Outpatient parenteral antimicrobial therapy

INTRODUCTION — Outpatient parenteral antimicrobial therapy (OPAT) refers to administration of intravenous antimicrobial therapy on at least two separate days without an intervening hospitalization [1].

OPAT is now routinely employed for patients requiring intravenous antimicrobial therapy for a variety of infective conditions, including endocarditis, osteomyelitis, and prosthetic joint infection (PJI), enabling early hospital discharge. For some conditions (such as cellulitis), initiation of OPAT in the outpatient or emergency department setting may allow avoidance of hospitalization entirely [2,3].

Issues related to OPAT in adults (including logistics, patient selection, antibiotic selection, vascular access, and monitoring) are reviewed here.

Issues related to treatment of infective endocarditis, osteomyelitis, PJI, cellulitis, and other common OPAT diagnoses are discussed separately, as are details related to individual antimicrobial agents. (See "Overview of management of infective endocarditis in adults" and "Nonvertebral osteomyelitis in adults: Treatment" and "Prosthetic joint infection: Treatment" and "Acute cellulitis and erysipelas in adults: Treatment".)

GENERAL PRINCIPLES

Benefits and risks — OPAT benefits patients and health care systems by facilitating outpatient follow-up care, reducing the length of inpatient stays, and limiting exposure to nosocomial pathogens, reducing costs [4,5].

Risks of OPAT include catheter-related complications, adverse drug events, reduced frequency of clinical oversight in comparison to hospitalization, and risks posed by provision of care by patients and caregivers who may have no formal medical training.

Program components — Key components of OPAT programs include [6-8]:

●The presence of a structured program with a formal OPAT team.

●Careful attention to patient assessment and the adequacy of care after hospitalization.

●Education of patients and caregivers (if applicable) about the OPAT plan.

●Assessment by an infectious diseases specialist and a pharmacist (which are known to reduce costs and re-admission rates) to identify patients who may be treated with oral therapy.

●Communication and monitoring after hospital discharge. Monitoring includes devising mechanisms for rapid communication between patients and team members, timely review of laboratory studies, and prompt mitigation if laboratory test results are not obtained or reviewed.

●Monitoring of programmatic quality, including hospital readmission rates and complications, such as adverse drug effects.

Logistics of delivery — Options for the location of administering OPAT include administration in the patient's home, at an infusion center (clinic or hospital based), or a skilled nursing/rehabilitation facility. Each option has advantages and disadvantages; the choice depends on local resources, insurance coverage, and patient factors:

●OPAT in a patient's home allows administration of antimicrobial therapy in a familiar environment. This arrangement may require the patient or caregiver to administer at least some of the treatment without the assistance of medical personnel.

●OPAT in an infusion center allows routine clinical assessments and obviates the need for the patient to perform his or her own infusions; however, this arrangement is not practical for antimicrobial regimens with multiple doses per day. Furthermore, patients must be mobile, and transportation to the infusion center must be feasible.

●OPAT in a skilled nursing facility (SNF) or rehabilitation facility allows the patient to receive antimicrobial regimens that are too cumbersome to administer at home or at an infusion center. SNFs provide the greatest degree of medical oversight; however, care in SNFs may be associated with increased risk of exposure to nosocomial pathogens and reduced patient satisfaction [9].

CLINICAL APPROACH

Patient selection — Infections that can be managed with OPAT include infective endocarditis, osteomyelitis and other orthopedic infections, and skin and soft tissue infections [2,10-13]. (See "Overview of management of infective endocarditis in adults" and "Nonvertebral osteomyelitis in adults: Treatment" and "Prosthetic joint infection: Treatment" and "Acute cellulitis and erysipelas in adults: Treatment".)

Traditionally, certain infections, such as endocarditis, osteomyelitis, and orthopedic device infections, have been treated with long courses of IV antimicrobial therapy. With increasing data supporting the use of oral antimicrobial therapy for such infections [14,15], patients being considered for OPAT should first be assessed to ensure that there is not an equally efficacious oral alternative. OPAT should be reserved for patients in whom there is not an equally effective oral alternative based on microbiology, tolerability or drug-drug interactions, oral bioavailability, or based on impaired enteral absorption. Patients with life-threatening infections, such as endocarditis and meningitis, and those with highly resistant infections are often still best served with IV antimicrobial therapy.

After the need for IV antimicrobial therapy has been verified, candidates for home-based OPAT must be evaluated to ensure they are medically stable and able and willing to self-administer intravenous antimicrobial therapy (or have a caregiver who can do so). In general, OPAT is not feasible for homeless patients; in such cases, OPAT in a medical respite setting may be feasible [16].

