Intervenciones para tratar las septicemias relacionadas con el catéter en personas que reciben hemodiálisis de mantenimiento
Resumen
Antecedentes
Los pacientes con insuficiencia renal necesitan un acceso vascular para recibir hemodiálisis (HD) de mantenimiento, que puede conseguirse mediante una fístula arteriovenosa o un catéter venoso central (CVC). El uso de CVC está relacionado con complicaciones frecuentes como la estenosis venosa y la infección. La estenosis venosa se produce principalmente por el traumatismo causado por la entrada del catéter en la luz venosa y el contacto repetido con la pared de la vena. La biopelícula, una colonia de microorganismos irreversibles adheridos y autosuficientes integrados en una matriz autoproducida de exopolisacáridos, está asociada a la aparición de infecciones en los pacientes con catéteres permanentes. A pesar de su relevancia clínica, el tratamiento de las septicemias relacionadas con el catéter (SRC) en pacientes que reciben HD de mantenimiento sigue siendo controvertido, especialmente en lo que respecta al manejo del catéter. Las soluciones de sellado antibiótico podrían esterilizar el catéter, tratar la infección y evitar procedimientos innecesarios con el catéter. Sin embargo, este tratamiento también podría provocar una resistencia a los antibióticos o incluso un empeoramiento clínico en ciertos patógenos más virulentos. La retirada del catéter y el retraso en su sustitución podrían eliminar la fuente de infección, mejorando los desenlaces infecciosos, pero este enfoque también podría aumentar la estenosis del acceso vascular, la trombosis o ambas, o incluso el fracaso del acceso venoso central. El cambio de catéter con guía trata de eliminar la fuente de infección a la vez que mantiene el acceso a la misma vena y, por lo tanto, podría mejorar los desenlaces clínicos y conservar las venas centrales para futuros accesos.
Objetivos
Evaluar los efectos beneficiosos y perjudiciales de las diferentes intervenciones para el tratamiento de la SRC en pacientes que reciben HD de mantenimiento a través de un CVC permanente, como los antibióticos sistémicos solos o los antibióticos sistémicos combinados con soluciones de sellado o el cambio del catéter con guía o el reemplazo del catéter.
Métodos de búsqueda
Se realizaron búsquedas en el registro de estudios del Grupo Cochrane de Riñón y trasplante (Cochrane Kidney and Transplant) hasta el 21 de diciembre de 2021 mediante el contacto con el documentalista y el uso de términos de búsqueda relevantes para esta revisión. Los estudios del registro se identificaron mediante búsquedas en CENTRAL, MEDLINE y EMBASE, resúmenes de congresos, la Plataforma de registros internacionales de ensayos clínicos (ICTRP) y ClinicalTrials.gov.
Criterios de selección
Se incluyeron todos los ensayos controlados aleatorizados (ECA) y cuasialeatorizados que evaluaron el tratamiento de la SRC en CVC permanentes en personas que recibían HD de mantenimiento.
Obtención y análisis de los datos
Dos autores de la revisión de forma independiente seleccionaron los estudios para inclusión, evaluaron el riesgo de sesgo y extrajeron los datos. Los resultados se expresaron como razones de riesgos (RR) o cociente de riesgos instantáneos (CRI) para los desenlaces dicotómicos y como diferencia de medias (DM) para los desenlaces continuos, con intervalos de confianza (IC) del 95%. La certeza de la evidencia se evaluó mediante el método GRADE.
Resultados principales
Se identificaron dos ECA y un ensayo controlado cuasialeatorizado que incluyeron 760 participantes y que abordaron el tratamiento de las SRC en personas (niños y adultos) que recibieron HD de mantenimiento a través de CVC. No hubo dos estudios que compararan las mismas intervenciones. El ensayo controlado cuasialeatorizado comparó dos soluciones diferentes de sellado (activador tisular de plasminógeno [ATP] y heparina) con antibióticos sistémicos concomitantes. Un ECA comparó los antibióticos sistémicos solos y asociados con una solución de sellado de etanol, y el otro comparó antibióticos sistémicos con diferentes estrategias de manejo del catéter (cambio con guía versus la extracción y sustitución). La certeza general de la evidencia se redujo debido al bajo número de participantes, el alto riesgo de sesgo en muchos dominios, especialmente la asignación al azar, la asignación y otras fuentes de sesgo, así como a la falta de datos de desenlaces. Se desconoce si la solución de sellado de etanol utilizada con antibióticos sistémicos simultáneos mejoró la erradicación de la SRC en comparación con los antibióticos solos (RR 1,61; IC del 95%: 1,16 a 2,23) debido a que la certeza de la evidencia es muy baja. No se registraron diferencias entre los efectos del ATP y las soluciones de sellado con heparina en las tasas de curación (RR 0,92; IC del 95%: 0,74 a 1,15) ni entre el cambio de catéteres con guía versus la retirada de los catéteres con reemplazo diferido, expresado como supervivencia del catéter sin infección (CRI 0,88; IC del 95%: 0,43 a 1,79). Hasta la fecha, no hay resultados disponibles que comparen otras intervenciones.
No se informó sobre desenlaces como estenosis venosa y trombosis, resistencia de los antibióticos, muerte ni eventos adversos.
Conclusiones de los autores
En la actualidad no se dispone de evidencia de certeza alta que apoye un tratamiento sobre otro para la SRC. El beneficio de utilizar el tratamiento de sellado con etanol en combinación con antibióticos sistémicos en comparación con los antibióticos sistémicos solos para las SRC en pacientes que reciben HD de mantenimiento sigue siendo incierto debido a la certeza muy baja de la evidencia. Por ello, se necesitan más ECA para identificar los efectos beneficiosos y perjudiciales de las diferentes opciones terapéuticas para la SRC. Los estudios futuros deben unificar las definiciones de SRC y cura y mejorar el diseño metodológico.
