Risk of New Bloodstream Infections and Mortality Among People Who Inject Drugs With Infective Endocarditis | Cardiology | JAMA Network Open | JAMA Network
[Skip to Navigation]
Table 1.  Demographic Characteristics, Site of Endocardial Involvement, and IVDU Variables of Infective Endocarditis Episodes, Stratified by New BSI
Demographic Characteristics, Site of Endocardial Involvement, and IVDU Variables of Infective Endocarditis Episodes, Stratified by New BSI
Table 2.  Microbiological Origin of New BSIs, Stratified by Treatment Location
Microbiological Origin of New BSIs, Stratified by Treatment Location
Table 3.  Hospital Care Variables, Complications, and Outcomes of Infective Endocarditis Episodes, Stratified by New BSI
Hospital Care Variables, Complications, and Outcomes of Infective Endocarditis Episodes, Stratified by New BSI
Table 4.  Multivariable Cox Regression for Factors Associated With New BSIs
Multivariable Cox Regression for Factors Associated With New BSIs
Table 5.  Multivariable Cox Regression for Factors Associated With 90-Day Mortality
Multivariable Cox Regression for Factors Associated With 90-Day Mortality
1.
Weir  MA, Slater  J, Jandoc  R, Koivu  S, Garg  AX, Silverman  M.  The risk of infective endocarditis among people who inject drugs: a retrospective, population-based time series analysis.   CMAJ. 2019;191(4):E93-E99. doi:10.1503/cmaj.180694 PubMedGoogle ScholarCrossref
2.
Hartman  L, Barnes  E, Bachmann  L, Schafer  K, Lovato  J, Files  DC.  Opiate injection-associated infective endocarditis in the southeastern United States.   Am J Med Sci. 2016;352(6):603-608. doi:10.1016/j.amjms.2016.08.010 PubMedGoogle ScholarCrossref
3.
Cooper  HL, Brady  JE, Ciccarone  D, Tempalski  B, Gostnell  K, Friedman  SR.  Nationwide increase in the number of hospitalizations for illicit injection drug use-related infective endocarditis.   Clin Infect Dis. 2007;45(9):1200-1203. doi:10.1086/522176 PubMedGoogle ScholarCrossref
4.
Fleischauer  AT, Ruhl  L, Rhea  S, Barnes  E.  Hospitalizations for endocarditis and associated health care costs among persons with diagnosed drug dependence—North Carolina, 2010-2015.   MMWR Morb Mortal Wkly Rep. 2017;66(22):569-573. doi:10.15585/mmwr.mm6622a1 PubMedGoogle ScholarCrossref
5.
Wurcel  AG, Anderson  JE, Chui  KK,  et al.  Increasing infectious endocarditis admissions among young people who inject drugs.   Open Forum Infect Dis. 2016;3(3):ofw157. doi:10.1093/ofid/ofw157 PubMedGoogle Scholar
6.
Schranz  AJ, Fleischauer  A, Chu  VH, Wu  LT, Rosen  DL.  Trends in drug use–associated infective endocarditis and heart valve surgery, 2007 to 2017: a study of statewide discharge data.   Ann Intern Med. 2019;170(1):31-40. doi:10.7326/M18-2124 PubMedGoogle ScholarCrossref
7.
Iversen  K, Ihlemann  N, Gill  SU,  et al.  Partial oral versus intravenous antibiotic treatment of endocarditis.   N Engl J Med. 2019;380(5):415-424. doi:10.1056/NEJMoa1808312 PubMedGoogle ScholarCrossref
8.
Paladino  JA, Poretz  D.  Outpatient parenteral antimicrobial therapy today.   Clin Infect Dis. 2010;51(suppl 2):S198-S208. doi:10.1086/653520 PubMedGoogle ScholarCrossref
9.
Tice  AD, Rehm  SJ, Dalovisio  JR,  et al; IDSA.  Practice guidelines for outpatient parenteral antimicrobial therapy: IDSA guidelines.   Clin Infect Dis. 2004;38(12):1651-1672. doi:10.1086/420939 PubMedGoogle ScholarCrossref
10.
MacKenzie  M, Rae  N, Nathwani  D.  Outcomes from global adult outpatient parenteral antimicrobial therapy programmes: a review of the last decade.   Int J Antimicrob Agents. 2014;43(1):7-16. doi:10.1016/j.ijantimicag.2013.09.006 PubMedGoogle ScholarCrossref
11.
Pericà S  JM, Llopis  J, González-Ramallo  V,  et al; Spanish Collaboration on Endocarditis-Grupo de Apoyo al Manejo de la Endocarditis Infecciosa en España (GAMES) Investigators.  Outpatient parenteral antibiotic treatment for infective endocarditis: a prospective cohort study from the GAMES cohort.   Clin Infect Dis. 2019;69(10):1690-1700. doi:10.1093/cid/ciz030 PubMedGoogle ScholarCrossref
12.
Seaton  RA, Barr  DA.  Outpatient parenteral antibiotic therapy: principles and practice.   Eur J Intern Med. 2013;24(7):617-623. doi:10.1016/j.ejim.2013.03.014 PubMedGoogle ScholarCrossref
13.
Andrews  MM, von Reyn  CF.  Patient selection criteria and management guidelines for outpatient parenteral antibiotic therapy for native valve infective endocarditis.   Clin Infect Dis. 2001;33(2):203-209. doi:10.1086/321814 PubMedGoogle ScholarCrossref
14.
Norris  AH, Shrestha  NK, Allison  GM,  et al.  2018 Infectious Diseases Society of America clinical practice guideline for the management of outpatient parenteral antimicrobial therapy.   Clin Infect Dis. 2019;68(1):e1-e35. doi:10.1093/cid/ciy745 PubMedGoogle ScholarCrossref
15.
Rodger  L, Glockler-Lauf  SD, Shojaei  E,  et al.  Clinical characteristics and factors associated with mortality in first-episode infective endocarditis among persons who inject drugs.   JAMA Netw Open. 2018;1(7):e185220. doi:10.1001/jamanetworkopen.2018.5220 PubMedGoogle Scholar
16.
Vandenbroucke  JP, von Elm  E, Altman  DG,  et al; STROBE Initiative.  Strengthening the Reporting of Observational Studies in Epidemiology (STROBE): explanation and elaboration.   Int J Surg. 2014;12(12):1500-1524. doi:10.1016/j.ijsu.2014.07.014 PubMedGoogle ScholarCrossref
17.