It is difficult to determine whether people who inject drugs (PWID) are suitable candidates for home-based OPAT; such decisions should be made on a case-by-case basis [17-19]. Concerns regarding OPAT in PWID include misuse of the vascular access device, increased risk for catheter complications, and drug overdose. In one study including more than 1400 patients managed with home-based OPAT, vascular access complications occurred in 9 percent of cases; the risk was increased in PWID (incidence rate ratio [IRR] 3.32, 95% CI 1.16-7.46) [20]; however, the study did not compare PWID managed with OPAT with those who received comparable inpatient therapy, so it is unclear whether this risk was attributable to OPAT.

Several small retrospective studies have demonstrated successful OPAT for PWID [21-23]. However, studies comparing outcomes for OPAT in PWID with patients who do not inject drugs have yielded variable results. In a retrospective cohort of 39 patients with current injection drug use and 117 propensity-matched controls with no current injection drug use, there were no differences in OPAT outcome; most of the patients with current injection drug use were discharged to a SNF, whereas the controls were discharged home [24]. Likewise, in a retrospective cohort study of 159 PWID and 6493 with no known injection drug use (all of whom received home-based OPAT), no difference in long-term outcomes was observed; all of the PWID had their treatments administered by visiting nurses [25]. In contrast, in a single-center urban OPAT program, line tampering, secondary bacteremia, and OPAT-related readmissions were more frequent among 101 PWID compared with 495 OPAT patients without a history of injection drug use [26]. Decisions about discharge on OPAT for PWID should also consider the safety of alternative options. Use of risk stratification tools may be helpful; in one academic medical center, use of a risk stratification questionnaire for PWID was associated with reduced length of stay without increasing readmission rate [27].

Combining OPAT with addiction care for PWID, including the use of medication-assisted treatment such as methadone or buprenorphine, may improve outcomes. In one retrospective study including more than 200 patients who received OPAT at a residential addictions treatment facility, 73 percent completed their course of antibiotic treatment at substantial cost savings to the health care system [28]. A small pilot study randomly assigned 10 PWID in need of intravenous antibiotics to receive buprenorphine and OPAT and 10 others to inpatient intravenous therapy; the OPAT with buprenorphine approach was deemed safe and reduced length of stay [29].

Antibiotic selection — In addition to usual considerations for antimicrobial selection including efficacy, toxicity, and drug-drug interactions, considerations for selection of OPAT antibiotics also include the dosing schedule and drug stability at room temperature. More than one antimicrobial agent may be required in the setting of polymicrobial infection and/or desire for synergistic therapy. The logistics of OPAT delivery may impact antimicrobial selection; as an example, patients who receive OPAT at an infusion center typically require a regimen with once-daily administration. When nonadherence is a concern, less frequent antibiotic dosing may also be preferred [30].

Some medications that require multiple daily infusions or continuous infusion therapy can be administered on an infusion pump. In such cases, the patient carries an infusion pump programmed for timed antimicrobial delivery (often in a satchel that can be worn); the antibiotic infusion cartridge needs to be changed once per day.

In selecting the OPAT regimen, antimicrobial stewardship concerns are weighed against cost, convenience, and/or tolerability [31]. Broader spectrum drugs that require less frequent administration may be preferred in the home setting. For example, serious streptococcal infections may be treated with intravenous penicillin G (which is administered continuously or up to six times per day) or ceftriaxone (which has a broader spectrum but lower cost and once-daily dosing) [32]. Likewise, once daily ertapenem is logistically easier to administer than comparable antimicrobial regimens such as cefepime with metronidazole or piperacillin-tazobactam. The relative impact of different OPAT antimicrobial regimens on future antimicrobial resistance are not well studied.

Long-acting lipoglycopeptides (dalbavancin and oritavancin) are useful agents for OPAT, given their infrequent dosing schedule and broad gram-positive spectrum. While small case series describe their use for osteomyelitis and endocarditis, thus far these medications are approved by the US Food and Drug Administration only for skin and soft tissue infections. Medication cost and insurance barriers have limited their utility for other indications. However, recently published data supports their use for more serious infections, including in patients with osteomyelitis, drug-resistant enterococcal infections, and endocarditis [33,34]. There has been some experience with these therapies in vulnerable populations including PWID [35-37], as these agents eliminate the need for long-term vascular access devices.

