In an analysis of a prospective cohort of outpatient parenteral antimicrobial therapy patients, 18% experienced an adverse drug event (ADE) and 14.5% experienced a significant ADE. Patients with longer courses of therapy had lower ADE rates.
Keywords: OPAT, adverse drug events, vancomycin, antibiotic side effect, drug monitoring
Abstract
Background
To better monitor patients on outpatient parenteral antimicrobial therapy (OPAT), we need an improved understanding of risk factors for and timing of OPAT-associated adverse drug events (ADEs).
Methods
We analyzed a prospective cohort of patients on OPAT discharged from 2 academic medical centers. Patients underwent chart abstraction and a telephone survey. Multivariable analyses estimated adjusted incident rate ratios (aIRR) between clinical and demographic risk factors and clinician-determined clinically significant ADEs. Descriptive data were used to present patient-reported ADEs.
Results
Of 339 patients enrolled in the study, 18.0% experienced an ADE (N = 65), of which 49 were significant (14.5%, 2.24/1000 home-OPAT days). Patients with longer courses of therapy had lower rates of ADEs compared with patients treated for 0–13 days (14–27 days: aIRR, 0.44; 95% confidence interval [CI], 0.20–0.99; at least 28 days: aIRR, 0.11; 95% CI, 0.056–0.21). Risk factors for ADEs included female gender and receipt of daptomycin or vancomycin, while treatment for uncomplicated bacteremia and empiric treatment were associated with lower rates of ADEs.
Conclusions
OPAT-related ADEs were common and often occurred within 2 weeks of hospital discharge. Patients on OPAT should be monitored more closely for ADEs, including clinical assessment and laboratory monitoring, especially within the first weeks after hospital discharge and particularly among women and patients who receive vancomycin.
Outpatient parenteral antimicrobial therapy (OPAT) allows patients to receive parenteral antimicrobials outside of acute care hospitals [1]. Home OPAT is typically delivered by a patient or caregiver (ie, friend, family, or neighbor) with support from home health nurses and home infusion agencies. In the hospital, clinical assessment and laboratory testing are readily accessible to detect potential adverse drug events (ADEs), but in OPAT, testing occurs less frequently and providers may not receive test results until one or more days later [2–5]. Furthermore, the responsibility for medication infusion and catheter care transitions from healthcare professionals to the patient and caregiver.
Thirty-day readmissions are common among OPAT patients, ranging from 17% to 27% [2, 6–10]. Up to 1 in 4 of these readmissions are due to ADEs [7, 8]. Understanding how to reduce ADEs is therefore essential to reducing readmissions. Most studies have used retrospective chart reviews and may not have captured all ADEs. Reported incidences of ADEs in OPAT ranges from 2% to 44%, depending on the definition of ADE [6, 11–14]. Fewer studies have focused on ADE rates, which may be 4.5/1000 antimicrobial-days [12].
Understanding the timing of ADEs in OPAT is important in understanding when clinical assessment and laboratory testing are most essential. However, we do not understand when ADEs may occur in OPAT when patients are receiving antimicrobial agents for weeks to months. To know how frequently patients should be monitored, we need to understand when ADEs occur in OPAT and what ADEs occur. Current monitoring guidelines are based on expert consensus [15].
We analyzed a prospective cohort of OPAT patients to determine rates of OPAT-related ADEs, risk factors for OPAT-related ADEs, and when these ADEs are most likely to occur.
METHODS
Patient Population and Setting
We performed a subanalysis of a prospective cohort of patients receiving home infusion therapy, expanding a previously described cohort [16]. Eligible patients were aged ≥18 years and discharged from 1 of 2 tertiary-care academic medical centers in Baltimore, Maryland (Johns Hopkins Hospital and Johns Hopkins Bayview Medical Center) between March 2015 and February 2017 to home with assistance from home infusion agencies. Although any clinician at either hospital could prescribe OPAT, infectious diseases physicians manage most OPAT patients in separate services at each hospital. Patients were required to have peripherally inserted central catheters, tunneled central venous catheters (CVCs), or midline catheters for OPAT. Patients were ineligible if they were in hospice care. As patients orally consented for the study and completed a telephone survey, we excluded patients who did not speak English or could not verbally consent. Eligible patients were contacted for telephone consent 2 weeks after hospital discharge, as previously described [16]. Three attempts were made to contact each patient during business hours. Patients could have used any home infusion or home nursing agency for antimicrobial agents and supplies and for training and support in CVC care, respectively.
Survey Instrument
Consenting patients underwent a 10-minute telephone survey that focused on complications such as readmissions and ADEs (Supplementary Table S1). The survey instrument was piloted among 10 patients prior to the study, with changes made based on feedback, focusing on OPAT outcomes [16].
The electronic health record (EHR) was abstracted for demographic information, CVC characteristics, OPAT indication, clinical data, readmissions, and ADEs through 1 month after OPAT completion.
Variables
Age was dichotomized at 65 years, focusing on the population of patients eligible for Medicare [17]. As few enrolled patients were Asian American, Hispanic, or other racial or ethnic groups, racial or ethnic group was categorized as white non-Hispanic, black non-Hispanic, or other. Insurer was characterized as Medicare, Medicaid, or private, as few were self-pay or uninsured. The Charlson comorbidity index [18] was calculated and dichotomized at 2 based on its distribution with the outcome. OPAT days were calculated as the number of days between hospital discharge and documented OPAT completion. We counted each day of OPAT, no matter the number of antimicrobials received, as 1 day of OPAT.
The indications for OPAT were characterized by infection site. Patients could have had more than 1 indication. Only parenteral agents were recorded.
The primary outcome was a clinically significant OPAT-related ADE, that is, an ADE documented in the EHR by a physician, physician assistant, or nurse practitioner that resulted in a hospital admission, change in antimicrobial agent, early termination of antimicrobial therapy, or Clostridium difficile infection. If a patient was on more than 1 antimicrobial agent, we attributed the ADE to the agent that the clinician had determined caused the ADE. We were concerned that patients may have misattributed symptoms to their antimicrobial agents or attributed symptoms that were OPAT-related ADEs to other causes, so we included patient-reported ADEs as descriptive data only. Patients were asked whether they had experienced any side effects and to freely respond with what these side effects were including symptoms (eg, diarrhea) or laboratory abnormalities (eg, “kidney troubles”). We also reported all-cause and ADE-related hospitalizations within 30 days of hospital discharge.