Li  JS, Sexton  DJ, Mick  N,  et al.  Proposed modifications to the Duke criteria for the diagnosis of infective endocarditis.   Clin Infect Dis. 2000;30(4):633-638. doi:10.1086/313753 PubMedGoogle ScholarCrossref
18.
Tan  C, Hansen  M, Cohen  G, Boyle  K, Daneman  N, Adhikari  NK.  Accuracy of administrative data for identification of patients with infective endocarditis.   Int J Cardiol. 2016;224:162-164. doi:10.1016/j.ijcard.2016.09.030 PubMedGoogle ScholarCrossref
19.
Centers for Disease Control and Prevention. FAQs: bloodstream infection (BSI) events: central line–associated bloodstream infection and non—central line–associated bloodstream infection. Reviewed April 14, 2020. Accessed May 12, 2019. https://www.cdc.gov/nhsn/faqs/faq-bsi.html
20.
Vittinghoff  E, McCulloch  CE.  Relaxing the rule of ten events per variable in logistic and Cox regression.   Am J Epidemiol. 2007;165(6):710-718. doi:10.1093/aje/kwk052 PubMedGoogle ScholarCrossref
21.
Jones  M, Fowler  R.  Immortal time bias in observational studies of time-to-event outcomes.   J Crit Care. 2016;36:195-199. doi:10.1016/j.jcrc.2016.07.017 PubMedGoogle ScholarCrossref
22.
Boucher  HW, Talbot  GH, Bradley  JS,  et al.  Bad bugs, no drugs: no ESKAPE! an update from the Infectious Diseases Society of America.   Clin Infect Dis. 2009;48(1):1-12. doi:10.1086/595011 PubMedGoogle ScholarCrossref
23.
Barter  DM, Johnston  HL, Williams  SR, Tsay  SV, Vallabhaneni  S, Bamberg  WM.  Candida bloodstream infections among persons who inject drugs: Denver metropolitan area, Colorado, 2017-2018.   MMWR Morb Mortal Wkly Rep. 2019;68(12):285-288. doi:10.15585/mmwr.mm6812a3 PubMedGoogle ScholarCrossref
24.
Tobudic  S, Forstner  C, Burgmann  H,  et al.  Dalbavancin as primary and sequential treatment for gram-positive infective endocarditis: 2-year experience at the General Hospital of Vienna.   Clin Infect Dis. 2018;67(5):795-798. doi:10.1093/cid/ciy279 PubMedGoogle ScholarCrossref
25.
Grewal  HK, Ti  L, Hayashi  K, Dobrer  S, Wood  E, Kerr  T.  Illicit drug use in acute care settings.   Drug Alcohol Rev. 2015;34(5):499-502. doi:10.1111/dar.12270 PubMedGoogle ScholarCrossref
26.
Rosenthal  ES, Karchmer  AW, Theisen-Toupal  J, Castillo  RA, Rowley  CF.  Suboptimal addiction interventions for patients hospitalized with injection drug use-associated infective endocarditis.   Am J Med. 2016;129(5):481-485. doi:10.1016/j.amjmed.2015.09.024 PubMedGoogle ScholarCrossref
27.
Serota  DP, Niehaus  ED, Schechter  MC,  et al.  Disparity in quality of infectious disease vs addiction care among patients with injection drug use–associated Staphylococcus aureus bacteremia.   Open Forum Infect Dis. 2019;6(7):ofz289. doi:10.1093/ofid/ofz289 PubMedGoogle Scholar
28.
Sharma  M, Lamba  W, Cauderella  A, Guimond  TH, Bayoumi  AM.  Harm reduction in hospitals.   Harm Reduct J. 2017;14(1):32. doi:10.1186/s12954-017-0163-0 PubMedGoogle ScholarCrossref
29.
Straw  S, Baig  MW, Gillott  R,  et al.  Long-term outcomes are poor in intravenous drug users following infective endocarditis, even after surgery.   Clin Infect Dis. 2019;ciz869. doi:10.1093/cid/ciz869 PubMedGoogle Scholar
30.
Thakarar  K, Rokas  KE, Lucas  FL,  et al.  Mortality, morbidity, and cardiac surgery in injection drug use (IDU)–associated versus non-IDU infective endocarditis: the need to expand substance use disorder treatment and harm reduction services.   PLoS One. 2019;14(11):e0225460. doi:10.1371/journal.pone.0225460 PubMedGoogle Scholar
31.
Goodman-Meza  D, Weiss  RE, Gamboa  S,  et al.  Long term surgical outcomes for infective endocarditis in people who inject drugs: a systematic review and meta-analysis.   BMC Infect Dis. 2019;19(1):918. doi:10.1186/s12879-019-4558-2 PubMedGoogle ScholarCrossref
32.
Leahey  PA, LaSalvia  MT, Rosenthal  ES, Karchmer  AW, Rowley  CF.  High morbidity and mortality among patients with sentinel admission for injection drug use-related infective endocarditis.   Open Forum Infect Dis. 2019;6(4):ofz089. doi:10.1093/ofid/ofz089 PubMedGoogle Scholar
33.
Rudasill  SE, Sanaiha  Y, Mardock  AL,  et al.  Clinical outcomes of infective endocarditis in injection drug users.   J Am Coll Cardiol. 2019;73(5):559-570. doi:10.1016/j.jacc.2018.10.082 PubMedGoogle ScholarCrossref
34.
Suzuki  J, Johnson  J, Montgomery  M, Hayden  M, Price  C.  Outpatient parenteral antimicrobial therapy among people who inject drugs: a review of the literature.   Open Forum Infect Dis. 2018;5(9):ofy194. doi:10.1093/ofid/ofy194 PubMedGoogle Scholar
35.
Eaton  EF, Mathews  RE, Lane  PS,  et al.  A 9-point risk assessment for patients who inject drugs and require intravenous antibiotics: focusing inpatient resources on patients at greatest risk of ongoing drug use.   Clin Infect Dis. 2019;68(6):1041-1043. doi:10.1093/cid/ciy722 PubMedGoogle ScholarCrossref
36.
Fanucchi  LC, Walsh  SL, Thornton  AC, Nuzzo  PA, Lofwall  MR.  Outpatient parenteral antimicrobial therapy plus buprenorphine for opioid use disorder and severe injection-related infections.   Clin Infect Dis. 2020;70(6):1226-1229. doi:10.1093/cid/ciz654PubMedGoogle Scholar
37.
Buehrle  DJ, Shields  RK, Shah  N, Shoff  C, Sheridan  K.  Risk factors associated with outpatient parenteral antibiotic therapy program failure among intravenous drug users.   Open Forum Infect Dis. 2017;4(3):ofx102. doi:10.1093/ofid/ofx102 PubMedGoogle Scholar
Limit 200 characters
Limit 25 characters
Conflicts of Interest Disclosure