Vascular access — Patients receiving OPAT require intravenous access, often for weeks. Considerations for selection of a vascular access device include the drug(s) to be administered, the anticipated duration of therapy, the complications associated with different access types, and the presence of comorbidities such as renal failure. (See "Central venous access in adults: General principles" and "Central venous access: Device and site selection in adults" and "Peripheral venous access in adults".)

Vascular access devices for OPAT include (table 1):

●Peripheral intravenous catheters (PIVs) – PIVs are usually used for treatment courses <7 days. PIVs are placed in a peripheral vein at the bedside, without radiographic imaging. If the OPAT plan includes vesicants (medications associated with tissue damage caused by extravasation), PIVs should be avoided.

●Midline catheters – Midline catheters may be used for treatment courses less than two to four weeks [1,38]. Midline catheters are peripheral catheters inserted into the upper arm, with the tip ending distal to the deep veins; placement often requires ultrasound guidance. It is uncertain whether complication rates differ between midline and PICCs; observational studies have yielded conflicting findings [39-44]. However, meta-analyses have found no difference in the incidence of phlebitis or catheter-related bloodstream infection between midline catheters and PICCs [45,46]. In contrast, a meta-analysis of 12 studies comparing risk of venous thromboembolism (VTE) with midline catheters versus PICCs found increased risk of VTE with midline catheters among adults (relative risk 1.53, 95% CI 1.33-1.76) [47].

●Peripherally inserted central catheters (PICCs) – PICCs can remain in place for many weeks, provided they continue to function well. PICCs are inserted into the upper arm with the tip ending in the superior vena cava or at the cavoatrial junction; placement requires ultrasound guidance. PICCs can be used for administration of vesicant antimicrobials, given their central location. A large retrospective study suggested that infectious disease physician approval of PICC placement for intravenous antimicrobial therapy prior to placement was associated with significantly greater odds of appropriate use and lower odds of PICC-related complications compared with lack of infectious disease physician approval [48].

For patients with chronic kidney disease who may require future arteriovenous access for hemodialysis, tunneled central venous catheters are preferred over PICCs, because PICC placement is associated with increased risk of venous complications including stenosis and thrombosis [1]. In a case control study including more than 280 patients on hemodialysis, prior PICC placement was associated with a higher likelihood of lacking a functioning arteriovenous fistula (adjusted odds ratio [aOR] 2.8, 95% CI 1.5-5.5) [49].

●Other central venous catheters − Other central catheters used for OPAT include tunneled catheters and implanted ports. Implanted ports are rarely placed for OPAT alone but may be used for antimicrobial administration if already in place. In one retrospective study including more than 1460 courses of home OPAT, complication rates were observed less frequently among patients with Hickman catheters (IRR 0.20, 95% CI 0.03-0.63) and ports (IRR 0.25, 95% CI 0.04-0.78) than among patients with PICCs [20].

Among patients receiving long-term hemodialysis, if feasible, clinicians may opt for an antimicrobial that can be administered with dialysis (examples include vancomycin, cefazolin, ceftazidime, and cefepime), thereby forgoing the need for placement of an additional vascular access device.

At the time of antibiotic completion, removal of the vascular access device should be arranged, unless it is being used for another purpose.

Monitoring — Patients managed with OPAT require monitoring to ensure that antimicrobial therapy is delivered safely.

The approach to monitoring varies depending on the setting. For patients on home-based OPAT, monitoring is usually provided by visiting nurses in association with office-based clinical care. Monitoring may be performed by the staff at infusion centers, SNFs, or hemodialysis centers, with input from the supervising clinician.

Key components of OPAT monitoring include:

●Monitoring treatment response of the infection

●Prompt identification and management of adverse drug effects and drug allergy

●Therapeutic monitoring for some drugs (such as vancomycin, aminoglycosides) to ensure efficacy and avoid toxicity

●Care of the vascular access device (including dressing changes and assessment for complications including thrombosis and catheter-associated infection)

●Frequent consideration of the continued need for IV therapy, including readiness for transition to oral therapy.

Successful OPAT monitoring often requires a dedicated OPAT team, including administration for tracking laboratory studies and ensuring followup care, nursing staff to assist with side effect management, and pharmacists to monitor antimicrobial dosing, particularly with agents requiring therapeutic drug levels. Logistic challenges to effective monitoring include the lack of a dedicated OPAT team, insufficient financial and/or administrative support, the diversity of locations where patients receive OPAT, and timely communication between patient care providers and the OPAT oversight team [30,50].