Data Analyses
Descriptive statistics were used for demographic, clinical, and outcome data (Stata ver. 14.0, College Station, Texas). Predictors included demographic and clinical variables. For significant ADEs, univariable and multivariable Poisson regressions were used to estimate predictors of ADEs/1000 OPAT-days. Covariates were considered if the association with the outcome was P ≤ 0.20 (2-sided) and were removed in a stepwise fashion if the covariate’s association with the outcome was P > .20 (2-sided). In addition, descriptive statistics were used to present patient-reported ADEs. Kappa scores were reported between patient-reported ADEs and ADEs identified through the EHR. A total of 322 patients were needed to achieve 80% power to calculate a 50% difference in the primary relationship.
The study was approved as expedited with oral consent by the Johns Hopkins University School of Medicine Institutional Review Board.
RESULTS
Of 672 patients discharged on OPAT, 570 were eligible for the study and 339 enrolled in the study (59.5%, 21878 home OPAT-days; Supplementary Figure S1). Osteomyelitis was the most common indication for OPAT (N = 102, 30.1%). The most common antimicrobial agents prescribed were vancomycin (N = 89, 26.3%), ceftriaxone (N = 41, 12.1%), and piperacillin–tazobactam (N = 39, 11.5%). Patients were on OPAT for a median of 29 days (Table 1).
Table 1.
Demographic and Clinical Characteristics Among 339 Patients Discharged From 2 Hospitals on Outpatient Parenteral Antimicrobial Therapy
| Demographic | N (% of 339) |
|---|---|
| Age, y (median, IQR) | 55 (41–63) |
| Female gender | 159 (46.9) |
| Race/Ethnicity | |
| White non-Hispanic | 238 (70.2) |
| Black non-Hispanic | 78 (23.0) |
| Other racial/ethnic group | 23 (6.8) |
| Insurance | |
| Private (missing = 2 self-pay) | 215 (63.8) |
| Medicaid | 40 (11.9) |
| Medicare | 82 (24.3) |
| Type of catheter in the chart | |
| Peripherally inserted central catheter | 241 (71.1) |
| Tunneled central venous catheter | 88 (26.0) |
| Midline | 10 (3.0) |
| Recent malignancy | 91 (26.8) |
| Diabetes | 100 (29.5) |
| Solid organ transplant | 41 (12.1) |
| Charlson comorbidity index (mean, median, IQR) | 4.1, 3.0, 1–6 |
| >2 | 207 (61.1) |
| Days on OPAT (median, IQR, total days) | 29, 15–44, 21878 |
| 0–13 | 75 (21.1) |
| 14–27 | 84 (24.8) |
| At least 28 | 180 (53.1) |
| Site of infection | |
| Uncomplicated bacteremia | 71 (20.9) |
| Endocarditis or endovascular infection | 24 (7.1) |
| Cellulitis | 19 (5.6) |
| Epidural infection or discitis | 27 (8.0) |
| Hardware or prosthesis infection | 47 (13.9) |
| Abdominal abscess or other abdominal infection | 30 (8.9) |
| Meningitis | 27 (8.0) |
| Osteomyelitis | 102 (30.1) |
| Pneumonia | 49 (14.5) |
| Septic arthritis | 25 (7.4) |
| Organism being treated | |
| Coagulase-negative Staphylococcus | 37 (10.9) |
| Methicillin-resistant Staphylococcus aureus | 28 (8.3) |
| Methicillin-susceptible S. aureus | 47 (13.9) |
| Other gram-positive cocci (Streptococcus, 49; Enterococcus, 20; other, 19) | 88 (26.0) |
| Any gram-negative rod (Acinetobacter, 1; Burkholderia, 1; Escherichia coli, 18; Enterobacter, 5; Klebsiella, 15; Proteus, 7; Pseudomonas, 39; Salmonella, 1; Serratia, 4) | 85 (25.1) |
| Other bacterial (Borrelia or Syphilis, 13; Propionibacterium acnes, 15; other anaerobe, 7; Mycobacteria, 5) | 40 (11.8) |
| Fungal (Candida, 7 or Cryptococcus, 1) | 8 (2.4) |
| Virus (Cytomegalovirus, 11; herpes simplex virus, 5; varicella zoster virus, 3) | 19 (5.6) |
| Empiric | 64 (18.9) |
| Polymicrobial | 22 (6.5) |
| Antibiotic prescribed for OPAT | |
| Ampicillin and penicillin (ampicillin, 6; ampicillin-sulbactam, 21; penicillin, 18) | 45 (13.3) |
| Other penicillins (nafcillin, 3; oxacillin, 22; piperacillin–tazobactam, 30) | 55 (16.2) |
| Cefazolin | 15 (4.4) |
| Ceftriaxone | 41 (12.1) |
| Cefepime | 26 (7.7) |
| Ceftaroline | 2 (0.6) |
| Carbapenem (ertapenem, 26; imipenem, 4; meropenem, 23) | 53 (15.6) |
| Daptomycin | 8 (2.4) |
| Vancomycin | 89 (26.3) |
| Aminoglycoside | 28 (8.3) |
| Other antibiotics (azithromycin, 2; aztreonam, 1; ciprofloxacin, 2; clindamycin, 2; colistin, 1; tigecycline, 1) | 9 (2.7) |
| Antiviral medication | 21 (6.2) |
| Antifungal medication | 12 (3.5) |
| More than 1 antimicrobial agent | 66 (19.5) |
Abbreviations: IQR, interquartile range; OPAT, outpatient parenteral antimicrobial therapy.
ADEs were common (Table 2); 61 patients had a clinician-documented ADE (18.0%). Of these, 49 were considered clinically significant and were included in the primary outcome (14.5%, 2.24/1000 OPAT-days).
Table 2.
