failure readmission. We found that while traditional risk factors predict heart failure readmission and mortality in the MethHF cohort, opioid use and mood/anxiety disorders also independently contribute to heightened risk. Opioid use was highly prevalent in the MethHF cohort and, in line with the national opioid abuse epidemic, a source of significant harm. Psychiatric disorders such as depression are associated with poor clinical outcomes among patients with cardiovascular disease including heart failure.16 Clinical trials utilizing selective serotonin reuptake inhibitors to reduce the risk of cardiovascular outcomes in patients with mood disorders and cardiovascular disease have shown mixed results.17, 18 Nevertheless, an emphasis on the detection and treatment of psychiatric comorbidities and concomitant substance abuse is likely necessary to improve outcomes among patients with MethHF. Effective prevention of MethHF will likely center on efforts toward early methamphetamine abuse detection and cessation, with a heightened emphasis on patients at greatest risk of developing heart failure. In patients who abuse methamphetamine, we found that heart failure was strongly
8
associated with older age, African American and Asian/Pacific Islander races, and a lack of medical insurance. While relatively uncommon among these patients, factors such as atrial fibrillation, chronic kidney disease, and ischemic heart disease were strongly associated with heart failure. This latter finding highlights the fact that although MethHF is understood to be due to a dilated cardiomyopathy secondary to a toxic exposure, the contributions of traditional heart failure risk factors should be considered, particularly in older patients. Surprisingly, we found alcohol abuse, mood/anxiety disorders, and schizophrenia to be inversely associated with heart failure among patients who abuse methamphetamine. Although alcohol itself has a known toxic effect on the myocardium, concomitant alcohol abuse may lessen the intake of methamphetamine, thereby exerting a “protective” effect. Alternatively, these co-morbidities may be frequently diagnosed among patients with methamphetamine abuse and under-diagnosed among patients with heart failure, leading to a spurious inverse association. As substance abuse and psychiatric disorders affect treatment strategies and outcomes among patients with MethHF, providers should screen for these disorders and treat accordingly. Substance abuse is often hidden by patients and difficult for providers to diagnose.19 In patients with a diagnosis of heart failure, we found that a reduction in LVEF, as well as younger age, male sex, white race, concomitant substance abuse, and schizophrenia were strongly correlated with methamphetamine abuse. As methamphetamine and its metabolites are only detectable in the urine for approximately 48 hours after exposure,20 urine drug screens alone may not be sufficient to detect methamphetamine abuse, and repeated inquiry and urine testing of patients at high risk of methamphetamine abuse are often necessary. Our findings add to the relatively limited literature on patients with MethHF. In a recent study, Schürer et al. characterized tissue obtained from endomyocardial biopsies on 30 patients with MethHF and found that the duration of methamphetamine abuse correlated with the extent of tissue fibrosis,
9
and that the extent of tissue fibrosis predicted the likelihood of recovery of LV function after methamphetamine cessation.5 These findings demonstrate the importance of early detection and intervention among patients with MethHF. Zhao et al. described clinical characteristics for patients with MethHF as well as patients with methamphetamine associated pulmonary arterial hypertension, and showed that the mortality risk for both groups was high.8 We found that PASP was elevated in 25% of patients who abuse methamphetamine without a diagnosis of HF, as possibly an early manifestation of the effect of methamphetamine on the cardiovascular system. Pulmonary arterial hypertension in the absence of LV systolic dysfunction is an increasingly recognized manifestation of methamphetamine abuse, and the development of cardiomyopathy versus pulmonary arterial hypertension may be linked to the mode of consumption (i.e. snorting, smoking, ingestion or intravenous injection).21
Limitations Our study evaluates a large cohort of patients with MethHF and identifies important predictors of adverse clinical outcomes in these patients. Our study also has limitations. Since our study relies on the diagnosis of methamphetamine abuse via ICD-9 coding and screening with urine toxicology, we have potentially underestimated the true prevalence of methamphetamine abuse in our health care system, and given rising trends in methamphetamine abuse, many patients classified in the heart failure cohort may have also been abusing methamphetamines. Moreover, heart failure may not have been investigated or diagnosed in many young patients in the methamphetamine abuse cohort. Rising rates of methamphetamine abuse that we observed in this study may be in part due to increased awareness and diagnosis of methamphetamine abuse, rather than simply due to increased abuse of methamphetamines. However, these trends have been paralleled by increasing arrests for methamphetamine sales and possession reported by local law enforcement.4 As outcomes were obtained from the UCSD electronic health record, individuals who abuse methamphetamine who do not seek medical attention are not represented, hospitalizations that occurred outside of our health care
10
system were not captured, and deaths that occurred outside of our health care system were likely under-reported. ICD-9 codes and urine toxicology for the majority of the study period were for amphetamines in general (not methamphetamine in particular), and thus we could not exclude the use or abuse of other amphetamine–type stimulants, such as MDMA (ecstasy), or amphetamine containing prescription drugs. Finally, our study population is from a region with a high prevalence of methamphetamine abuse, and thus these results may not be generalizable.