The optimal frequency of clinical monitoring is variable; however, patients should be evaluated at least weekly for vascular access care. For patients at increased risk of complications, such as those with higher acuity illness and/or significant underlying comorbidities, more frequent monitoring may be needed. The frequency of monitoring may also be influenced by patient mobility and transportation limitations. Patients who may be candidates for early transition to oral therapy may also benefit from early outpatient evaluation to determine if and when completion of OPAT is appropriate.

Infectious disease outpatient follow-up care for patients receiving OPAT has been associated with a lower readmission rate [51,52]. In one study including more than 380 patients receiving OPAT, early infectious disease outpatient follow-up (within 14 days of hospital discharge) was associated with a lower risk of 30-day readmission relative to later follow-up or no infectious disease follow-up (aOR 0.33, 95% CI 0.19-0.59) [52].

Patients receiving prolonged oral antimicrobial therapy usually need warrant laboratory surveillance monitoring. Such testing may include complete blood counts and differential (to evaluate for cytopenia), liver function studies (to assess for drug-induced hepatitis), renal indices (to identify nephrotoxicity), electrolyte levels (for patients receiving beta-lactams, amphotericin, and trimethoprim-sulfamethoxazole), creatine phosphokinase (for patients receiving daptomycin), and therapeutic drug levels (for patients on vancomycin and aminoglycosides). Laboratory monitoring for patients receiving OPAT has been associated with reduced readmission rates [53,54] (see "Vancomycin: Parenteral dosing, monitoring, and adverse effects in adults" and "Dosing and administration of parenteral aminoglycosides"). While United States OPAT guidelines [1] recommend weekly monitoring for most drugs, emerging data suggests that may not always be necessary for some agents [55]. The type and frequency of laboratory monitoring is often individualized based on the antibiotic regimen, the duration of therapy, individual patient risk factors, and baseline laboratory abnormalities (table 2).

OPAT monitoring also includes timely assessments for transition to oral antimicrobial therapy. Infections for which transition to oral therapy may be appropriate include skin and soft tissue infections, bone and joint infections, intra-abdominal infections, and urinary tract infections. At times, logistic challenges, adverse drug effects, or vascular complications may also lead to an early transition to oral therapy. Transition to oral therapy is an important target of outpatient stewardship efforts and is typically associated with lower cost and toxicity. Criteria for early transition to oral therapy have not been fully established, but programs to support these efforts, often called complex outpatient antimicrobial therapy (COPAT), are being developed [56,57].

COMMONLY ENCOUNTERED PROBLEMS — Commonly encountered problems among patients receiving OPAT include adverse drug effects (ADEs) and complications associated with vascular access devices.

Adverse drug effects — Patients with ADEs may present with clinical symptoms (such as fever, diarrhea, or rash) or may have laboratory abnormalities in the absence of symptoms [58-63]. Patients with fever (≥101ºF or 38.3ºC), significant diarrhea (≥3 loose stools per day), or rash should be instructed to seek prompt clinical evaluation. For patients with temperature <101ºF in the absence of diarrhea, rash, or other symptoms, evaluation by the visiting nurse may prevent the need for a visit.

Fever may be related to the underlying infection, development of a new infection (such as catheter-associated infection or Clostridioides difficile), or be drug induced. Patients with fever and central venous access should have blood cultures collected. Further evaluation of fever should be tailored to individual patient circumstances. (See "Intravascular non-hemodialysis catheter-related infection: Clinical manifestations and diagnosis".)

Diarrhea is a common adverse effect of many antimicrobials; it may reflect C. difficile infection or a perturbation in the gastrointestinal microbiome. Patients should be counseled that they may have more frequent stools or a change in stool consistency while on antibiotics and that they should not take stool softeners or laxatives. Patients with significant diarrhea (≥3 loose stools per day) and/or abdominal pain should have stool testing for C. difficile. Empiric treatment for C. difficile is reasonable in the setting of high clinical suspicion if logistics preclude prompt stool testing. (See "Clostridioides difficile infection in adults: Clinical manifestations and diagnosis".)