PICO
Resumen en términos sencillos
Intervenciones para tratar las septicemias relacionadas con el catéter en personas que reciben hemodiálisis de mantenimiento
¿Cuál es el problema?
Los pacientes con insuficiencia renal necesitan un tratamiento de sustitución renal para mantenerse con vida. Entre las opciones de tratamiento, la hemodiálisis (HD) es el principal método de diálisis. Los pacientes que reciben HD a través de un catéter permanente tienen un mayor riesgo de presentar infecciones de la sangre (septicemia). Existen varias opciones para tratarlas. Entre ellas se encuentran las soluciones de sellado (la infusión de altas dosis de antibiótico en el interior de cada uno de los puertos del catéter entre las sesiones de diálisis), la retirada del catéter seguida de una nueva inserción tras la mejora clínica inicial, el cambio del catéter por uno nuevo mediante una guía (insertada a través de uno de los puertos del catéter en la misma vena, lo que permite conservar el sitio venoso) y el uso de antibióticos sistémicos (utilizados de forma aislada o combinados con otros tratamientos). Cada tratamiento tiene sus propios riesgos inherentes.
¿Qué se hizo?
Se realizaron búsquedas en el Registro especializado del Grupo Cochrane de Riñón y trasplante hasta el 21 de diciembre de 2021 y se realizó una revisión sistemática de los estudios que investigaron las opciones terapéuticas para la septicemia relacionada con el catéter en pacientes sometidos a HD.
¿Qué se encontró?
Se encontraron tres estudios pequeños, con 760 participantes, que compararon varios tratamientos para las septicemias relacionadas con el catéter. Ningún estudio comparó estrategias de tratamiento similares ni tuvo desenlaces similares y, por lo tanto, no fue posible combinar los datos en los metanálisis. Las comparaciones incluyeron antibióticos sistémicos con dos soluciones de sellado distintas, antibióticos sistémicos solos y antibióticos sistémicos más extracción del catéter comparado con antibióticos sistémicos más cambio del catéter.
Un estudio informó una mayor tasa de éxito en la eliminación de la infección con antibióticos sistémicos más tratamiento de sellado con etanol que con antibióticos sistémicos solos. Los otros estudios informaron que no hubo diferencias entre sus dos grupos de tratamiento. No se informó sobre desenlaces como estenosis venosa y trombosis, resistencia de los antibióticos, muerte ni eventos adversos.
Conclusiones
Se necesitan más estudios aleatorizados para identificar los efectos beneficiosos y perjudiciales de las infecciones sanguíneas relacionadas con el catéter.
Authors' conclusions
Summary of findings
| Systemic antibiotics plus antibiotic lock solutions plus TPA versus systemic antibiotics plus antibiotic lock solutions plus heparin | |||||
| Patient or population: patients on maintenance HD with CRBSI Settings: inpatient Intervention: systemic antibiotics plus antibiotic lock solution plus TPA Comparison: systemic antibiotics plus antibiotic lock solution plus heparin | |||||
| Outcomes | Anticipated absolute effects* (95% CI) | Relative effect | No. of Participants | Quality of the evidence | |
|---|---|---|---|---|---|
| Risk with heparin | Risk with TPA | ||||
| Cure: short‐term success Follow‐up: 2 weeks | 1000 per 1000 | 920 per 1000 (740 to 1150) | RR 0.92 (0.74 to 1.15) | 18; 24 catheters | ⊕⊝⊝⊝ |
| Cure: long‐term success Follow‐up: 6 weeks | 917 per 1000 | 587 per 1000 (348 to 972) | RR 064 (0.38 to 1.06) | 18; 24 catheters | ⊕⊝⊝⊝ very low 1,2 |
| Cure: infection‐free survival Follow‐up: between the final dose of antibiotics and the first positive blood culture obtained from the catheter | Infection‐free survival was 27.7 days less with TPA (88.68 days less to 33.28 days more) than with heparin | ‐ | 18; 24 catheters | ⊕⊝⊝⊝ very low 1,2 | |
| Stenosis or thrombosis of vascular access site after catheter removal | Not reported | Not reported | ‐ | ‐ | ‐ |
| Death | Not reported | Not reported | ‐ | ‐ | ‐ |
| Development of antibiotic resistance | Not reported | Not reported | ‐ | ‐ | ‐ |
| Adverse effects | Not reported | Not reported | ‐ | ‐ | ‐ |
| *The risk in the intervention group (and its 95% CI) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI). | |||||
| GRADE Working Group grades of evidence | |||||
| 1High risk of bias in random sequence generation and allocation concealment (selection bias), blinding of participants and personnel (performance bias), inconsistencies regarding the description of the data analysis and the tables, protocol break and also conflicting information between the author's reply and the article (other bias); downgraded by 2 levels (methodological limitation) 2Small number of participants and studies (doubt about the reproducibility of the data); downgraded by 1 level (imprecision) | |||||
| Systemic antibiotics alone versus systematic antibiotics plus antibiotic lock solution | |||||
| Patient or population: patients on maintenance HD with CRBSI Settings: inpatient Intervention: systemic antibiotics alone Comparison: systemic antibiotics and antibiotic lock solution (ethanol) | |||||
| Outcomes | Anticipated absolute effects* (95% CI) | Relative effect | No. of participants | Quality of the evidence | |
|---|---|---|---|---|---|
| Risk with systemic antibiotics alone | Risk with systemic antibiotic plus | ||||
| Cure: successful eradication of the infection Follow‐up: 2 days | 563 per 1000
| 906 per 1000 (653 to 1000) | RR 1.61 (1.16 to 2.23) | 64 (1) | ⊕⊝⊝⊝ 1,2,3 |
| Stenosis or thrombosis of vascular access site after catheter removal | Not reported | Not reported | ‐ | ‐ | ‐ |
| Death | Not reported | Not reported | ‐ | ‐ | ‐ |
| Development of antibiotic resistance | Not reported | Not reported | ‐ | ‐ | ‐ |
| Adverse effects | Not reported | Not reported | ‐ | ‐ | ‐ |
| *The risk in the intervention group (and its 95% CI) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI). | |||||
| GRADE Working Group grades of evidence | |||||