Identify all potential conflicts of interest that might be relevant to your comment.

Conflicts of interest comprise financial interests, activities, and relationships within the past 3 years including but not limited to employment, affiliation, grants or funding, consultancies, honoraria or payment, speaker's bureaus, stock ownership or options, expert testimony, royalties, donation of medical equipment, or patents planned, pending, or issued.

Err on the side of full disclosure.

If you have no conflicts of interest, check "No potential conflicts of interest" in the box below. The information will be posted with your response.

Not all submitted comments are published. Please see our commenting policy for details.

Limit 140 characters
Limit 3600 characters or approximately 600 words
    Original Investigation
    Infectious Diseases
    August 12, 2020

    在注射治疗感染性心内膜炎的药物的人群中发生新血流感染和死亡的风险

    Author Affiliations
    • 1Department of Medicine, Schulich School of Medicine and Dentistry, London, Ontario, Canada
    • 2Division of Infectious Diseases, St Joseph’s Health Care, London, Ontario, Canada
    • 3Department of Epidemiology and Biostatistics, Schulich School of Medicine and Dentistry, London, Ontario, Canada
    • 4currently a medical student at Schulich School of Medicine and Dentistry, London, Ontario, Canada
    • 5Department of Family Medicine, Schulich School of Medicine and Dentistry, London, Ontario, Canada
    • 6Division of Infectious Diseases, St Joseph’s Health Care and London Health Sciences Centre, London, Ontario, Canada
    JAMA Netw Open. 2020;3(8):e2012974. doi:10.1001/jamanetworkopen.2020.12974
    关键点 español English

    问题  在注射药物和正在接受感染性心内膜炎治疗的人群中,新的血流感染中发现了哪些菌群?住院与门诊治疗环境是否与新的血流感染率和死亡风险增加相关?

    结果  这项队列研究涉及 420 例感染性心内膜炎以及 309 名注射药物的患者。研究发现,82 例并发新的血流感染 (20%) 且主要由革兰氏阴性杆菌和念珠菌属引起。在接受门诊治疗的患者中,新的血流感染和死亡并不常见。

    意义  这些调查结果表明,在注射治疗感染性心内膜炎的药物的人群中,新的血流感染十分常见,但经过精心选择的低风险患者可能适合接受门诊肠外抗菌药物治疗。

    Abstract

    Importance  People who inject drugs (PWID) who are being treated for infective endocarditis remain at risk of new bloodstream infections (BSIs) due to ongoing intravenous drug use (IVDU).

    Objectives  To characterize new BSIs in PWID receiving treatment for infective endocarditis, to determine the clinical factors associated with their development, and to determine whether new BSIs and treatment setting are associated with mortality.

    Design, Setting, and Participants  This retrospective cohort study was performed at 3 tertiary care hospitals in London, Ontario, Canada, from April 1, 2007, to March 31, 2018. Participants included a consecutive sample of all PWID 18 years or older admitted with infective endocarditis. Data were analyzed from April 1, 2007, to June 29, 2018.

    Main Outcomes and Measures  New BSIs and factors associated with their development, treatment setting of infective endocarditis episodes (ie, inpatient vs outpatient), and 90-day mortality.

    Results  The analysis identified 420 unique episodes of infective endocarditis in 309 PWID (mean [SD] patient age, 35.7 [9.7] years; 213 episodes [50.7%] involving male patients), with 82 (19.5%) complicated by new BSIs. There were 138 independent new BSIs, of which 68 (49.3%) were polymicrobial and 266 were unique isolates. Aerobic gram-negative bacilli (143 of 266 [53.8%]) and Candida species (75 of 266 [28.2%]) were the most common microorganisms. Ongoing inpatient IVDU was documented by a physician in 194 infective endocarditis episodes (46.2%), and 127 of these (65.5%) were confirmed by urine toxicology results. Multivariable time-dependent Cox regression demonstrated that previous infective endocarditis (hazard ratio [HR], 1.89; 95% CI, 1.20-2.98), inpatient treatment (HR, 4.49; 95% CI, 2.30-8.76), and physician-documented inpatient IVDU (HR, 5.07; 95% CI, 2.68-9.60) were associated with a significantly higher rate of new BSIs, whereas inpatient addiction treatment was associated with a significantly lower rate (HR, 0.53; 95% CI, 0.32-0.88). New BSIs were not significantly associated with 90-day mortality (HR, 1.76; 95% CI, 0.78-4.02); significant factors associated with mortality included inpatient infective endocarditis treatment (HR, 3.39; 95% CI, 1.53-7.53), intensive care unit admission (HR, 9.51; 95% CI, 4.91-18.42), and methicillin-resistant Staphylococcus aureus infective endocarditis (HR, 1.77; 95% CI, 1.03-3.03), whereas right-sided infective endocarditis was associated with a significantly lower mortality rate (HR, 0.41; 95% CI, 0.25-0.67).

    Conclusions and Relevance  In this study, new BSIs were common in PWID receiving parenteral treatment for infective endocarditis. Discharging patients to outpatient treatment was not associated with an increase in new BSI incidence or mortality; carefully selected PWID may therefore be considered for such treatment.

    Introduction

    Infective endocarditis in people who inject drugs (PWID) is rising in incidence, coinciding with the opioid epidemic.1-6 Prolonged parenteral antimicrobial treatment is the standard of care for infective endocarditis in PWID. Although oral therapy has been shown to be effective for infective endocarditis in non-PWID,7 concerns regarding adherence have limited this approach in PWID. Outpatient parenteral antimicrobial therapy (OPAT) is safe, efficacious, and cost-effective for treating many infections,8-11 but PWID are generally not considered candidates. This decision is in part due to the risk of new bloodstream infections (BSIs) from ongoing intravenous drug use (IVDU), commonly through central venous catheters inserted for antimicrobial treatment.9,12,13 New BSIs are difficult to manage, with the selected antimicrobial regimen needing to treat both the initial infective endocarditis and the superimposed infection; they may also necessitate lengthening of antimicrobial therapy and hospital admission. Therefore, it is often mandated that infective endocarditis in PWID be treated in monitored inpatient settings to optimize adherence, deter IVDU, and prevent new BSIs. However, whether remaining admitted actually decreases IVDU or improves outcomes is unknown. Hence, recent Infectious Diseases Society of America guidelines for OPAT concluded that “there is insufficient evidence to make a recommendation for or against treating PWID with OPAT at home.”14(p17)

    The objectives of this study were to characterize new BSIs in PWID receiving treatment for infective endocarditis and determine the clinical factors associated with their development. We also sought to compare the rates of new BSIs in inpatient and outpatient (ie, OPAT) settings and to determine whether new BSIs and treatment setting were associated with mortality.

    Methods
    Study Design

    This retrospective cohort study was conducted across all 3 acute care hospitals in London, Ontario, Canada, from April 1, 2007, to March 31, 2018. Data were recently published on 202 first episodes of infective endocarditis in PWID presenting from April 1, 2007, to March 31, 2016.15 For the present analysis, we extended the time for enrollment and added recurrent episodes. Research ethics approval was obtained from the Lawson Research institutional review board. Informed consent from study participants was waived for this retrospective collection of deidentified data. This study adhered to the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) reporting guideline for cohort studies.16

    Patient Identification and Data Set Design

    We included all PWID 18 years or older admitted during the study period who met modified Duke criteria for definite infective endocarditis.17 The hospital medical records were screened for patients who received a code from the International Classification of Diseases, Ninth Revision, Clinical Modification (ICD-9-CM), or International Statistical Classification of Diseases, Tenth Revision, Clinical Modification (ICD-10-CM), for infective endocarditis in their discharge diagnoses; the accuracy of these codes has been validated.18 Medical records were reviewed by 2 infectious diseases physicians (including E.S.), with data abstracted into a standardized data set. The data set was restricted to patients who met modified Duke criteria for definite infective endocarditis and reported active IVDU within 3 months of admission.15

    The data set was organized into unique episodes of infective endocarditis, defined as occurring at least 6 months after a previous episode or associated with both a new vegetation and a new microorganism within 6 months.15 Each episode of infective endocarditis may consist of several admissions (eg, for complications, after leaving against medical advice), and individual patients may have multiple unique episodes. Follow-up was 90 days from the last hospital discharge associated with the episode, gathered from inpatient and outpatient health records (including physician visits, laboratory results, and radiography), pharmacy dispensing records, and local obituary records.