Drug rash may develop during the course of antimicrobial therapy. In such cases, clinical evaluation is warranted; photographs are not always sufficient to discern whether a rash is likely to represent drug allergy. Antibiotics should be stopped immediately in the setting of any blistering reaction, rash with mucosal involvement, or rash associated with new laboratory abnormalities (such as eosinophilia, cytopenia, nephrotoxicity, or hepatotoxicity). Life-threatening cutaneous drug reactions include urticaria, drug reaction with eosinophilia and systemic symptoms, and Stevens-Johnson syndrome/toxic epidermal necrolysis [50,64]. Other cutaneous reactions include morbilliform eruptions, photosensitivity, cutaneous vasculitis, and erythema multiforme. Referral to dermatology and/or allergy can be helpful if there is uncertainty regarding whether the rash is attributable to a drug reaction. (See "New-onset urticaria" and "Drug reaction with eosinophilia and systemic symptoms (DRESS)" and "Stevens-Johnson syndrome and toxic epidermal necrolysis: Pathogenesis, clinical manifestations, and diagnosis" and "Drug eruptions".)

Drug effects detected by laboratory monitoring include eosinophilia, cytopenia, nephrotoxicity, and hepatotoxicity [65]. These are discussed further separately. (See "Approach to the patient with unexplained eosinophilia", section on 'Suspect medication' and "Drug-induced neutropenia and agranulocytosis" and "Drug-induced immune thrombocytopenia" and "Clinical manifestations and diagnosis of acute interstitial nephritis" and "Drug-induced liver injury".)

Issues related to drug hypersensitivity associated with specific antibiotic agents are discussed further separately. (See "Penicillin allergy: Delayed hypersensitivity reactions" and "Cephalosporin hypersensitivity: Clinical manifestations and diagnosis" and "Vancomycin hypersensitivity" and "Hypersensitivity reactions to fluoroquinolones" and "Sulfonamide allergy in HIV-uninfected patients".)

The decision to change the antibiotic regimen depends on a number of factors including the severity of the reaction, the likelihood that the drug is causing the toxicity, the likelihood of continued harm in the event of drug continuation, the treatment response of the infection, and the availability, efficacy, and safety of alternative agents [59,63]. In the setting of suspected drug reaction, an empiric antibiotic change is appropriate as a diagnostic trial if other causes of fever (including infection) have been excluded. (See "Drug fever".)

Vascular access complications — Complications associated with vascular access devices include infection, thrombophlebitis, and dislodgement [50,66]:

●Intravascular catheter infection usually presents with isolated fever; occasionally, erythema and or drainage at the insertion site may be observed. The diagnosis is established by blood cultures. (See "Intravascular non-hemodialysis catheter-related infection: Clinical manifestations and diagnosis".)

●Thrombophlebitis may manifest with line occlusion that does not resolve with catheter-instilled thrombolytics, palpable cord, extremity pain, and distal swelling. We have a low threshold to obtain venous ultrasound when one of these symptoms or signs is present. (See "Peripherally inserted central catheter (PICC)-related venous thrombosis in adults" and "Catheter-related upper extremity venous thrombosis in adults".)

●Dislodgement may occur inadvertently at the time of a peripherally inserted central catheter line dressing change or when snagged. Any concern about line position should prompt chest radiograph to assess position. Line malposition is a risk factor for catheter-related thrombosis [67].

Lack of adherence — Compliance with OPAT is challenging for some patients [30]. Factors that may contribute to poor adherence include cost-related issues, inability to self-administer the medications, lack of housing stability, and active substance abuse disorder [18]. These issues can often be identified prior to initiation of OPAT and, in some cases, can be mitigated with specialized local support and interventions. Adherence is not always improved with oral therapy; however, transition to oral therapy may be considered when aligned with patients' preferences.

OUTCOME MEASURES — Ideally, clinicians managing OPAT should track outcome measures including complications (rates of adverse drug effects, C. difficile infection, line complications, and readmission), as well as the rates of successful treatment response [6,7,68]. However, few clinicians have such tracking systems in place [50]. Furthermore, meaningful comparison of treatment outcomes is limited by heterogeneity of OPAT populations and lack of standardized definitions of adverse events [69].

Readmission rates are the most frequently reported OPAT outcomes; they range from 3 to 32 percent [8,51-53,60,64,69-73]. Factors associated with a higher likelihood of readmission include both systems factors and patient factors.

Systems factors associated with higher readmission rates include OPAT administration by individuals without infectious disease expertise [8], OPAT administration at a skilled nursing facility [70,73], unavailability of laboratory tests for monitoring [53], and lack of or delayed clinical follow-up visit [51,52]. At one hospital, adoption of an OPAT bundle (including a multidisciplinary infectious disease care team; optimal patient selection, education, and disposition; early OPAT follow-up appointment; standardized communication to outpatient providers; and standardized outpatient monitoring and tracking) was associated with a 50 percent reduction in readmission (odds ratio 0.51, 95% CI 0.27-0.94) [74].