| 1High risk of bias in selective reporting because of retrospective trial registration (reporting bias) and other bias once 45% of the included population have negative blood cultures; downgraded by 1 level (methodological limitation). 2Small number of participants and studies (doubt about the reproducibility of the data); downgraded by 1 level (imprecision). 3Outcome cure defined only by clinical improvement in signs and symptoms, without laboratory confirmation; downgraded by 1 level (indirectness). | |||||
| Catheter guidewire exchange versus catheter removal and replacement | |||||
| Patient or population: patients on maintenance HD with CRBSI Settings: inpatient Intervention: systemic antibiotics and catheter guidewire exchange Comparison: systemic antibiotics and catheter removal and replacement | |||||
| Outcomes | Anticipated absolute effects* (95% CI) | Relative effect | No. of participants | Quality of the evidence | |
|---|---|---|---|---|---|
| Risk with systemic catheter removal | Risk with guidewire exchange | ||||
| Cure: infection‐free survival Follow‐up: 45 days | 720 per 1000 | 749 per 1000 (555 to 868) | HR 0.881 (0.43 to 1.79) | 678 (1) | ⊕⊝⊝⊝ 2,3,4 |
| Stenosis or thrombosis of vascular access site after catheter removal | Not reported | Not reported | ‐ | ‐ | ‐ |
| Death | Not reported
| Not reported | ‐ | ‐ | ‐ |
| Development of antibiotic resistance | Not reported | Not reported | ‐ | ‐ | ‐ |
| Adverse effects | Not reported | Not reported | ‐ | ‐ | ‐ |
| *The risk in the intervention group (and its 95% CI) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI). | |||||
| GRADE Working Group grades of evidence | |||||
| 1Outcome cure was measured as a time‐to‐event outcome. Comparator group risk of event‐free survival at 90 days. Baseline risk line of 0.72 comes from the study's Kaplan‐Meier survival curves at 90 days time point, which represents the middle of the follow‐up 2High risk of bias in blinding of participants and personnel (performance bias), incomplete outcome data (attrition bias), selective reporting because of retrospective trial registration (reporting bias) and other bias; downgraded by 1 level (methodological limitation). 3Small number of participants and studies (doubt about the reproducibility of the data); downgraded by 1 level (imprecision) 4Use of catheter infection‐free survival as an indication of cure; downgraded by 1 level (indirectness) | |||||
Background
Description of the condition
Patients with kidney failure require kidney replacement therapy (KRT) to sustain life. Among KRT options, haemodialysis (HD) is the primary method of dialysis used in 79% of countries with a frequency of 80% (Saran 2019). The creation and maintenance of a point of access into the patient’s vascular system are crucial for achieving safe and effective HD. In their latest guidelines, the National Kidney Foundation’s Dialysis Outcomes Quality Initiative (NKF‐DOQI) and the European Society of Vascular Surgeons (ESVS) recommend the use of an arteriovenous fistula (AVF) as the primary option for achieving vascular access for HD. However, there are situations in which the use of an AVF is not feasible, either because of the patient’s limited life expectancy or their comorbidities (e.g. congestive heart failure or atherosclerosis with ischaemia), AVF maturation time, the absence of venous outflow or reduced venous diameter. In such cases, the use of a tunnelled and cuffed central venous catheter (CVC) for vascular access is acceptable. Additionally, compared to AVF, tunnelled and cuffed CVC has benefits such as needle‐free connection to the dialysis circuit, the possibility of immediate use, the elimination of cannulation‐related complications, and the simplicity of device implantation.
However, patients receiving maintenance HD through CVC frequently present with clinically relevant complications, such as venous thrombosis, central venous stenosis, and especially infection (Liangos 2006). Sepsis in these patients is related to a higher occurrence of cardiovascular events and lead to a higher death rate (Foley 2004). Catheter‐related infections (CRI) (also termed bacteraemia or catheter‐related bloodstream infection (CRBSI)) may occur at the exit site, at the subcutaneous tunnel, or in the bloodstream.
The clinical features of CRBSI might not always be present. The most sensitive. although nonspecific manifestations. are fever and chills, while others, such as haemodynamic instability, catheter dysfunction, hypothermia, nausea and vomiting, and generalized malaise may also occur. Especially for surveillance purposes, any patient with a HD catheter and signs and symptoms of bloodstream infection without an alternate source of infection is suspected to have CRBSI (CDC 2018 Dialysis Event Surveillance Protocol). Although there is no consensus related to CRBSI definition, most guidelines and associations require laboratory confirmation (Mermel 2009; O'Grady 2011; Vascular Access 2006 Work Group 2006). In this systematic review, definite CRBSI is defined as suspected CRBSI with concurrent positive blood cultures of the same organism from the catheter and a peripheral vein. The blood from a peripheral source can be obtained from a vein or the dialysis circuit. The colony count is considered positive if there are > 15 colony forming units (CFU)/catheter segment (in the hub or tip) in semi‐quantitative culture or > 100 CFU/catheter segment (hub or tip) in quantitative culture. In addition, the differential time to positivity of two hours, which is the differential period of catheter culture versus peripheral blood culture, is considered positive. If available, simultaneous quantitative cultures of blood samples with a ratio of ≥ 3:1 (catheter hub/tip versus peripheral dialysis circuit/vein) would be supportive.
Description of the intervention
Patients receiving maintenance HD through CVC have increased susceptibility to infections as a result of the acquired immunodeficiency status, a characteristic of kidney failure (due to uraemia, advanced age and associated comorbidities such as diabetes mellitus), the high bacterial virulence (related to biofilm formation), and the frequent and repeated exposure to risk factors inherent in the typical HD process (Sarnak 2000).