Outcomes among 339 Patients Discharged from 2 Hospitals on Outpatient Parenteral Antimicrobial Therapy
| Outcomes among OPAT Patients | Total (% of 339) |
|---|---|
| Any readmission within 30 days of discharge | 61 (18.0) |
| Readmission within 30 days of hospital discharge due to OPAT-related ADE | 12 (3.5) |
| Any ADE | 61 (18.0) |
| ADE with no treatment (% of 61) | 5 (8.2) |
| ADE where clinician treated symptoms only (% of 61) | 5 (8.2) |
| ADE where clinician changed medication dose (% of 61) | 5 (8.2) |
| ADE where clinician changed antimicrobial agent (% of 61) | 23 (37.7) |
| ADE where clinician stopped medications early (% of 61) | 19 (31.1) |
| ADE of Clostridium difficile infection (% of 61) | 5 (8.2) |
| Any significant ADE (change of medication, stopping medication, C. difficile infection, hospitalization due to ADE) (% of 339) | 49 (14.5) |
| Significant ADEs per 1000 home OPAT days (95% confidence interval) | 2.24 (1.69–2.96) |
Adverse drug event (ADE) due to OPAT, as documented by a physician, nurse practitioner, or physician assistant in the medical record.
Abbreviations: ADE, adverse drug event; OPAT, outpatient parenteral antimicrobial therapy.
We assessed clinician-documented ADEs, including all ADEs and clinically significant ADEs (Table 3). Only 1.5% of patients in this study developed C. difficile infections. Vancomycin led to ADEs in 18.0% of patients (N = 61), of which 19 (21.3%) were clinically significant, most commonly nephrotoxicity (N = 11, 18.0%). Ceftriaxone led to only 1 ADE (2.5%), which was not clinically significant (nausea, which was not treated). Piperacillin–tazobactam led to ADEs in 5 patients (16.7%), of which 4 (10.3%) were clinically significant (C. difficile infection, N = 1; nephrotoxicity, N = 1; hepatotoxicity, N = 1; edema necessitating hospitalization, N = 1). Only 6 patients had ADEs that led to early termination within 0–13 days (8.0% of N = 75 patients on 0–13 OPAT-days).
Table 3.
Adverse Drug Events among 339 Patients Discharged from the Hospital on Outpatient Parenteral Antimicrobial Therapy as Recorded by a Clinician*
| Parenteral Antimicrobial Agent | ADEa | Significant ADEb | Type of ADE | Response to ADE |
|---|---|---|---|---|
| Amikacin (N = 4) | 1 (25.0%) | 1 (25.0%) | Ototoxicity (N = 1); electrolyte abnormality (N = 1) | Changed medication (N = 2) |
| Ampicillin–sulbactam (N = 21) | 4 (19.1%) | 4 (19.1%) | Rash (N = 1); nausea (N = 1); diarrhea (N = 1); nephrotoxicity (N = 1); edema (N = 1); | Changed medication (N = 2); stopped all medication (N = 2) |
| Cefazolin (N = 15) | 1 (6.7%) | 1 (6.7%) | Diarrhea (N = 1) | Changed medication (N = 1) |
| Cefepime (N = 26) | 4 (15.4%) | 3 (11.5%) | Rash (N = 1); Clostridium difficile (N = 2); cutaneous candida (N = 1) | Treated symptoms (N = 1); stopped all medication (N = 1); C. difficile treated (N = 2); readmitted due to ADE (N = 1) |
| Ciprofloxacin (N = 2) | 1 (50.0%) | 1 (50.0%) | C. difficile (N = 1) | Treated symptoms (N = 1); C. difficile treated (N = 1) |
| Ceftriaxone (N = 41) | 1 (2.5%) | 0 (0.0%) | Nausea (N = 1) | No treatment (N = 1) |
| Ceftazidime (N = 2) | 2 (100%) | 2 (100%) | Hepatotoxicity (N = 1); delirium (N = 1) | Changed medications (N = 2) |
| Daptomycin (N = 8) | 3 (37.5%) | 3 (37.5%) | Creatinine kinase elevation (N = 3)c | Changed medications (N = 2); stopped all medications (N = 1) |
| Ertapenem (N = 26) | 4 (15.4%) | 4 (15.4%) | C. difficile (N = 1); diarrhea (N = 1); hepatotoxicity (N = 2) | Treated symptoms (N = 1); stopped all medication (N = 3); C. difficile treated (N = 1) |
| Gentamicin (N = 3) | 1 (33.3%) | 1 (33.3%) | Nephrotoxicity (N = 1) | Changed medication (N = 1) |
| Liposomal amphotericin (N = 6) | 4 (66.7%) | 4 (66.7%) | Nausea (N = 1); nephrotoxicity (N = 2); hepatotoxicity (N = 1) | Stopped all medication (N = 4) |
| Meropenem (N = 23) | 1 (4.4%) | 0 (0.0%) | Diarrhea (N = 1) | Treated symptoms (N = 1) |
| Nafcillin (N = 3) | 2 (66.7%) | 2 (66.7%) | Nausea (N = 1); nephrotoxicity (N = 1) | Changed medication (N = 2); readmitted due to ADE (N = 2) |
| Oxacillin (N = 22) | 4 (18.2%) | 2 (9.1%) | Rash (N = 1); hepatotoxicity (N = 3); drug–drug interaction causing hemorrhagic stroke (N = 1) | No treatment (N = 2); changed dose (N = 1); stopped medication (N = 1); readmitted due to ADE (N = 1) |
| Penicillin (N = 18) | 2 (11.1%) | 1 (5.6%) | Nephrotoxicity (N = 1); electrolyte abnormality (N = 1) | Stopped all medication (N = 1) |
| Piperacillin–tazobactam (N = 39) | 5 (16.7%) | 4 (10.3%) | Rash (N = 1); nausea (N = 1); C. difficile (N = 1); diarrhea (N = 1); nephrotoxicity (N = 1); hepatotoxicity (N = 1); edema (N = 1) | No treatment (N = 1); changed dose (N = 1); stopped all medication (N = 2); C. difficile (N = 1); readmitted (N = 1) |
| Tobramycin (N = 21) | 3 (14.3%) | 1 (4.8%) | Ototoxicity (N = 2); dizziness (N = 2); nephrotoxicity (N = 1) | Changed dose (N = 3) |
| Vancomycin (N = 89) | 22 (24.7%) | 19 (21.3%) | Rash (N = 1); elevated levels (N = 2); ototoxicity (N = 1); nausea (N = 2); C. difficile (N = 1); nephrotoxicity (N = 11); cytopenias (N = 6); drug fevers (N = 2); edema (N = 1) | Nothing (N = 1); treated symptoms (N=1); changed dose (N=2); changed medication (N=9); stopped all medication (N=5); readmitted due to ADE (N=5); C. difficile treated (N=1) |
| Total ADEs | 61 (18.0%) | 49 (14.5%) |
*A clinician is a physician, nurse practitioner, or physician assistant.