Patient factors associated with higher readmission rates include increased patient age, presence of comorbidities, aminoglycoside use, treatment with more than one intravenous antimicrobial agent, presence of resistant organisms, presence of endovascular infection, and more frequent hospitalization prior to OPAT course [71,72,75,76]. In one study, the readmission rate for patients treated with cephalosporins was lower than for patients treated with other drug classes [72]. In another study including more than 380 patients receiving OPAT, complications (including adverse drug reactions and worsening infection) were the most common reason for readmission (reported in 46 percent of cases) [52]. Readmissions in patients receiving OPAT are often due to factors not directly attributable to the infection or its treatment [76].

SOCIETY GUIDELINE LINKS — Links to society and government-sponsored guidelines from selected countries and regions around the world are provided separately. (See "Society guideline links: Outpatient parenteral antimicrobial therapy".)

SUMMARY AND RECOMMENDATIONS

●Benefits and risks – Outpatient parenteral antimicrobial therapy (OPAT) has become the standard of care for treatment with intravenous antimicrobial therapy when hospitalization is not required. OPAT can reduce inpatient length of stay and is more cost-effective than hospitalization; risks include catheter-related complications and adverse drug events. (See 'Introduction' above and 'Benefits and risks' above.)

●Patient selection – Patients with infections for whom there are not equally efficacious oral antimicrobials may be considered for OPAT. Candidates for home-based OPAT must be medically stable and able and willing to self-administer intravenous antimicrobial therapy (or have a caregiver who can do so). It can be difficult to determine whether people who inject drugs are suitable candidates for home-based OPAT; such decisions should be made on a case-by-case basis. (See 'Logistics of delivery' above and 'Patient selection' above.)

●Antibiotic selection – Considerations for antimicrobial selection include efficacy, dosing schedule, toxicity, drug-drug interactions, and drug stability. Considerations for selection of a vascular access device include the drug(s) to be administered, the anticipated duration of therapy, the complications associated with different access types, and the presence of comorbidities such as renal failure. (See 'Antibiotic selection' above and 'Vascular access' above.)

●Monitoring – Patients managed with OPAT require monitoring to ensure that antimicrobial therapy is delivered safely. Patients should be evaluated at least weekly for vascular access care; more frequent monitoring may be needed for patients at increased risk of complications. The approach to laboratory monitoring should be individualized based on the antimicrobial regimen and individual patient factors (table 2) (see 'Monitoring' above). Patients should be frequently reassessed for ongoing need for antimicrobials and readiness to transition to oral therapy.

●Clinical adverse effects – Patients with adverse drug effects may present with clinical symptoms (such as fever, diarrhea, or rash) or may have laboratory abnormalities in the absence of symptoms. Patients with fever (≥101ºF or 38.3ºC), significant diarrhea (≥3 loose stools per day), or rash should be instructed to seek prompt clinical evaluation (see 'Adverse drug effects' above):

●Laboratory adverse effects – Drug effects detected by laboratory monitoring include eosinophilia, cytopenia, nephrotoxicity, and hepatotoxicity. These are discussed further separately. (See "Approach to the patient with unexplained eosinophilia", section on 'Suspect medication' and "Drug-induced neutropenia and agranulocytosis" and "Drug-induced immune thrombocytopenia" and "Clinical manifestations and diagnosis of acute interstitial nephritis" and "Drug-induced liver injury".)

●Switching antibiotics – The decision to change the antibiotic regimen depends on a number of factors including the severity of the reaction, the likelihood that the drug is causing the toxicity, the likelihood of continued harm in the event of drug continuation, the treatment response of the infection, and the availability, efficacy, and safety of alternative agents. In the setting of suspected drug reaction, an empiric antibiotic change is appropriate as a diagnostic trial if other causes of fever (including infection) have been excluded. Transition to oral therapy should be considered as appropriate. (See 'Adverse drug effects' above.)

●Vascular access complications – Complications associated with vascular access devices include infection, thrombophlebitis, and dislodgement. Intravascular catheter infection usually presents with isolated fever; the diagnosis is established by blood cultures. Thrombophlebitis may present with line occlusion, palpable cord, extremity pain, and distal swelling; in such cases, we obtain venous ultrasound. Dislodgement may occur inadvertently at the time of a peripherally inserted central catheter line dressing change or when snagged; concerns regarding line position should prompt chest radiograph to assess position. (See 'Vascular access complications' above.)