Several pathogens can cause CRBSI. Coagulase‐negative staphylococci and Staphylococcus aureus are the most frequent causes and account for 40% to 80% of cases (Allon 2004). Non‐staphylococcal infections are mostly due to enterococci and gram‐negative bacteria. Up to 12% to 38% of CRBSIs are caused by methicillin‐resistant S. aureus (MRSA) (Lok 2011). Although less common, fungal infection can also occur and, in 7% to 21% of cases, multiple organisms are responsible, making treatment even more challenging.
The severity of the infection is closely linked to the type of organism present. Serious metastatic infections, including osteomyelitis, thrombophlebitis, endocarditis, septic arthritis, epidural abscess, and large atrial thrombi occur more frequently due to S. aureus. S. aureus is also commonly associated with significant morbidity and death rates, with some series showing a three‐fold higher risk of infectious complications and a 20% higher death rate compared to all other pathogens (Allon 2004). MRSA, however, has been associated with three to five times higher death rates compared with methicillin‐sensitive strains (Beigi 2010).
The isolation of the pathogen also indicates the treatment required. Although the use of systemic empirical antibiotic therapy is similar at the beginning of treatment, tailoring might be necessary following the microbiological results. In addition, some guidelines suggest different catheter management depending on each aetiological agent (Mermel 2009; Vascular Access 2006 Work Group 2006).
Since the use of the Dacron cuff leads to a local inflammatory response and is thought to create a mechanical barrier avoiding bacterial migration from the skin, biofilm formation is the presumed source of CRI (Valliant 2014). As such, CRBSI treatment should also address catheter sterilisation, which can be achieved through using antibiotic lock solutions or by the removal of the CVC, either over a guidewire in the same vein or with replacement in a new vein. Although there is evidence of inferior rates of cure through the use of systemic antibiotics alone, it remains an intervention option in many centres (Aslam 2014).
Antibiotic lock solutions are high concentrations (approximately 100‐fold higher than those used systemically) of antibiotic or antimicrobial solutions in small volumes with or without an anticoagulant. This intervention is adjunctive to systemic antibiotic therapy and consists of catheter salvage with the instillation of a solution into each catheter lumen at the end of each dialysis session for the same duration as the systemic therapy. Common antibiotic lock solutions contain gentamicin, cefazolin, cefotaxime, vancomycin, linezolid, vancomycin plus gentamicin, cefazolin plus gentamicin, trimethoprim plus sulfamethoxazole, or minocycline. Antimicrobial lock solutions may contain taurolidine, 30% citrate, ethanol, EDTA, or methylene blue. The anticoagulants typically used in collaboration with lock treatment are heparin, citrate or tissue plasminogen activator (TPA) (Justo 2014).
CVC removal with CVC exchange over a guidewire at the same site is a single procedure performed by inserting a guidewire through the infected catheter lumen to the accessed vein, followed by CVC removal and placement of a new CVC over the guidewire, therefore, maintaining the same access site (i.e. the same vein) (Duszak 1998).
CVC removal with a new CVC placed at a new site involves the removal of the infected catheter and the placement of a new HD catheter, which can be done in one procedure by positioning a new CVC for HD, or in two separate procedures: one to place a CVC that is neither tunnelled nor cuffed and a second to exchange it for a new tunnelled and cuffed CVC. Additionally, it may or may not include a CVC‐free duration (i.e. a period when there is no CVC in situ) (Lok 2019).
How the intervention might work
Biofilm is associated with the development of infections in patients with indwelling catheters. Within 24 hours of catheter insertion, all vascular devices show micro‐organism colonization. The process of biofilm formation is of irreversible adhesion of micro‐organisms to a surface such as medical devices (e.g. venous and urinary catheters); the micro‐organisms can multiply slowly and create a self‐sufficient flora protected by a self‐produced matrix of exopolysaccharides. Unlike planktonic micro‐organisms, biofilm flora have a lower sensitivity to antibiotics, requiring high concentrations for its eradication (Donlan 2001).
The systematic review by Arechabala 2018 showed that the use of antibiotics and combined antimicrobial lock solutions reduced the risk of CRI compared to the use of control lock solutions (usually containing only an anticoagulant). Benefits were also observed in tunnelled HD CVC. Thus, the use of antibiotic lock solutions to treat CRI represents a logical strategy.
Central venous stenosis is a frequent complication among patients receiving maintenance HD (MacRae 2005). The trauma caused by the entrance of the catheter into the venous lumen, the torque of the CVC over the distal third of the subclavian vein at the first rib topography, and its repeated contact with the vein wall, aggravated by the cardiac cycle, are contributing factors to the development of venous stenosis (Liangos 2006).
Patients with venous stenosis requiring CVC removal may develop thrombosis and occlusion of the venous access. In addition to the risk of developing superior vena cava stenosis and superior vena cava syndrome (facial oedema and dyspnoea), central venous occlusion reduces the chances of posterior vascular access due to failure of AVF maturation, inadequate flow or the impossibility of new CVC (MacRae 2005).
Therefore, the treatment of CRI should include the caveat of trying to preserve venous sites. The passage of a guidewire through the CVC ensures access to the vein lumen, allowing the passage of a new device in the same place without thrombosis and removing the infection and its source altogether.
Why it is important to do this review
Infection is a major cause of CVC‐related morbidity and death, with an incidence of one to six episodes of CRI/1000 catheter‐days (Lok 2011; Ravani 2013; Saad 1999; Tanriover 2000). Infection is the second highest cause of death among patients receiving HD, behind only cardiovascular causes (Perl 2011; Saran 2018). Moreover, the occurrence of sepsis in patients with kidney failure results in higher death than in the general population (Sarnak 2000) and can be considered a harbinger of cardiovascular events (Foley 2004; Ishani 2005). Additionally, CVC carries a 1.5‐ to three‐fold increased risk of death from infectious causes and a two‐ to three‐fold greater risk of death from all causes (Polkinghorne 2004).