    Patient Characteristics

    Demographic characteristics included age, sex, and homelessness; comorbidities included previous infective endocarditis, prosthetic cardiac valve, and HIV, hepatitis B virus, or hepatitis C virus infection. The microbiological origin of each infective endocarditis episode was determined by index blood cultures and cultures of operative specimens, when available. Endocardial involvement on echocardiography was documented. Variables for IVDU included substances used before admission, documented (by a physician in the medical record) and confirmed (by urine toxicology screen result) inpatient drug misuse, and referral to addictions treatment.

    Definition of BSIs

    New BSIs were defined as the identification of a microorganism in blood culture not secondary to an infection at another body site (as per Centers for Disease Control and Prevention definitions for laboratory-confirmed BSIs19) and different from those grown on index blood cultures at the time of infective endocarditis diagnosis. The positive blood cultures must have been obtained at least 48 hours after index blood cultures and while the patient was receiving parenteral antimicrobials. They could not be associated with a new vegetation; these were classified as novel episodes of infective endocarditis. Single isolates of skin and oral microflora were excluded. We documented the microbiological origins and susceptibility profiles of new BSIs and whether they occurred during inpatient (≥48 hours after admission) or outpatient (≥48 hours after discharge) treatment.

    Outcomes and Hospital Care Measures

    Outcomes included embolic complications (embolic stroke, intracerebral hemorrhage, mycotic aneurysm, pulmonary embolism, and intra-abdominal embolism), cardiac complications (heart failure, conduction delay, and intracardiac abscess), other infections (nonembolic central nervous system, respiratory, and musculoskeletal), acute kidney injury (≥1.5-fold elevation in creatinine level from baseline or need for renal replacement therapy), and hepatic injury (≥3-fold elevation in liver enzyme level from baseline). We also evaluated septic shock, intensive care unit (ICU) admission, in-hospital mortality, and 90-day mortality.

    Hospital care measures included a peripherally inserted central catheter (PICC), PICC complications, and surgical management of infective endocarditis (ie, cardiac surgery). We determined whether patients received inpatient or outpatient treatment for infective endocarditis or left against medical advice. Patients who were treated predominantly as inpatients, defined as less than 14 days of outpatient antimicrobials or leaving against medical advice with less than 14 days of antimicrobials remaining, were considered to have received inpatient treatment.

    We determined the total inpatient and outpatient days of intravenous access (central or peripheral) associated with each episode. This was determined via chest x-rays (for dates of central venous catheter insertion and removal), clinical notes, and medication administration records (with patients assumed to have intravenous access if administered an intravenous medication).

    Statistical Analysis

    Data were analyzed from April 1, 2007, to June 29, 2018. We performed descriptive analyses of episodes of infective endocarditis with and without new BSIs, comparing demographic characteristics, comorbidities, microbiological origin, endocardial involvement, IVDU variables, complications, mortality, and hospital care measures. The analyses were repeated comparing episodes with and without new candidemia. Descriptive analyses were also performed comparing patients who died and who survived at 90 days after discharge. Variables for IVDU were not compared in mortality analyses because in-hospital IVDU and addiction treatment referral would not be possible for patients who died. Similarly, PICC use was not compared because PICCs were generally placed in clinically stable patients, with a citywide policy of requiring negative blood cultures before insertion. Discrete variables were coded dichotomously as present or absent (with additional categorizations where appropriate) and reported as frequencies and percentages. Continuous variables were reported as means with SDs. Missing data were documented for all variables. We also compared the incidence of new BSIs by treatment setting, defined as the number of BSIs per 1000 days of intravenous access, using Poisson regression.

    We conducted multivariable time-dependent Cox proportional hazards regression analyses for clinical characteristics associated with new BSIs and 90-day mortality. The regression models included variables selected a priori based on expected clinical significance; both models were designed targeting at least 5 events for each variable.20 The model for new BSIs included previous infective endocarditis, inpatient treatment of infective endocarditis, physician-documented inpatient IVDU, PICC use, ICU admission, and inpatient addiction treatment. Time at risk was defined as time of admission until line removal at discharge or in the community. The model for 90-day mortality included new BSIs, inpatient treatment of infective endocarditis, right-sided infective endocarditis, methicillin-resistant Staphylococcus aureus (MRSA) infective endocarditis, ICU admission, surgical management, and inpatient addiction treatment referral. Time at risk was defined as from the time of admission until death or 90 days after discharge. To account for immortal time bias,21 new BSIs, inpatient addiction treatment, and surgical management were treated as time-dependent covariates. For both models, patients who left against medical advice were excluded owing to absence of appropriate follow-up; treatment setting was therefore a direct comparison between inpatient and outpatient treatment of infective endocarditis. All statistical analyses were performed with Excel, version 16.16.23 (Microsoft Corporation), and STATA statistical software, version 16.1 (StataCorp LLC). Two-sided P < .05 indicated significance.

    Results
    Description of Study Population

    Our study included 420 unique episodes of infective endocarditis in 309 PWID (mean [SD] patient age, 35.7 [9.7] years; 213 episodes [50.7%] involving male patients and 207 [49.3%] involving female patients). We screened 2027 discharges with an ICD-9-CM or an ICD-10-CM code for infective endocarditis, identifying 688 episodes fulfilling definite modified Duke criteria. Of these, 268 were excluded for lack of IVDU within 3 months, yielding the final cohort. Twelve episodes were recurrences with a new organism within 3 months of a prior episode, although none occurred while patients still had an intravenous catheter. Two hundred fifty-three episodes (60.2%) involved inpatient treatment, 131 (31.2%) involved outpatient treatment, and 36 (8.6%) resulted in the patient leaving against medical advice.

    The characteristics of the cohort are shown in Table 1. New BSIs were identified in 82 of 420 infective endocarditis episodes (19.5%; new BSI group). The predominant microbiological origin of all infective endocarditis episodes was S aureus (326 of 420 [77.6%]), with MRSA constituting 88 of 420 episodes (21.0%). New BSIs were more common among episodes in which the patient had a history of infective endocarditis (37 of 82 [45.1%] vs 91 of 338 [26.9%]). New BSIs were also more common among episodes with right-sided infective endocarditis (72 of 80 [90.0%] vs 229 of 329 [69.6%]); left-sided disease was more common among episodes without new BSIs (16 of 80 [20.0%] vs 125 of 329 [38.0%]).