Abbreviation: ADE, adverse drug event.
aNo ADEs were reported for acyclovir (10 patients), ampicillin (6 patients), azithromycin (2 patients), aztreonam (1 patient), ceftaroline (2 patients), clindamycin (1 patient), fluconazole (1 patient), ganciclovir (11 patients), imipenem (4 patients), colistin (1 patient), micafungin (5 patients), or tigecycline (1 patient).
bSignificant ADEs were those that led to a change in antimicrobial agent, early termination of medications, readmission, or Clostridium difficile infection.
cElevation in creatinine kinase was defined as recorded by a clinician and requiring a change in therapy (change in antimicrobial agent or early termination of therapy).
Adjusted incidence risk ratios (aIRRs) were calculated for significant ADEs/1000 OPAT-days (Table 4). Independent predictors of having a significant ADE included female gender (aIRR, 2.56; 95% confidence interval [CI], 1.40–4.67), receipt of daptomycin (aIRR, 11.77; 95% CI, 2.67–51.84), and receipt of vancomycin (aIRR, 2.19; 95% CI, 1.78–5.72). Independent predictors of not having a significant ADE included uncomplicated bacteremia (aIRR, 0.30; 95% CI, 0.13–0.68) and being empirically treated with antimicrobial therapy (aIRR, 0.35; 95% CI, 0.16–0.76). The longer patients were on OPAT, the less likely they were to have a significant ADE (vs being on OPAT for 0–13 days; 14–27 days: aIRR, 0.35; 95% CI, 0.14–0.87 and ≥28 days: aIRR, 0.15; 95% CI, 0.070–0.32).
Table 4.
Association of Covariates and Rate of Significant Adverse Drug Events among 339 Patients Discharged on Outpatient Parenteral Antimicrobial Therapy
| Characteristic | Significant ADE/1000 OPAT-days (IRR, 95% CI) | Significant ADE/1000 OPAT-days (aIRR, 95% CI) |
|---|---|---|
| Age >65 y | 1.08 (0.55–2.12) | Not included |
| Female | 1.95 (1.09–3.49) | 2.56 (1.40–4.67) |
| Race/ethnicity | ||
| White non-Hispanic | – | Not included |
| African American non-Hispanic | 1.03 (0.48–2.21) | Not included |
| Other race | 0.65 (0.23–1.81) | Not included |
| Insurance | ||
| Private | – | Not included |
| Medicaid | 2.38 (1.04–5.41) | Not included |
| Medicare | 1.48 (0.76–2.88) | Not included |
| Diabetes | 1.46 (0.80–2.68) | Not included |
| Malignancy | 0.40 (0.21–0.75) | Not included |
| Charlson comorbidity index ≥2 | 0.63 (0.36–1.12) | Not included |
| Uncomplicated bacteremia | 0.46 (0.21–1.02) | 0.30 (0.13–0.68) |
| Hardware involvement | 2.43 (1.21–4.87) | Not included |
| Abdominal | 0.27 (0.067–1.13) | Not included |
| Meningitis | 3.14 (1.13–8.72) | Not included |
| Osteomyelitis | 2.99 (1.71–5.25) | Not included |
| Sepsis | 2.49 (1.06–5.84) | Not included |
| Coagulase-negative Staphylococcus | 1.56 (0.66–3.66) | Not included |
| Any gram-negative rod | 1.10 (0.57–2.11) | Not included |
| Enterococcus | 2.59 (0.93–7.20) | Not included |
| Methicillin-resistant Staphylococcus aureus | 3.45 (1.47–8.11) | Not included |
| Methicillin-sensitive S. aureus | 1.51 (0.64–3.55) | Not included |
| Propionibacterium acnes | 2.98 (1.18–7.51) | Not included |
| Empiric treatment | 0.42 (0.20–0.89) | 0.35 (0.16–0.76) |
| Polymicrobial infection | 1.69 (0.72–3.97) | Not included |
| Ampicillin–sulbactam | 2.77 (1.00–7.70) | Not included |
| Cefepime | 0.43 (0.13–1.37) | Not included |
| Ceftriaxone | 0.39 (0.095–1.62) | Not included |
| Daptomycin | 4.86 (1.18–20.00) | 11.77 (2.67–51.84) |
| Meropenem | 0.22 (0.031–1.62) | Not included |
| Vancomycin | 2.20 (1.25–3.87) | 2.19 (1.78–5.72) |
| More than 1 antibiotic | 1.07 (0.53–2.14) | Not included |
| Days on OPAT | ||
| 0–13 | – | Not included |
| 14–27 | 0.38 (0.15–0.95) | 0.35 (0.14–0.87) |
| At least 28 | 0.11 (0.053–0.24) | 0.15 (0.070–0.32) |
Abbreviations: ADE, adverse drug event; aIRR: adjusted incidence rate ratio; CI, confidence interval; IRR, incidence rate ratio; OPAT, outpatient parenteral antimicrobial therapy.
Significant ADEs resulted in change in antimicrobial agent, stopping antimicrobial therapy early, readmission due to ADE, or Clostridium difficile infection. ADEs were recorded in the medical record by a clinician (physician, nurse practitioner, or physician assistant).