By the end of 2016, 63.1% of patients with prevalent kidney failure in the USA underwent HD. Four out of five of these patients started HD through CVC, and up to three months later, more than two‐thirds of them had maintained the CVC as vascular access. Approximately 39% of the AVFs created between June 2014 and May 2016 in the USA had maturation failure (Saran 2018). And so, despite the risks of catheter use and the policies to encourage fistulae creation, the use of tunnelled and cuffed CVC has increased in many countries (Rayner 2010).
The ideal CRBSI treatment should be able to resolve the potentially life‐threatening infection, without leading to antibiotic resistance or the development of venous stenosis or occlusion. This systematic review aimed to evaluate the current state of the evidence on CRBSI management, as this information is essential for vascular surgeons, nephrologists and interventional radiologists who can treat CRBSI and patients with kidney failure for whom the presence of a viable access site is a lifeline.
Objectives
This review aimed to look at the benefits and harms of the different interventions proposed for bloodstream infections in permanent CVC for patients on maintenance HD, such as systemic antibiotics alone or systemic antibiotics combined with either lock solutions or catheter guidewire exchange or catheter replacement.
Methods
Criteria for considering studies for this review
Types of studies
We included all randomised controlled trials (RCTs) and quasi‐RCTs (RCTs in which allocation to treatment was obtained by alternation, use of alternate medical records, date of birth or other predictable methods) looking at the management of CRBSI in permanent CVCs in people receiving maintenance HD.
Types of participants
Inclusion criteria
Patients with kidney failure on maintenance HD by tunnelled and cuffed CVC with CRBSI.
Exclusion criteria
-
Patients with kidney failure on maintenance HD using subcutaneously tunnelled and cuffed CVC with exit site infections or tunnel infections without associated CRBSI
-
Patients with kidney failure on maintenance HD using subcutaneously tunnelled and cuffed CVC with other sources of infection
-
Patients with kidney failure on other types of KRT
-
Patients on HD with CVC infection but with acute kidney injury
-
Patients with kidney failure on maintenance HD with short‐term CVC use.
Types of interventions
All types of interventions compared included systemic antibiotic use, either used alone or in combination with a catheter management option.
Planned comparisons
-
Systemic antibiotic alone (any substance, dose or administration) versus catheter guidewire exchange
-
Systemic antibiotic alone (any substance, dose or administration) versus catheter removal and replacement
-
Systemic antibiotic alone (any substance, dose or administration) versus lock treatment (any substance, preparation, dose or administration)
-
Catheter guidewire exchange versus catheter removal and replacement
-
Catheter guidewire exchange versus catheter retainment and lock treatment (any substance, preparation, dose or administration)
-
Catheter removal replacement versus catheter retainment and lock treatment (any substance, preparation, dose or administration)
-
Catheter removal and replacement, single procedure (new tunnelled and cuffed CVC) versus catheter removal and replacement, two different procedures (non tunnelled CVC followed by tunnelled and cuffed CVC)
-
Catheter retainment and lock treatment (any substance, preparation, dose or administration) versus catheter retainment and other lock treatment (different substance, preparation, dose or administration).
Types of outcome measures
This review did not exclude studies based on the non‐reporting of outcomes of interest.
The outcomes selected included the relevant SONG core outcome sets as specified by the Standardized Outcomes in Nephrology initiative (SONG 2017).
Primary outcomes
-
Cure: clinical resolution associated with negative blood cultures at least 72 hours after completion of systemic antibiotic treatment
-
Stenosis or thrombosis of vascular access site after catheter removal: narrowing or occlusion of a central vein used for tunnelled and cuffed CVC insertion (e.g. internal jugular vein, subclavian vein, femoral vein).
Secondary outcomes
-
Death
-
Catheter malfunction (failure to attain and maintain an extracorporeal blood flow ≥ 300 mL/min in an adult patient to perform HD without significantly lengthening the HD treatment) (Vascular Access 2006 Work Group 2006)
-
Duration of hospitalisation
-
Development of antibiotic resistance
-
Fatigue
-
Cardiovascular disease
-
Adverse outcomes, such as bleeding (major or minor), allergic reactions, urticaria, anaphylaxis, or other adverse effects defined by the study authors.
Search methods for identification of studies
Electronic searches
We searched the Cochrane Kidney and Transplant Register of Studies up to 21 December 2021 through contact with the Information Specialist using search terms relevant to this review. The Register contains studies identified from the following sources.
-
Monthly searches of the Cochrane Central Register of Controlled Trials (CENTRAL)
-
Weekly searches of MEDLINE OVID SP
-
Searches of kidney and transplant journals, and the proceedings and abstracts from major kidney and transplant conferences
-
Searching of the current year of EMBASE OVID SP
-
Weekly current awareness alerts for selected kidney and transplant journals
-
Searches of the International Clinical Trials Register (ICTRP) Search Portal and ClinicalTrials.gov.
Studies contained in the Register are identified through searches of CENTRAL, MEDLINE, and EMBASE based on the scope of Cochrane Kidney and Transplant. Details of search strategies, as well as a list of handsearched journals, conference proceedings and current awareness alerts, are available on the Cochrane Kidney and Transplant website under CKT Register of Studies
See Appendix 1 for search terms used in strategies for this review.
Searching other resources
-
Reference lists of review articles, relevant studies, and clinical practice guidelines.
-
Contacting relevant individuals/organizations seeking information about unpublished or incomplete studies.
-
Grey literature sources (e.g. abstracts, dissertations, and theses), in addition to those already included in the Cochrane Kidney and Transplant Register of Studies, were not searched.