    Physician-documented inpatient IVDU occurred in 194 of 420 infective endocarditis episodes (46.2%), whereas 127 of these (65.5%; 30.2% of all 420 episodes) had confirmed drug use by urine toxicology results. Drug use was associated with treatment setting; IVDU during admission was more frequent in episodes when patients remained in the hospital for the duration of antimicrobial therapy (120 of 253 [47.4%]) than in episodes when patients were discharged with OPAT (45 of 131 [34.4%]), with the highest rate in those who left against medical advice (29 of 36 episodes [80.6%]). Episodes complicated by new BSIs, compared with episodes without BSIs, had higher rates of physician-documented (69 of 82 [84.1%] vs 125 of 338 [37.0%]) and confirmed (45 of 82 [54.9%] vs 82 of 338 [24.3%]) drug use. Opiate (80 of 82 [97.6%] vs 285 of 338 [84.3%]) and polysubstance (72 of 82 [87.8%] vs 249 of 338 [73.7%]) use were also more common in the new BSI group (Table 1).

    Microbiological Origin of New BSIs

    Among the 82 episodes of infective endocarditis complicated by new BSIs, 138 independent new BSIs with 266 unique isolates were noted (Table 2). Almost half of the new BSIs were polymicrobial (68 of 138 [49.3%]). Aerobic gram-negative bacilli were most frequent (143 of 266 [53.8%]), with high rates of resistance to penicillins and cephalosporins and many species with inducible AmpC β-lactamase activity. Candida species constituted 75 of 266 isolates (28.2%), with low rates of azole and amphotericin B resistance. Gram-positive bacteria constituted 48 of 266 isolates (18.0%), more than half of which were Enterococcus species. The susceptibility profiles of new BSIs caused by gram-negative bacteria, gram-positive bacteria, and Candida species are shown in eTables 1 to 3 in the Supplement. The antimicrobials used to treat new BSIs are shown in eTable 4 in the Supplement.

    Outcomes and Hospital Care Measures

    Episodes of infective endocarditis complicated by new BSIs, compared with episodes without BSIs, had higher rates of septic pulmonary emboli (60 of 82 [73.2%] vs 187 of 338 [55.3%]), whereas embolic strokes (6 of 82 [7.3%] vs 57 of 338 [16.9%]) and intra-abdominal emboli (11 of 82 [13.4%] vs 74 of 338 [21.9%]) were more frequent in episodes without new BSIs (non-BSI group) (Table 3). No other differences in embolic complications, cardiac complications, nonembolic infections, and end-organ damage were observed. Admission to the ICU was less frequent in the new BSI group (21 of 82 [25.6%] vs 127 of 338 [37.6%]). Overall 90-day mortality was similar between the new BSI (11 of 82 [13.4%]) and non-BSI groups (53 of 338 [15.7%]).

    Use of PICCs was more frequent among episodes complicated by new BSIs compared with episodes without BSIs (80 of 82 [97.6%] vs 287 of 338 [84.9%]) (Table 3); PICC complications are shown in eTable 5 in the Supplement. More episodes in the new BSI group involved inpatient treatment (70 of 82 [85.4%] vs 183 of 338 [54.1%]), whereas more episodes in the non-BSI group involved outpatient treatment (12 of 82 [14.6%] vs 119 of 338 [35.2%]) or involved the patient leaving against medical advice (0 of 82 vs 36 of 338 [10.7%]). Fewer episodes involved surgical management in the new BSI group (8 of 82 [9.8%]) compared with the non-BSI group (56 of 338 [16.6%]).

    Secondary Analysis of Episodes Complicated by Candidemia

    New candidemia was identified in 55 of 420 infective endocarditis episodes (13.1%) and was associated with previous infective endocarditis (25 of 55 [45.5%] vs 103 of 365 [28.2%]), HIV (10 of 55 [18.2%] vs 35 of 365 [9.6%]), and right-sided infective endocarditis (49 of 54 [90.7%] vs 252 of 355 [71.0%]) (eTable 6 in the Supplement). Physician-documented (46 of 55 [83.6%] vs 148 of 365 [40.5%]) and urine toxicology–confirmed (31 of 55 [56.4%] vs 95 of 365 [26.0%]) drug use were more common in the new candidemia group (eTable 6 in the Supplement). With regard to outcomes, septic pulmonary emboli were more common among episodes with new candidemia (41 of 55 [74.5%] vs 206 of 365 [56.4%]); there was no difference in 90-day mortality (9 of 55 [16.4%] vs 55 of 365 [15.1%]) (eTable 7 in the Supplement). More episodes complicated by new candidemia were treated in the inpatient setting (49 of 55 [89.1%] vs 204 of 365 [55.9%]) (eTable 7 in the Supplement).

    Rates of Bloodstream Infections

    The inpatient new BSI rate was 9.60 (95% CI, 7.95-11.5) BSIs per 1000 days of intravenous access, with 115 BSIs and 11 985 days of intravenous access. In comparison, the outpatient new BSI rate was 5.23 (95% CI, 3.50-7.46) BSIs per 1000 days of intravenous access, with 27 BSIs and 5165 days of intravenous access. The incidence rate ratio comparing inpatient with outpatient rates of new BSIs was 1.84 (95% CI, 1.20-2.90; P = .003).

    Variables Associated With BSIs

    The multivariable Cox regression for predictors of new BSIs is shown in Table 4. Clinical factors associated with a significantly higher rate of new BSIs were previous infective endocarditis (hazard ratio [HR], 1.89; 95% CI, 1.20-2.98), inpatient treatment of infective endocarditis (HR, 4.49; 95% CI, 2.30-8.76), and physician-documented inpatient IVDU (HR, 5.07; 95% CI, 2.68-9.60). Inpatient addiction treatment was associated with a significantly lower rate of new BSIs (HR, 0.53; 95% CI, 0.32-0.88).

    Variables Associated With 90-Day Mortality

    The multivariable Cox regression for variables associated with 90-day mortality is shown in Table 5. New BSIs were not significantly associated with 90-day mortality (HR, 1.76; 95% CI, 0.78-4.02). Clinical factors associated with a significantly higher 90-day mortality rate were inpatient treatment of infective endocarditis (HR, 3.39; 95% CI, 1.53-7.53), ICU admission (HR, 9.51; 95% CI, 4.91-18.42), and MRSA infective endocarditis (HR, 1.77; 95% CI, 1.03-3.03). Right-sided infective endocarditis (HR, 0.41; 95% CI, 0.25-0.67) was associated with a significantly lower 90-day mortality rate. Clinical characteristics associated with 90-day mortality are shown in eTable 8 in the Supplement.