Patient-reported ADEs were more frequent than clinician-reported ADEs (Table 5). A total of 97 patients (28.6%) reported any ADE, only 39 of which were also clinician reported. Meanwhile, clinicians reported 22 ADEs that were unreported by patients. Interrater reliability between patient-reported and clinician-reported total ADEs was only κ = 0.35. Many ADEs reported by both patients and clinicians were laboratory abnormalities (N = 10). Notably, patients frequently reported ADEs such as loss of appetite, fatigue, malaise, and mucocutaneous candidiasis that clinicians did not report. Of the most commonly prescribed OPAT agents, 43.8% of patients on vancomycin reported an ADE (N = 39), 26.8% of patients on ceftriaxone reported an ADE (N = 11), and 33.3% of patients on piperacillin–tazobactam reported an ADE (N = 13).
Table 5.
Patient-Reported Adverse Drug Events Related to Outpatient Parenteral Antimicrobial Therapy in Surveys of 339 Patients Discharged from 2 Hospitals
| Parenteral = Antimicrobial Agenta | ADE (% on agent) | Type of ADE |
|---|---|---|
| Acyclovir (N = 10) | 1 (10.0) | Swelling (N=1) |
| Ampicillin (N = 6) | 1 (16.7) | Poor appetite (N=1) |
| Ampicillin–sulbactam (N = 21) | 7 (33.3) | Rash (N = 4), trouble breathing (N = 1), diarrhea (N = 1), mouth itching (N = 1), cutaneous candidiasis (N = 1) |
| Azithromycin (N = 2) | 1 (50.0) | Dizziness (N = 1) |
| Cefepime (N = 26) | 12 (46.2) | Rash (N = 1), dose too high (N = 1), nausea and vomiting (N = 4), diarrhea (N = 4), hives (N = 1), loss of appetite (N = 1), fatigue (N = 1), cutaneous candidiasis (N = 1) |
| Ceftaroline (N = 2) | 2 (100.0) | Nausea and vomiting (N = 1), diarrhea (N = 1) |
| Ciprofloxacin (N = 2) | 1 (50.0) | Diarrhea (N = 1), weight loss (N = 1) |
| Clindamycin (N = 1) | 1 (100.0) | Itching (N = 1) |
| Ceftriaxone (N = 41) | 11 (26.8) | Nausea and vomiting (N = 2), dizziness (N = 1), renal insufficiency (N = 1), muscle pain (N = 1), trouble breathing (N = 1), headache (N = 1), diarrhea (N = 5), “Herxheimer reaction” (N = 1), dehydration (N = 1), metallic taste (N = 1), fever (N = 1), cutaneous candidiasis (N = 1) |
| Daptomycin (N = 8) | 3 (37.5) | Rash (N = 1), diarrhea (N = 1), hives (N = 1), hepatotoxicity (N = 1) |
| Ertapenem (N = 26) | 5 (19.2) | Rash (N = 2), nausea and vomiting (N = 1), nephrotoxicity (N = 1), hepatotoxicity (N = 1), diarrhea (N = 1) |
| Ganciclovir (N = 11) | 1 (9.1) | Cytopenia (N = 1) |
| Gentamicin (N = 3) | 2 (66.7) | Nausea and vomiting (N = 1), nephrotoxicity (N = 1) |
| Imipenem (N = 4) | 1 (25.0) | Feeling jittery (N = 1) |
| Liposomal amphotericin (N = 6) | 2 (33.3) | Dizziness (N = 1), nephrotoxicity (N = 1), hepatotoxicity (N = 1), headache (N = 1), general malaise (N = 1), loss of appetite (N = 1) |
| Meropenem (N = 23) | 4 (17.4) | Rash (N = 1), muscle pain (N = 1), diarrhea (N = 1), fatigue (N = 1) |
| Micafungin (N = 5) | 1 (20.0) | Fatigue (N = 1) |
| Nafcillin (N = 3) | 1 (33.3) | Cytopenia (N = 1) |
| Oxacillin (N = 22) | 4 (18.2) | Nausea and vomiting (N = 1), diarrhea (N = 1), itching (N = 1) |
| Penicillin (N = 18) | 5 (27.8) | Rash (N = 2), nausea and vomiting (N = 1), muscle pain (N = 1), cytopenia (N = 1), fatigue (N = 1) |
| Piperacillin–tazobactam (N = 39) | 13 (33.3) | Rash (N = 1), difficulty hearing (N = 1), nausea and vomiting (N = 3), dizziness (N = 1), nephrotoxicity (N = 1), muscle pain (N = 1), hepatotoxicity (N = 1), diarrhea (N = 7), Clostridium difficile (N = 1), pain (N = 1), cutaneous candidiasis (N = 1) |
| Tobramycin (N = 21) | 8 (38.1) | Rash (N = 2), ototoxicity (N = 1), nausea and vomiting (N = 1), dizziness (N = 1), nephrotoxicity (N = 1), hepatotoxicity (N = 1), diarrhea (N = 1), C. difficile (N = 1) |
| Vancomycin (N = 89) | 39 (43.8) | Rash (N = 3), elevated dose (N = 4), trouble hearing (N = 4), nausea and vomiting (N = 9), dizziness (N = 2), nephrotoxicity (N = 6), muscle trouble (N = 1), hepatotoxicity (N = 1), cytopenia (N = 4), headache (N = 3), diarrhea (N = 8), dehydration (N = 1), C. difficile (N = 1), fatigue (N = 1), general malaise (N = 1), fever (N = 2), weakness (N = 1), weight loss (N = 1) |
| Total ADEs | 97 (28.6) |
Abbreviation: ADE, adverse drug event.
aPatients on several antimicrobial agents did not report ADEs: amikacin (N = 4), aztreonam (N = 1), cefazolin (N = 15), ceftazidime (N = 1), colistin (N = 1), fluconazole (N = 1), tigecycline (N = 1).
Basic demographic, CVC information, and 30-day all-cause readmissions were abstracted for the 231 patients who were eligible for the study but either declined participation or were unable to be reached. No differences were detected.