Data collection and analysis
Selection of studies
The search strategy described was used to obtain the titles and abstracts of studies potentially relevant to the review. The titles and abstracts were screened independently by two authors (BMA, DGC) and discarded when not applicable; studies and reviews that could include relevant data were initially retained. We independently assessed the retrieved abstracts and, when necessary, the full text of these studies to determine which ones satisfied our inclusion criteria. Disagreements were resolved in consultation with a third author (DHM). If a study did not meet our inclusion criteria, it was excluded and the reasons documented.
Data extraction and management
Two authors (BMA, DGC) independently carried out data extraction using standard data extraction forms. Data extraction included characteristics of the population, the intervention described, the outcomes assessed, and the duration of the follow‐up. Disagreements were resolved in consultation with a third author (DHM). Studies reported in non‐English language journals were to be translated before assessment. Where more than one publication of a study existed, reports were grouped together and the publication with the most complete data was used in the analyses. Where relevant outcomes were only published in earlier versions these data were used. Any discrepancies between published versions were to be highlighted.
Assessment of risk of bias in included studies
The following items were independently assessed by two authors (BMA, DGC) using the risk of bias assessment tool (Higgins 2020) (see Appendix 2).
-
Was there adequate sequence generation (selection bias)?
-
Was allocation adequately concealed (selection bias)?
-
Was knowledge of the allocated interventions adequately prevented during the study?
-
Participants and personnel (performance bias)
-
Outcome assessors (detection bias)
-
-
Were incomplete outcome data adequately addressed (attrition bias)?
-
Are reports of the study free of suggestion of selective outcome reporting (reporting bias)?
-
Was the study apparently free of other problems that could put it at risk of bias?
Measures of treatment effect
We used the risk ratio (RR) for dichotomous outcomes (e.g. death, cure, stenosis or thrombosis of vascular access site after catheter removal, catheter malfunction outcomes), with 95% confidence intervals (CI). Where continuous scales of measurement were used to assess the effects of treatment (e.g. duration of hospitalisation, fatigue, cardiovascular disease, development of antibiotic resistance), we planned to use the mean difference (MD); and if different scales were used, the standardised mean difference (SMD).
We planned to tabulate and assess the adverse effects with descriptive techniques.
Unit of analysis issues
Outcomes were analysed at the individual patient level. If the unit of randomisation was not the same as the level of analysis, that is, the patient, adjustments were made to address the potential impact of clustering on the outcome.
Dealing with missing data
We included all available data and requested further information twice from the original authors by written correspondence (emailing corresponding author/s) and obtained a response from two. The relevant information obtained was included in the review. Evaluation of important numerical data such as screened, randomised patients, as well as intention‐to‐treat, as‐treated, and per‐protocol population, were carefully performed. Attrition rates such as drop‐outs, losses to follow‐up, and withdrawals were investigated. Issues of missing data and imputation methods (for example, last‐observation‐carried‐forward) were critically appraised (Higgins 2020).
Assessment of heterogeneity
The heterogeneity was first assessed by visual inspection of the forest plot, and after quantified by statistical heterogeneity using the I² statistic, which describes the percentage of total variation across studies that is due to heterogeneity rather than sampling error (Higgins 2003). A guide to the interpretation of I² values was used as follows.
-
0% to 30%: might not be important
-
30% to 50%: may represent moderate heterogeneity
-
50% to 75%: may represent substantial heterogeneity
-
75% to 100%: considerable heterogeneity.
The importance of the observed value of I² depends on the magnitude and direction of treatment effects and the strength of the evidence for heterogeneity (e.g. P‐value from the Chi² test, or a 95% CI for I²) (Higgins 2020).
Assessment of reporting biases
We planned to use funnel plots to assess the potential existence of publication bias and other bias reports if sufficient studies (more than 10) were identified for inclusion in the meta‐analysis (Higgins 2020).
Data synthesis
Data were intended to be pooled using the random‐effects model, but the fixed‐effect model would also have been used to ensure the robustness of the model chosen and susceptibility to outliers.
Subgroup analysis and investigation of heterogeneity
If sufficient data were available, we had intended to perform subgroup analysis to explore possible sources of heterogeneity, which were to include the following.
-
First versus multiple treatments and catheter salvage attempts
-
Initial catheter access site: either jugular, subclavian, or femoral
-
Presence of other CRI in association with CRBSI, such as exit‐site infection, infection of the tunnel track or CRBSI alone
-
Inpatient versus outpatient
-
Previous use or not of prophylactic antibiotic lock solutions
-
Aetiological agent of the infection (such as gram positive, gram negative, anaerobes, fungi).
Heterogeneity among participants could be related to age and renal pathology (gender, ethnicity, diabetics versus non‐diabetics, smoking, vascular disease, body mass index, age category, dialysis vintage and duration of catheter use before development of CRI). Heterogeneity in treatments could be related to a patient’s access site conditions (patients with femoral tunnelled and cuffed CVC usually have poor vascular access prognosis), to the placement or not of a not tunnelled CVC prior to a tunnelled and cuffed CVC at CVC removal, prior agent(s) used, and the agent, dose and duration of therapy (such as the lock solution substance, dose, preparation, and administration). Statistical heterogeneity in study design and risk of bias was evaluated.
Sensitivity analysis
We had planned to perform sensitivity analyses to explore the influence of the following factors on effect size.
-
Repeating the analysis excluding unpublished studies
-
Repeating the analysis considering the risk of bias, as specified
-
Repeating the analysis excluding any very long or large studies to establish how much they dominate the results
-
Repeating the analysis excluding studies using the following filters: diagnostic criteria, language of publication, source of funding (industry versus other), and country.