    Discussion

    To our knowledge, this study is the largest to date describing PWID with infective endocarditis by detailed medical record review. New BSIs were a frequent complication, affecting 19.5% of infective endocarditis episodes. Previous infective endocarditis was associated with new BSIs, suggesting that PWID with entrenched addictions are at higher risk. The association between inpatient drug misuse and new BSIs suggests that ongoing IVDU was likely responsible for most new BSIs, indicating that inpatient addiction treatment may serve as a protective factor. Furthermore, in-hospital treatment of infective endocarditis was not associated with fewer new BSIs, and there was no association between new BSIs and complications or mortality.

    Gram-negative bacteria were the most common cause of new BSIs, whereas gram-positive bacteria were cultured less frequently and were predominantly Enterococcus species; this is likely a reflection of almost all patients receiving antistaphylococcal and/or antistreptococcal therapy for their infective endocarditis. The new BSIs typically involved organisms that were resistant to patients’ infective endocarditis treatment. Gram-negative bacteria primarily consisted of the gram-negative components of ESKAPE organisms (Enterococcus faecium, S aureus, Klebsiella pneumoniae, Acinetobacter baumannii, Pseudomonas aeruginosa, and Enterobacter species)22; glucose nonfermenters; and organisms with inducible AmpC β-lactamase genes associated with antibiotic resistance, including Pseudomonas species, Stentrophomonas maltophilia, Acinetobacter species, Enterobacter species, and Klebsiella species. Candidemia was also common, particularly among inpatients; candidemia is classically associated with health care settings, but IVDU has emerged as an independent risk factor.23

    Empirical therapy in patients with suspected new BSI would ideally include an agent with activity against Pseudomonas species and S maltophilia, the 2 most common gram-negative isolates, such as levofloxacin or an antipseudomonal carbapenem with trimethoprim-sulfamethoxazole. Vancomycin may be added if not already part of the infective endocarditis regimen to cover Enterococcus species. Empirical antifungal therapy would include fluconazole or an echinocandin.

    Inpatient treatment of infective endocarditis was associated with increased rates of new BSIs and 90-day mortality compared with outpatient treatment. These findings may in part reflect detection bias, with admitted patients followed up more closely with serial clinical and microbiological assessments. However, follow-up to 90 days was comprehensive, including medical records capturing admissions and outpatient visits to all London hospitals and neighboring hospitals in southwestern Ontario, inpatient and outpatient pharmacy medication dispensing records, laboratory investigations (including blood cultures), radiography at regional inpatient and outpatient facilities, and local obituary records. Patients receiving OPAT also receive nursing visits and are evaluated by an infectious disease specialist at the end of treatment. This thorough assessment did not reveal any increase in new BSIs or mortality that would preclude PWID from OPAT, providing support that OPAT may be appropriate for carefully selected PWID. The development of long-acting antibiotics with broad gram-positive coverage including against MRSA (eg, dalbavancin, ortivancin) may also allow for safer delivery of OPAT to PWID in the future.24

    We found that ongoing inpatient IVDU was common, documented by a physician in almost half the episodes of infective endocarditis; this is likely an underestimate given that data collection was retrospective. We could not identify any previous studies evaluating physician-reported IVDU in hospitalized PWID, although a study of self-reported inpatient drug use found a similarly high rate.25 However, fewer than one-third of our patients were referred to inpatient addiction treatment. Poor access to addictions support for hospitalized PWID is a widespread concern.26,27 Similarly, provision of harm reduction materials for inpatients is uncommon28 and not practiced at our study hospitals. Although PWID with infective endocarditis tend to be younger and have fewer comorbidities than patients with non–IVDU-associated infective endocarditis, they often have worse outcomes, including higher rates of readmission and mortality.29-33 These disparities are largely due to inadequate treatment of substance use disorders and limited harm reduction services29-33; such interventions therefore need to be prioritized for hospitalized PWID.

    Previous studies have differed in whether OPAT is safe and effective in PWID. However, the studies tended to have small sample sizes and did not evaluate new BSIs as a complication. A literature review of 10 studies on OPAT in PWID34 found high treatment completion, low mortality, and few catheter-related complications, comparable to rates in non-PWID populations. Another study35 implemented a 9-point tool to identify PWID candidates for OPAT, with low-risk PWID discharged to complete treatment as outpatients. Implementation resulted in reduced length of stay and cost savings without increased readmissions. Similarly, a pilot randomized trial that combined OPAT with buprenorphine opioid agonist therapy found a significantly reduced length of hospital stay by 23.5 days.36 In contrast, a 2017 study found that 61.2% (41 of 67) of PWID discharged to receive OPAT experienced treatment failure.37 However, failure was defined broadly, including readmission, prolongation of antibiotic regimens, nonadherence to antibiotic regimens and clinic follow-up, and death. Furthermore, there was no comparator group, and 68.7% (46 of 67) of patients were discharged to a nursing facility, potentially indicating a population with high levels of comorbidities.

    Although evidence suggests that oral antibiotics can be used to treat infective endocarditis,7 this practice has not been evaluated in PWID. If validated, oral treatment may represent an alternative to OPAT, eliminating the need for prolonged venous access and consequent risk of new BSIs.

    Strengths and Limitations

    Our study has several strengths. This study is the first, to our knowledge, of PWID receiving long-term intravenous antimicrobial therapy to describe BSIs and compare outcomes by treatment setting. Our outcome measures were robust, and all cases were reviewed by infectious disease physicians and fulfilled modified Duke criteria for definite infective endocarditis. Comprehensive region-wide records allowed detailed follow-up of all patients to 90 days after discharge. We also accounted for immortal time bias by incorporating time-dependent variables.

    Our study also has some limitations. First, owing to retrospective data collection, the accuracy of data depended on the fidelity of prior documentation. For example, data on nonfatal drug overdoses were limited, available for only 4 outpatients and 2 inpatients. Second, outcomes may have been underestimated in patients who received OPAT, with admitted patients monitored more closely (ie, detection bias). Third, our comparison of infective endocarditis treatment setting was nonrandomized, with those receiving outpatient infective endocarditis treatment likely having lower-risk IVDU behaviors and fewer comorbidities (ie, selection bias). However, our intention was not to prove the superiority of outpatient treatment, but to demonstrate that discharging selected PWID with OPAT was not associated with increased morbidity and mortality. Last, our results may not be generalizable to different OPAT programs. In London, OPAT involves parenteral antimicrobials administered at the patient’s residence, whereas OPAT at some other centers involves patients presenting to an infusion clinic, with attendance potentially compromised in PWID.

    Conclusions

    We described new BSIs in PWID receiving treatment for infective endocarditis, finding higher frequency in patients with entrenched addictions and in the inpatient setting. Aerobic gram-negative bacilli and Candida species were the most common causative microorganisms. Ongoing IVDU was documented in almost half of inpatients being treated for infective endocarditis. Although our study was affected by detection and selection biases, we did not find higher rates of new BSIs or mortality among PWID discharged with OPAT, suggesting that OPAT may be appropriate for selected low-risk PWID. Prospective randomized studies of OPAT vs inpatient treatment or oral treatment for IVDU-associated infections are required. Furthermore, improved access to inpatient addictions and harm reduction services should be prioritized.