DISCUSSION
In this prospective cohort of OPAT patients, 14.5% (2.24/1000 OPAT-days) had a significant ADE that required a change in therapy, early termination of therapy, readmission, or resulted in C. difficile infection. Total ADEs were even more common, occurring in 18.0% of patients. In their first 2 weeks of OPAT, patients were at a higher risk of developing a significant ADE. These findings underscore the importance of judiciously prescribing OPAT agents, ensuring appropriate dosage of medication, educating patients about ADEs, and carefully monitoring for ADEs.
Previous studies on ADEs in OPAT have largely been retrospective chart reviews [2, 6, 19–27]. Incidences of ADEs in prior studies have varied, possibly depending on ADE definition or the population or condition examined [2, 6, 28, 29]. In this study, we enrolled consenting patients discharged on home-based OPAT and defined ADEs as those that were clinically significant (causing a change in antimicrobial agent, early termination of therapy, readmission, or C. difficile infection). We also reported patient-reported ADEs, although as descriptive data only as we were concerned about misattribution of symptoms.
Our data suggest that ADEs may be highest in the first 2 weeks of OPAT. The hospital-to-home transition is a time when medication errors are common [30, 31] and, in the absence of close monitoring, frequently occur among OPAT patients [2]. Dosing or other medication changes may occur just prior to hospital discharge, and patients who initiate therapy immediately after discharge may make errors as they learn to perform OPAT-related tasks [32]. Evidence-based education that starts at the hospital bedside is needed to mitigate this high-risk period. The transition home in OPAT is a time when clinical monitoring and timely laboratory testing is particularly important.
ADE incidence was lower than in several recent studies, such as one where 44.4% of patients experienced an ADE, although only 26% resulted in therapeutic changes [13]. Our rate of significant ADEs was also lower than a prior study’s rate of 4.5/1000 days [12], perhaps as our study focused on the most clinically significant ADEs. Our incidence of ADEs was slightly lower than in inpatients, where 20% of patients receiving an antibiotic for 24 hours or longer had an ADE [33]. In an inpatient population, many medications are initiated simultaneously in patients who may have been already critically ill, while patients in our study had already initiated therapy prior to study entry.
We also assessed which antimicrobial agents led to ADEs. Among antimicrobial agents received by at least 15 patients, those who received cefazolin, ceftriaxone, penicillin, and piperacillin–tazobactam had few clinically significant ADEs (6.7%, 0.0%, 5.6%, and 6.1%, respectively), while patients who received vancomycin, ampicillin–sulbactam, and ertapenem had frequent clinically significant ADEs (21.3%, 19.1%, and 15.4%, respectively). Vancomycin in particular frequently led to nephrotoxicity and cytopenias, as in prior studies [14, 34]. Although a prior study of OPAT patients suggested that patients discharged on vancomycin had more ADEs than patients discharged on daptomycin [14], we found an association between daptomycin and ADEs as well. In our patient population, patients were typically discharged on daptomycin only for complicated vancomycin-resistant Enterococcus infections or for methicillin-resistant Staphylococcus aureus infections that did not respond to vancomycin; patients who had had a prior ADE to vancomycin were also typically discharged. These patients may be sicker than other patients or require higher doses of daptomycin. Our data suggest that patients on vancomycin and daptomycin be monitored closely.
We identified risk factors for ADEs. As in at least 1 other study, we did not find an association between older age and ADEs [19]. However, women had a higher rate of ADEs. This may be related to a lower volume of distribution in women. Dosing of antimicrobials and monitoring for ADEs may be particularly important in women.
Uncomplicated bacteremia was associated with a lower ADE rate, as was empiric therapy. It is possible that patients with uncomplicated bacteremia were less ill than those with complicated bacteremia or infections in other locations, making these patients less susceptible to ADEs. In addition, it is possible that culture-negative infections were due to a lower burden of pathogens or less-virulent pathogens, and therefore patients who were empirically treated may also be less ill.
Although we focused on clinician-documented clinically significant ADEs, patient- and clinician-reported ADEs often did not correlate. Most patient-reported ADEs were symptoms (eg, poor appetite, nausea and vomiting, dizziness, diarrhea). It is possible that patients did not think many of these ADEs were significant enough to mention to a clinician or clinicians did not think them important enough to document. As OPAT occurs in the home without direct observation by clinicians, it is likely that ADEs are underreported. However, many clinically significant laboratory-based ADEs were reported by both clinicians and patients, perhaps indicating that clinicians had explained to patients why a change in antimicrobial agent was needed. OPAT may cause even more ADEs than previously recognized. Patients should be informed about the possibility of ADEs and asked to report these.
We believe that our study enhances the OPAT literature in several ways. We used a prospective cohort. Our primary analysis focused on clinician-defined, clinically significant ADEs as we felt that clinicians would be most interested in ADEs that result in a change in antimicrobial agent, early termination of therapy, readmission, or C. difficile infection. We also measured patient-reported ADEs to understand the patient experience.
Our study did have several limitations. First, our population may not represent other OPAT populations as this study was performed at 2 urban, academically affiliated hospitals where medically complex populations were cared for (12.1% had a solid organ transplant, 26.8% were undergoing treatment for a malignancy, and 7.7% had cystic fibrosis). We excluded non-English speakers because patients were required to orally consent and respond to a telephone survey. Some antimicrobials (eg, aminoglycosides, tigecycline, ceftaroline, liposomal amphotericin) were prescribed too infrequently to include in our analyses. Patients initiated their antimicrobial agents prior to hospital discharge; we did not include predischarge data as we were concerned that antimicrobial agents may frequently be changed in the hospital, making ADE attribution difficult. We likely underestimated ADEs as we required clinician documentation; patients may not inform clinicians of all symptoms, clinicians may not document all ADEs, and not all laboratory test results may be received by clinicians. There may have also been variation in clinician-reported ADEs; one clinician may have made a change in antimicrobial agent due to a symptom while another may have elected to monitor the patient closely. In addition, we were unable to evaluate data from patients who had medical care outside of the 9 hospitals and affiliated clinics linked to our EHR. We did not collect data on oral antimicrobial agents or on other medications, so some of the ADEs experienced may have been misattributed to OPAT. Some OPAT courses may have been terminated early due to ADEs. However, only 6 of 19 patients who experienced an ADE that led to early termination of therapy experienced this ADE within the first 2 weeks of hospital discharge.