Conclusions As methamphetamine abuse grows increasingly common, health care providers are likely to encounter an epidemic of MethHF in the coming years. Increasing amounts and duration of exposure to methamphetamine are associated with the development and severity of heart failure,13 suggesting that early intervention can reduce the burden of this disease. Innovative therapies directed at treating methamphetamine addiction are needed and an understanding of the underlying mechanisms of MethHF development are necessary in order to develop novel therapeutic interventions. Given the complex underpinnings of MethHF, comprehensive multidisciplinary efforts are needed to stem this emerging epidemic.
11
References 1.
United Nations Office on Drugs and Crime. World Drug Report 2017. ISBN: 978-92-1-148291-1,
eISBN: 978-92-1-060623-3, United Nations publication, Sales No. E.17.XI.6. 2.
United States Department of Justice Drug Enforcement Administration. 2017 National Drug
Threat Assessment. DEA Strategic Intelligence Section. October 2017. 3.
National Institute on Drug Abuse. Nationwide Trends. https://www.drugabuse.gov/
publications/drugfacts/nationwide-trends. Accessed July 30, 2018. 4.
Methamphetamine Strike Force. Local Statistics. San Diego County Health and Human Services
Administration, Behavioral Health Services. https://www.no2meth.org/local-statistics. Accessed November 11, 2018. 5.
Schurer S, Klingel K, Sandri M, Majunke N, Besler C, Kandolf R, et al. Clinical Characteristics,
Histopathological Features, and Clinical Outcome of Methamphetamine-Associated Cardiomyopathy. JACC Heart failure. 2017;5:435-445. 6.
Neeki MM, Kulczycki M, Toy J, Dong F, Lee C, Borger R, et al. Frequency of Methamphetamine
Use as a Major Contributor Toward the Severity of Cardiomyopathy in Adults ≤50 Years. American Journal of Cardiology. 2016;118:585-589. 7.
Frishman WH, Del Vecchio A, Sanal S and Ismail A. Cardiovascular manifestations of substance
abuse: part 2: alcohol, amphetamines, heroin, cannabis, and caffeine. Heart disease (Hagerstown, Md). 2003;5:253-71. 8.
Zhao SX, Kwong C, Swaminathan A, Gohil A and Crawford MH. Clinical Characteristics and
Outcome of Methamphetamine-Associated Pulmonary Arterial Hypertension and Dilated Cardiomyopathy. JACC: Heart Failure. 2018;6:209-218. 9.
Fine JP and Gray RJ. A Proportional Hazards Model for the Subdistribution of a Competing Risk.
Journal of the American Statistical Association. 1999;94:496-509.
12
10.
Smith HJ, Roche AH, Jausch MF and Herdson PB. Cardiomyopathy associated with amphetamine
administration. American heart journal. 1976;91:792-7. 11.
Hong R, Matsuyama E and Nur K. Cardiomyopathy associated with the smoking of crystal
methamphetamine. Jama. 1991;265:1152-4. 12.
Islam MN, Kuroki H, Hongcheng B, Ogura Y, Kawaguchi N, Onishi S, et al. Cardiac lesions and
their reversibility after long term administration of methamphetamine. Forensic science international. 1995;75:29-43. 13.
Sliman S, Waalen J and Shaw D. Methamphetamine-Associated Congestive Heart Failure:
Increasing Prevalence and Relationship of Clinical Outcomes to Continued Use or Abstinence. Cardiovascular toxicology. 2015. 14.
Won S, Hong RA, Shohet RV, Seto TB and Parikh NI. Methamphetamine-Associated
Cardiomyopathy. Clinical cardiology. 2013;36:737-742. 15.
Voskoboinik A, Ihle JF, Bloom JE and Kaye DM. Methamphetamine-associated cardiomyopathy:
patterns and predictors of recovery. Internal medicine journal. 2016;46:723-7. 16.
Rutledge T, Reis VA, Linke SE, Greenberg BH and Mills PJ. Depression in Heart Failure: A Meta-
Analytic Review of Prevalence, Intervention Effects, and Associations With Clinical Outcomes. Journal of the American College of Cardiology. 2006;48:1527-1537. 17.
Kim JM, Stewart R, Lee YS, Lee HJ, Kim MC, Kim JW, et al. Effect of Escitalopram vs Placebo
Treatment for Depression on Long-term Cardiac Outcomes in Patients With Acute Coronary Syndrome: A Randomized Clinical Trial. Jama. 2018;320:350-358. 18.
O'Connor CM, Jiang W, Kuchibhatla M, Silva SG, Cuffe MS, Callwood DD, et al. Safety and
Efficacy of Sertraline for Depression in Patients With Heart Failure: Results of the SADHART-CHF Trial. Journal of the American College of Cardiology. 2010;56:692-699.
13
19.
Diercks DB, Fonarow GC, Kirk JD, Jois-Bilowich P, Hollander JE, Weber JE, et al. Illicit stimulant
use in a United States heart failure population presenting to the emergency department (from the Acute Decompensated Heart Failure National Registry Emergency Module). The American journal of cardiology. 2008;102:1216-9. 20.
Oyler JM, Cone EJ, Joseph RE, Moolchan ET and Huestis MA. Duration of Detectable
Methamphetamine and Amphetamine Excretion in Urine after Controlled Oral Administration of Methamphetamine to Humans. Clinical Chemistry. 2002;48:1703. 21.
Ben-Yehuda O and Siecke N. Crystal Methamphetamine: A Drug and Cardiovascular Epidemic.
JACC Heart failure. 2018;6:219-221.
14