    Back to top
    Article Information

    Accepted for Publication: May 27, 2020.

    Published: August 12, 2020. doi:10.1001/jamanetworkopen.2020.12974

    Open Access: This is an open access article distributed under the terms of the CC-BY License. © 2020 Tan C et al. JAMA Network Open.

    Corresponding Author: Michael Silverman, MD, Division of Infectious Diseases, St Joseph’s Health Care, Room B3 404, 268 Grosvenor St, London, ON N6A 4V2, Canada (michael.silverman@sjhc.london.on.ca).

    Author Contributions: Drs Tan and Silverman had full access to all the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis.

    Concept and design: Tan, Koivu, Silverman.

    Acquisition, analysis, or interpretation of data: Tan, Shojaei, Wiener, Shah, Silverman.

    Drafting of the manuscript: Tan, Shojaei, Silverman.

    Critical revision of the manuscript for important intellectual content: All authors.

    Statistical analysis: Tan, Wiener, Silverman.

    Obtained funding: Silverman.

    Administrative, technical, or material support: Shojaei, Shah.

    Supervision: Koivu, Silverman.

    Conflict of Interest Disclosures: None reported.

    Funding/Support: This study was supported by the Ontario HIV Treatment Network and the St Joseph’s Health Care Foundation.

    Role of the Funder/Sponsor: The sponsors had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication.

    Additional Contributions: Adeel Sherazi, MD, MacKenzie Health, Vaughan, Ontario, Canada, helped with data acquisition and Susana Pearl, PhD, London Psychological Associates, London, Ontario, Canada, contributed to manuscript revision. Neither was compensated for this work.