Our data suggest that ADEs are common and often can be detected by laboratory testing or clinical assessment. Guidelines based on expert opinion suggest that OPAT patients receive weekly clinic visits and regular laboratory monitoring [15, 35]. This recommendation may be based on an assumption that laboratory testing will be received and reviewed by the ordering provider on a timely basis. However, most infectious diseases clinicians do not have a way to ensure receipt of ordered laboratory test results [2, 36]. In the absence of a care coordination program that allows for rapid communication between patients, inpatient clinicians, outpatient clinicians, home health agencies, and home infusion agencies, as few as 30% of laboratory tests may be received by the provider [2]. Physician nonreceipt of OPAT-related laboratory testing is associated with increased admissions [37]. Care coordination can improve compliance with laboratory testing [2, 3] and may prevent ADEs.
Our study indicates that close monitoring of OPAT patients for ADEs may be particularly essential in the first 2 weeks after hospital discharge. Patients on daptomycin or vancomycin as well as women may need closer monitoring for ADE development. To prevent harm from ADEs, patients should be monitored closely, especially in the first 2 weeks after discharge, with laboratory testing and clinic visits.
Supplementary Data
Supplementary materials are available at Clinical Infectious Diseases online. Consisting of data provided by the authors to benefit the reader, the posted materials are not copyedited and are the sole responsibility of the authors, so questions or comments should be addressed to the corresponding author.
Supplementary Material
Notes
Acknowledgments. We appreciate the contributions of Kathleen Pulice, BS; Amanda Krosche, BS; Matthew Naumann, BS; and Mayo Levering, BS in enrolling patients in the study. We acknowledge the contributions of Vicky Belotserkovsky and Nekia Murphy of the Johns Hopkins Home Care Group for their assistance with providing the study team a database of eligible participants and Terri Taylor and home care coordinators at Johns Hopkins Hospital and Johns Hopkins Bayview Medical Center for their assistance with introducing patients to the study.
Financial support. S. C. K. receives funding from the National Center for Advancing Translational Sciences/Johns Hopkins Institute for Clinical and Translational Research (KL2 award KL2TR001077). This work was supported by the Sherrilyn and Ken Fisher Center for Environmental Infectious Diseases Discovery award.
Potential conflicts of interest. S. E. C. is a committee member for Novartis and Theravance. All other authors: No reported conflicts. All authors have submitted the ICMJE Form for Disclosure of Potential Conflicts of Interest. Conflicts that the editors consider relevant to the content of the manuscript have been disclosed.
References
- 1. Paladino JA, Poretz D. Outpatient parenteral antimicrobial therapy today. Clin Infect Dis 2010; 51(Suppl 2):S198–208. [DOI] [PubMed] [Google Scholar]
- 2. Keller SC, Ciuffetelli D, Bilker W et al. The impact of an infectious diseases transition service on the care of outpatients on parenteral antimicrobial therapy. J Pharm Technol 2013; 29:205–14. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3. Shah PJ, Bergman SJ, Graham DR, Glenn S. Monitoring of outpatient parenteral antimicrobial therapy and implementation of clinical pharmacy services at a community hospital infusion unit. J Pharm Pract 2015; 28:462–8. [DOI] [PubMed] [Google Scholar]
- 4. Muldoon EG, Switkowski K, Tice A, Snydman DR, Allison GM. A national survey of infectious disease practitioners on their use of outpatient parenteral antimicrobial therapy (OPAT). Infect Dis (Lond) 2015; 47:39–45. [DOI] [PubMed] [Google Scholar]
- 5. Huck D, Ginsberg JP, Gordon SM, Nowacki AS, Rehm SJ, Shrestha NK. Association of laboratory test result availability and rehospitalizations in an outpatient parenteral antimicrobial therapy programme. J Antimicrob Chemother 2014; 69:228–33. [DOI] [PubMed] [Google Scholar]
- 6. Lee B, Tam I, Weigel B 4th et al. Comparative outcomes of β-lactam antibiotics in outpatient parenteral antibiotic therapy: treatment success, readmissions and antibiotic switches. J Antimicrob Chemother 2015; 70:2389–96. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7. Means L, Bleasdale S, Sikka M, Gross AE. Predictors of hospital readmission in patients receiving outpatient parenteral antimicrobial therapy. Pharmacotherapy 2016; 36:934–9. [DOI] [PubMed] [Google Scholar]
- 8. Yan M, Elligsen M, Simor AE, Daneman N. Patient characteristics and outcomes of outpatient parenteral antimicrobial therapy: a retrospective study. Can J Infect Dis Med Microbiol 2016; 2016:8435257. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9. Allison GM, Muldoon EG, Kent DM et al. Prediction model for 30-day hospital readmissions among patients discharged receiving outpatient parenteral antibiotic therapy. Clin Infect Dis 2014; 58:812–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10. Bhavan KP, Brown LS, Haley RW. Self-administered outpatient antimicrobial infusion by uninsured patients discharged from a safety-net hospital: a propensity-score-balanced retrospective cohort study. PLoS Med 2015; 12:e1001922. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11. Cervera C, Sanroma P, González-Ramallo V et al. ; DAPTODOM Investigators Safety and efficacy of daptomycin in outpatient parenteral antimicrobial therapy: a prospective and multicenter cohort study (DAPTODOM trial). Infect Dis (Lond) 2017; 49:200–7. [DOI] [PubMed] [Google Scholar]