Summary of findings and assessment of the certainty of the evidence
We have presented the main results of the review in the 'Summary of findings' tables. These tables present key information concerning the certainty of the evidence, the magnitude of the effects of the interventions examined, and the sum of the available data for the main outcomes (Schunemann 2020a). The 'Summary of findings' tables also include an overall grading of the evidence related to each of the main outcomes using the GRADE (Grades of Recommendation, Assessment, Development and Evaluation) approach (GRADE 2008; GRADE 2011). The GRADE approach defines the certainty of a body of evidence as to the extent to which one can be confident that an estimate of effect or association is close to the true quantity of specific interest. This approach will be assessed by two authors. The certainty of a body of evidence involves consideration of within‐trial risk of bias (methodological quality), directness of evidence, heterogeneity, precision of effect estimates and risk of publication bias (Schunemann 2020b). The following outcomes were intended to be presented in the 'Summary of findings' tables.
-
Cure
-
Stenosis or thrombosis of vascular access site after catheter removal
-
Death
-
Development of antibiotic resistance
-
Adverse effects.
Results
Description of studies
See: Characteristics of included studies, Characteristics of excluded studies and Characteristics of studies awaiting classification
Results of the search
After searching the Specialised Register, a total of 25 records were identified. After screening the titles and abstracts and then full‐text review, three studies (five reports) were included and 13 studies (19 reports) were excluded. One study (Khosroshahi 2006b) is as awaiting assessment (Figure 1).
Study flow diagram.
Included studies
We identified two RCTs (Khosroshahi 2015; Saleh 2017) and one quasi‐RCT (Onder 2008) (760 participants). See Characteristics of included studies.
-
Khosroshahi 2015 enrolled 64 patients with kidney failure on maintenance HD with suspected CRBSI. This single‐centre RCT compared an ethanol lock solution plus concomitant systemic antibiotics with systemic antibiotics alone to treat CRBSIs. The outcome measured was the successful eradication of the infection.
-
Onder 2008 enrolled 18 children with kidney failure on maintenance HD (a total of 24 catheters) with suspected CRBSI. This study compared two different anticoagulants (TPA and heparin) used with an antibiotic lock solution with concomitant systemic antibiotics to treat CRBSIs. The investigation was a single‐centre quasi‐RCT that evaluated successful eradication of infection, recurrence of infection, and infection‐free catheter survival as outcomes.
-
Saleh 2017 compared two different catheter management strategies (catheter exchange over a guidewire to catheter removal and replacement) with concomitant systemic antibiotics to treat CRBSIs. This multicentre RCT analysed 678 patients with kidney failure and suspected CRBSI for cure, treatment failure, and indeterminate outcomes.
No two studies compared the same interventions, so a meta‐analysis could not be undertaken. Additionally, the studies applied different definitions for confirmed CRBSIs and for outcomes such as cure or eradication of infection.
Excluded studies
Thirteen studies were excluded.
-
Eight studies investigated the prevention of CRBSI rather than its treatment (Bosma 2010; ELVIS 2015; Harris 1997; Hymes 2017; Levin 1989; Oliver 2007; Rosenblum 2014; Zwiech 2016).
-
Abdel Azim 2018 only included patients with kidney failure and non‐cuffed and non‐tunnelled CVC.
-
Three studies were abandoned early due to inadequate enrolment (NCT01483872; NCT02040818), one of which also did not evaluate the interventions of this review (NCT00108433)
-
Dahlberg 1986 included the wrong population and compared the incidence of CRBSI and not its treatment.
For individual details of the excluded studies, see Characteristics of excluded studies.
Studies awaiting classification
Khosroshahi 2006b was published as an abstract over 10 years ago and both its full‐text and raw data are not available according to the author.
Risk of bias in included studies
The 'Risk of bias' assessments for all included studies are shown in Figure 2 and Figure 3. Overall, most studies were classified as having a high risk of bias due to methodological limitations and poorly reported results.
Risk of bias graph: review authors' judgements about each risk of bias item presented as percentages across all included studies.
Risk of bias summary: review authors' judgements about each risk of bias item for each included study.
Allocation
Random sequence generation
Two studies were judged to be at low risk of bias. Khosroshahi 2015 used a computer‐generated list, and Saleh 2017 used a sealed envelope service to provide the randomised sequence. Onder 2008 was judged to be at high risk because although it used sealed envelopes for the first 18 patients, six recurrences were further included as a non‐planned cross‐over and were not randomised.
Allocation concealment
Two studies were judged to be at low risk of bias. Khosroshahi 2015 responded by correspondence that the randomised list was provided to the HD centre nurse, and the randomisation service used by Saleh 2017 also provided allocation concealment. Onder 2008 was judged to be at high risk of bias. The envelopes were not described as being opaque, and the allocation remains unclear because the information provided by the article conflicted with the author's response.
Blinding
Khosroshahi 2015 reported that the practitioner who evaluated the outcome was unaware of the patient’s treatment, but the researchers failed to report how and if the participants were blinded. We thus classified the study as having an unclear risk for performance bias and a low risk of detection bias.
Onder 2008 described the study as randomised, prospective and non‐blinded. We contacted the author and received only one reply with conflicting data. The masking was said to have included the patient, the practitioner, and the observer, but while describing the allocation, it was said that the sealed envelope was given to the patient’s care nurse – which disabled blinding of providers. Due to unreliable information, we classified the performance bias as high risk. Because the outcomes were direct and objective, although there were the mentioned discrepancies, we considered that the assessment was not affected and thus classified the detection bias as low risk.
Saleh 2017 compared two different surgical techniques for which it was impossible to mask the personnel and the patient. Therefore, the performance bias was classified as high risk. The article did not describe the detection bias, but the clinical study’s retrospective registration points out that there was single masking for the care provider, which probably concerns the outcome assessment observer. Consequently, we classified it as having a low risk for detection bias.
Incomplete outcome data
In Saleh 2017, participants consisted of those with suspected CRBSIs and positive blood cultures from both the catheter and a peripheral vein. The authors described that negative catheter tips were an exclusion criterion. There is no report of how many patients initially diagnosed with CRBSI were included, randomised, but had further negative catheter tips and were then excluded from the study. There is also no mention of withdrawals or deaths, or descriptions of which groups these patients belonged to. These facts led to a high risk for attrition bias in this study.