    References
    1.
    Weir  MA, Slater  J, Jandoc  R, Koivu  S, Garg  AX, Silverman  M.  The risk of infective endocarditis among people who inject drugs: a retrospective, population-based time series analysis.   CMAJ. 2019;191(4):E93-E99. doi:10.1503/cmaj.180694 PubMedGoogle ScholarCrossref
    2.
    Hartman  L, Barnes  E, Bachmann  L, Schafer  K, Lovato  J, Files  DC.  Opiate injection-associated infective endocarditis in the southeastern United States.   Am J Med Sci. 2016;352(6):603-608. doi:10.1016/j.amjms.2016.08.010 PubMedGoogle ScholarCrossref
    3.
    Cooper  HL, Brady  JE, Ciccarone  D, Tempalski  B, Gostnell  K, Friedman  SR.  Nationwide increase in the number of hospitalizations for illicit injection drug use-related infective endocarditis.   Clin Infect Dis. 2007;45(9):1200-1203. doi:10.1086/522176 PubMedGoogle ScholarCrossref
    4.
    Fleischauer  AT, Ruhl  L, Rhea  S, Barnes  E.  Hospitalizations for endocarditis and associated health care costs among persons with diagnosed drug dependence—North Carolina, 2010-2015.   MMWR Morb Mortal Wkly Rep. 2017;66(22):569-573. doi:10.15585/mmwr.mm6622a1 PubMedGoogle ScholarCrossref
    5.
    Wurcel  AG, Anderson  JE, Chui  KK,  et al.  Increasing infectious endocarditis admissions among young people who inject drugs.   Open Forum Infect Dis. 2016;3(3):ofw157. doi:10.1093/ofid/ofw157 PubMedGoogle Scholar
    6.
    Schranz  AJ, Fleischauer  A, Chu  VH, Wu  LT, Rosen  DL.  Trends in drug use–associated infective endocarditis and heart valve surgery, 2007 to 2017: a study of statewide discharge data.   Ann Intern Med. 2019;170(1):31-40. doi:10.7326/M18-2124 PubMedGoogle ScholarCrossref
    7.
    Iversen  K, Ihlemann  N, Gill  SU,  et al.  Partial oral versus intravenous antibiotic treatment of endocarditis.   N Engl J Med. 2019;380(5):415-424. doi:10.1056/NEJMoa1808312 PubMedGoogle ScholarCrossref
    8.
    Paladino  JA, Poretz  D.  Outpatient parenteral antimicrobial therapy today.   Clin Infect Dis. 2010;51(suppl 2):S198-S208. doi:10.1086/653520 PubMedGoogle ScholarCrossref
    9.
    Tice  AD, Rehm  SJ, Dalovisio  JR,  et al; IDSA.  Practice guidelines for outpatient parenteral antimicrobial therapy: IDSA guidelines.   Clin Infect Dis. 2004;38(12):1651-1672. doi:10.1086/420939 PubMedGoogle ScholarCrossref
    10.
    MacKenzie  M, Rae  N, Nathwani  D.  Outcomes from global adult outpatient parenteral antimicrobial therapy programmes: a review of the last decade.   Int J Antimicrob Agents. 2014;43(1):7-16. doi:10.1016/j.ijantimicag.2013.09.006 PubMedGoogle ScholarCrossref
    11.
    Pericà S  JM, Llopis  J, González-Ramallo  V,  et al; Spanish Collaboration on Endocarditis-Grupo de Apoyo al Manejo de la Endocarditis Infecciosa en España (GAMES) Investigators.  Outpatient parenteral antibiotic treatment for infective endocarditis: a prospective cohort study from the GAMES cohort.   Clin Infect Dis. 2019;69(10):1690-1700. doi:10.1093/cid/ciz030 PubMedGoogle ScholarCrossref
    12.
    Seaton  RA, Barr  DA.  Outpatient parenteral antibiotic therapy: principles and practice.   Eur J Intern Med. 2013;24(7):617-623. doi:10.1016/j.ejim.2013.03.014 PubMedGoogle ScholarCrossref
    13.
    Andrews  MM, von Reyn  CF.  Patient selection criteria and management guidelines for outpatient parenteral antibiotic therapy for native valve infective endocarditis.   Clin Infect Dis. 2001;33(2):203-209. doi:10.1086/321814 PubMedGoogle ScholarCrossref
    14.
    Norris  AH, Shrestha  NK, Allison  GM,  et al.  2018 Infectious Diseases Society of America clinical practice guideline for the management of outpatient parenteral antimicrobial therapy.   Clin Infect Dis. 2019;68(1):e1-e35. doi:10.1093/cid/ciy745 PubMedGoogle ScholarCrossref
    15.
    Rodger  L, Glockler-Lauf  SD, Shojaei  E,  et al.  Clinical characteristics and factors associated with mortality in first-episode infective endocarditis among persons who inject drugs.   JAMA Netw Open. 2018;1(7):e185220. doi:10.1001/jamanetworkopen.2018.5220 PubMedGoogle Scholar
    16.
    Vandenbroucke  JP, von Elm  E, Altman  DG,  et al; STROBE Initiative.  Strengthening the Reporting of Observational Studies in Epidemiology (STROBE): explanation and elaboration.   Int J Surg. 2014;12(12):1500-1524. doi:10.1016/j.ijsu.2014.07.014 PubMedGoogle ScholarCrossref
    17.
    Li  JS, Sexton  DJ, Mick  N,  et al.  Proposed modifications to the Duke criteria for the diagnosis of infective endocarditis.   Clin Infect Dis. 2000;30(4):633-638. doi:10.1086/313753 PubMedGoogle ScholarCrossref
    18.
    Tan  C, Hansen  M, Cohen  G, Boyle  K, Daneman  N, Adhikari  NK.  Accuracy of administrative data for identification of patients with infective endocarditis.   Int J Cardiol. 2016;224:162-164. doi:10.1016/j.ijcard.2016.09.030 PubMedGoogle ScholarCrossref
    19.
    Centers for Disease Control and Prevention. FAQs: bloodstream infection (BSI) events: central line–associated bloodstream infection and non—central line–associated bloodstream infection. Reviewed April 14, 2020. Accessed May 12, 2019. https://www.cdc.gov/nhsn/faqs/faq-bsi.html
    20.
    Vittinghoff  E, McCulloch  CE.  Relaxing the rule of ten events per variable in logistic and Cox regression.   Am J Epidemiol. 2007;165(6):710-718. doi:10.1093/aje/kwk052 PubMedGoogle ScholarCrossref
    21.
    Jones  M, Fowler  R.  Immortal time bias in observational studies of time-to-event outcomes.   J Crit Care. 2016;36:195-199. doi:10.1016/j.jcrc.2016.07.017 PubMedGoogle ScholarCrossref
    22.
    Boucher  HW, Talbot  GH, Bradley  JS,  et al.  Bad bugs, no drugs: no ESKAPE! an update from the Infectious Diseases Society of America.   Clin Infect Dis. 2009;48(1):1-12. doi:10.1086/595011 PubMedGoogle ScholarCrossref
    23.
    Barter  DM, Johnston  HL, Williams  SR, Tsay  SV, Vallabhaneni  S, Bamberg  WM.  Candida bloodstream infections among persons who inject drugs: Denver metropolitan area, Colorado, 2017-2018.   MMWR Morb Mortal Wkly Rep. 2019;68(12):285-288. doi:10.15585/mmwr.mm6812a3 PubMedGoogle ScholarCrossref
    24.
    Tobudic  S, Forstner  C, Burgmann  H,  et al.  Dalbavancin as primary and sequential treatment for gram-positive infective endocarditis: 2-year experience at the General Hospital of Vienna.   Clin Infect Dis. 2018;67(5):795-798. doi:10.1093/cid/ciy279 PubMedGoogle ScholarCrossref
    25.
    Grewal  HK, Ti  L, Hayashi  K, Dobrer  S, Wood  E, Kerr  T.  Illicit drug use in acute care settings.   Drug Alcohol Rev. 2015;34(5):499-502. doi:10.1111/dar.12270 PubMedGoogle ScholarCrossref
    26.
    Rosenthal  ES, Karchmer  AW, Theisen-Toupal  J, Castillo  RA, Rowley  CF.  Suboptimal addiction interventions for patients hospitalized with injection drug use-associated infective endocarditis.   Am J Med. 2016;129(5):481-485. doi:10.1016/j.amjmed.2015.09.024 PubMedGoogle ScholarCrossref
    27.
    Serota  DP, Niehaus  ED, Schechter  MC,  et al.  Disparity in quality of infectious disease vs addiction care among patients with injection drug use–associated Staphylococcus aureus bacteremia.   Open Forum Infect Dis. 2019;6(7):ofz289. doi:10.1093/ofid/ofz289 PubMedGoogle Scholar
    28.
    Sharma  M, Lamba  W, Cauderella  A, Guimond  TH, Bayoumi  AM.  Harm reduction in hospitals.   Harm Reduct J. 2017;14(1):32. doi:10.1186/s12954-017-0163-0 PubMedGoogle ScholarCrossref
    29.
    Straw  S, Baig  MW, Gillott  R,  et al.  Long-term outcomes are poor in intravenous drug users following infective endocarditis, even after surgery.   Clin Infect Dis. 2019;ciz869. doi:10.1093/cid/ciz869 PubMedGoogle Scholar
    30.
    Thakarar  K, Rokas  KE, Lucas  FL,  et al.  Mortality, morbidity, and cardiac surgery in injection drug use (IDU)–associated versus non-IDU infective endocarditis: the need to expand substance use disorder treatment and harm reduction services.   PLoS One. 2019;14(11):e0225460. doi:10.1371/journal.pone.0225460 PubMedGoogle Scholar
    31.
    Goodman-Meza  D, Weiss  RE, Gamboa  S,  et al.  Long term surgical outcomes for infective endocarditis in people who inject drugs: a systematic review and meta-analysis.   BMC Infect Dis. 2019;19(1):918. doi:10.1186/s12879-019-4558-2 PubMedGoogle ScholarCrossref
    32.
    Leahey  PA, LaSalvia  MT, Rosenthal  ES, Karchmer  AW, Rowley  CF.  High morbidity and mortality among patients with sentinel admission for injection drug use-related infective endocarditis.   Open Forum Infect Dis. 2019;6(4):ofz089. doi:10.1093/ofid/ofz089 PubMedGoogle Scholar
    33.
    Rudasill  SE, Sanaiha  Y, Mardock  AL,  et al.  Clinical outcomes of infective endocarditis in injection drug users.   J Am Coll Cardiol. 2019;73(5):559-570. doi:10.1016/j.jacc.2018.10.082 PubMedGoogle ScholarCrossref
    34.
    Suzuki  J, Johnson  J, Montgomery  M, Hayden  M, Price  C.  Outpatient parenteral antimicrobial therapy among people who inject drugs: a review of the literature.   Open Forum Infect Dis. 2018;5(9):ofy194. doi:10.1093/ofid/ofy194 PubMedGoogle Scholar
    35.
    Eaton  EF, Mathews  RE, Lane  PS,  et al.  A 9-point risk assessment for patients who inject drugs and require intravenous antibiotics: focusing inpatient resources on patients at greatest risk of ongoing drug use.   Clin Infect Dis. 2019;68(6):1041-1043. doi:10.1093/cid/ciy722 PubMedGoogle ScholarCrossref
    36.
    Fanucchi  LC, Walsh  SL, Thornton  AC, Nuzzo  PA, Lofwall  MR.  Outpatient parenteral antimicrobial therapy plus buprenorphine for opioid use disorder and severe injection-related infections.   Clin Infect Dis. 2020;70(6):1226-1229. doi:10.1093/cid/ciz654PubMedGoogle Scholar
    37.
    Buehrle  DJ, Shields  RK, Shah  N, Shoff  C, Sheridan  K.  Risk factors associated with outpatient parenteral antibiotic therapy program failure among intravenous drug users.   Open Forum Infect Dis. 2017;4(3):ofx102. doi:10.1093/ofid/ofx102 PubMedGoogle Scholar
    ×