- 12. Farry JK, Miles CD, Collins CD, Malani PN. Adverse events among renal transplant recipients receiving outpatient parenteral antimicrobial infusion therapy. Transpl Infect Dis 2009; 11:284–5. [DOI] [PubMed] [Google Scholar]
- 13. Hale CM, Steele JM, Seabury RW, Miller CD. Characterization of drug-related problems occurring in patients receiving outpatient antimicrobial therapy. J Pharm Pract 2017: 897190016688771. [DOI] [PubMed] [Google Scholar]
- 14. Shrestha NK, Mason P, Gordon SM et al. Adverse events, healthcare interventions and healthcare utilization during home infusion therapy with daptomycin and vancomycin: a propensity score-matched cohort study. J Antimicrob Chemother 2014; 69:1407–15. [DOI] [PubMed] [Google Scholar]
- 15. Tice AD, Rehm SJ, Dalovisio JR et al. ; Infectious Diseases Society of America Practice guidelines for outpatient parenteral antimicrobial therapy. IDSA guidelines. Clin Infect Dis 2004; 38:1651–72. [DOI] [PubMed] [Google Scholar]
- 16. Keller SC, Williams D, Gavgani M et al. Environmental exposures and the risk of central venous catheter complications and readmissions in home infusion therapy patients. Infect Control Hosp Epidemiol 2016:1–8. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17. Keller S, Pronovost P, Cosgrove S. What Medicare is missing. Clin Infect Dis 2015; 61:1890–1. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18. D’Hoore W, Sicotte C, Tilquin C. Risk adjustment in outcome assessment: the Charlson comorbidity index. Methods Inf Med 1993; 32:382–7. [PubMed] [Google Scholar]
- 19. Mujal A, Sola J, Hernandez M et al. Safety and effectiveness of outpatient parenteral antimicrobial therapy in older people. J Antimicrob Chemother 2016; 71:1402–7. [DOI] [PubMed] [Google Scholar]
- 20. Matthews PC, Conlon CP, Berendt AR et al. Outpatient parenteral antimicrobial therapy (OPAT): is it safe for selected patients to self-administer at home? A retrospective analysis of a large cohort over 13 years. J Antimicrob Chemother 2007; 60:356–62. [DOI] [PubMed] [Google Scholar]
- 21. Tice AD, Strait K, Ramey R, Hoaglund PA. Outpatient parenteral antimicrobial therapy for central nervous system infections. Clin Infect Dis 1999; 29:1394–9. [DOI] [PubMed] [Google Scholar]
- 22. Lin JW, Kacker A, Anand VK, Levine H. Catheter- and antibiotic-related complications of ambulatory intravenous antibiotic therapy for chronic refractory rhinosinusitis. Am J Rhinol 2005; 19:365–9. [PubMed] [Google Scholar]
- 23. Upton A, Ellis-Pegler RB, Woodhouse A. Outpatient parenteral antimicrobial therapy (OPAT): a review of experience at Auckland Hospital. N Z Med J 2004; 117:U1020. [PubMed] [Google Scholar]
- 24. Gomez M, Maraqa N, Alvarez A, Rathore M. Complications of outpatient parenteral antibiotic therapy in childhood. Pediatr Infect Dis J 2001; 20:541–3. [DOI] [PubMed] [Google Scholar]
- 25. Hoffman-Terry ML, Fraimow HS, Fox TR, Swift BG, Wolf JE. Adverse effects of outpatient parenteral antibiotic therapy. Am J Med 1999; 106:44–9. [DOI] [PubMed] [Google Scholar]
- 26. Ho J, Archuleta S, Sulaiman Z, Fisher D. Safe and successful treatment of intravenous drug users with a peripherally inserted central catheter in an outpatient parenteral antibiotic treatment service. J Antimicrob Chemother 2010; 65:2641–4. [DOI] [PubMed] [Google Scholar]
- 27. Kieran J, O’Reilly A, Parker J, Clarke S, Bergin C. Self-administered outpatient parenteral antimicrobial therapy: a report of three years experience in the Irish healthcare setting. Eur J Clin Microbiol Infect Dis 2009; 28:1369–74. [DOI] [PubMed] [Google Scholar]
- 28. Mujal A, Sola J, Hernandez M et al. Safety and effectiveness of home intravenous antibiotic therapy for multidrug-resistant bacterial infections. Eur J Clin Microbiol Infect Dis 2015; 34:1125–33. [DOI] [PubMed] [Google Scholar]
- 29. Beieler AM, Dellit TH, Chan JD et al. Successful implementation of outpatient parenteral antimicrobial therapy at a medical respite facility for homeless patients. J Hosp Med 2016; 11:531–5. [DOI] [PubMed] [Google Scholar]
- 30. Bell CM, Rahimi-Darabad P, Orner AI. Discontinuity of chronic medications in patients discharged from the intensive care unit. J Gen Intern Med 2006; 21:937–41. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31. Kripalani S, Jackson AT, Schnipper JL, Coleman EA. Promoting effective transitions of care at hospital discharge: a review of key issues for hospitalists. J Hosp Med 2007; 2:314–23. [DOI] [PubMed] [Google Scholar]
- 32. Arthur W, Bennett W, Stanush PL, McNelly TL. Factors that influence skill decay and retention: a quantitative review and analysis. Human Performance 1998; 11: 57–101. [Google Scholar]
- 33. Tamma PD, Avdic E, Li DX, Dzintars K, Cosgrove SE. Adverse events associated with antibiotic use in hospitalized patients. JAMA Intern Med 2017; 177:1–8. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34. Felder KK, Marshall LM, Vaz LE, Barnes PD. Risk factors for complications during outpatient parenteral antimicrobial therapy for adult orthopedic and neurosurgical infections. South Med J 2016; 109:53–60. [DOI] [PubMed] [Google Scholar]
- 35. Huminer D, Bishara J, Pitlik S. Home intravenous antibiotic therapy for patients with infective endocarditis. Eur J Clin Microbiol Infect Dis 1999; 18:330–4. [DOI] [PubMed] [Google Scholar]
- 36. Lane MA, Marschall J, Beekmann SE et al. Outpatient parenteral antimicrobial therapy practices among adult infectious disease physicians. Infect Control Hosp Epidemiol 2014; 35:839–44. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 37. Gordon SM, Shrestha NK, Rehm SJ. Transitioning antimicrobial stewardship beyond the hospital: the Cleveland Clinic’s community-based parenteral anti-infective therapy (CoPAT) program. J Hosp Med 2011; 6(Suppl 1):S24–30. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.