Onder 2008 did not describe if there were losses during the follow‐up. For this reason, we classified the study as having an unclear risk for this domain.
Khosroshahi 2015 replied all participants completed the treatment and the follow‐up. Therefore, the study was classified as having a low risk for attrition bias.
Selective reporting
Onder 2008 had an unclear risk of bias for selective reporting because although the authors reported the planned outcomes, there was no published trial protocol.
Both Khosroshahi 2015 and Saleh 2017 had retrospective trial registration, and neither study thoroughly reported the outcomes. Therefore, we classified both studies as having a high risk of reporting bias.
Other potential sources of bias
There were differences regarding the description of the data analysis and the tables, protocol breaks, and conflicting information between the author's reply and the published article by Onder 2008. Therefore we classified this study as having a high risk of other potential sources of bias.
Khosroshahi 2015 was judged to be at high risk of bias. The study considered all suspected CRBSIs as their included population, without the need for laboratory confirmation, which resulted in 45% of unconfirmed cases (negative blood cultures). Although the authors did not exclude patients with negative blood cultures, they failed to describe or compare which groups these patients belonged to, which could yield misleading conclusions.
Saleh 2017 reported that 65% of the incident and 41% of the prevalent patients on maintenance HD use catheters as vascular access in their centres and that this is a substantially higher rate than those observed in America and Europe. Although these peculiarities influence the study's external validity, they do not alter its risk of bias. However, patients whose catheter was not replaced within 10 days because of death were among the exclusion criteria. This exclusion criterion could be concealing a critical adverse event for this intervention. Hence, this study was classified as high risk for other potential sources of bias.
Effects of interventions
See: Summary of findings 1 Systemic antibiotics plus antibiotic lock solutions plus tissue plasminogen activator versus systemic antibiotics plus antibiotic lock solutions plus heparin; Summary of findings 2 Systemic antibiotics alone versus systematic antibiotics plus antibiotic lock solution; Summary of findings 3 Catheter guidewire exchange versus catheter removal and replacement
Antibiotic lock treatment plus tissue plasminogen activator versus antibiotic lock treatment plus heparin
Onder 2008 compared different antibiotic lock treatments in association with systemic antibiotics. Eighteen children with confirmed infection of tunnelled and cuffed HD catheters were included. The patients were all treated with systemic antibiotics (initially empiric and tailored afterwards if needed) and randomised into two groups based on the type of antibiotic lock solution used, which was a mixture of an antibiotic (tobramycin or vancomycin, depending on the blood culture) with either TPA or heparin. Thirty‐five patients with suspected catheter‐related bacteraemia entered the study with only 24 completions due to laboratory confirmations from positive blood cultures (69%). Additionally, a significant proportion of the participants were not randomised (6/18 patients had recurrences and underwent an unplanned cross‐over treatment). The unit of analysis was the episode of catheter‐related bacteraemia. The primary outcomes were short‐ and long‐term success, defined as clearance of infection at the end of two weeks and infection‐free status at the end of six weeks, respectively. Blood cultures were collected during the two weeks of maintained systemic antibiotic treatment, and no further surveillance cultures were obtained in asymptomatic children. Thus, these outcomes cannot be included as the cure definition for this review. The study’s authors reported that 23 catheters were cleared of infection after two weeks of treatment and that one was cleared after four weeks (because of antifungal treatment) (Analysis 1.1 (24 catheters): RR 0.92, 95% CI 0.74 to 1.15); the long‐term success was 58% (7 catheters) for the TPA group and 91% (11 catheters) for the heparin group (Analysis 1.2).
The secondary outcomes of this study were infection‐free catheter survival (defined as the time period between the final dose of antibiotics), the first subsequent positive blood culture obtained from the catheter, and overall catheter survival, with no reported definition. The mean infection‐free survival was 126.8 ± 81.6 days for the TPA group and 154.5 ± 70.4 days for the heparin group (Analysis 1.3). Overall, Onder 2008 did not demonstrate a difference in short‐ or long‐term success or infection‐free survival between TPA and heparin.
Systemic antibiotics alone versus systemic antibiotics plus antibiotic lock solution treatment
Khosroshahi 2015 compared systemic antibiotics alone versus systemic antibiotics with concurrent antibiotic lock solution, randomising 64 patients with kidney failure and suspected infections of tunnelled and cuffed HD catheters. All participants were initially treated with empiric antibiotics (vancomycin and third‐generation cephalosporin) further modified by blood culture results. Participants were followed for three weeks. Laboratory confirmation of CRBSI was obtained thereafter, leading to 45% of the included patients presenting negative blood cultures, and, therefore, outside this review’s inclusion criteria. The primary outcome measured was successful eradication of infection, defined as clinical improvement in signs and symptoms within 48 hours of initiating treatment, which does not comply with this review’s definition of cure. Overall, the study reported a success rate for clearing infection of 90.6% (29 patients) in the systemic antibiotic and ethanol lock group and 56.2% (18 patients) in the systemic antibiotics alone group (P = 0.002) (Analysis 2.1).
Catheter guidewire exchange versus catheter removal and replacement
Saleh 2017 compared over‐the‐wire catheter exchange with catheter removal and late replacement (3 to 7 days later), randomising 678 with kidney failure and suspected infection of tunnelled and cuffed HD catheters. Concurrent positive blood cultures from the catheter and a peripheral vein were obtained for laboratory confirmation of CRBSI, and catheter tips were also sent for culture. The number of negative blood cultures (including patients with no laboratory CRBSI confirmation) was not reported. Negative catheter tip cultures were considered an exclusion criterion, and the number of these excluded patients was also not reported. The primary outcome was catheter infection‐free survival, without a clear definition. The study’s authors presented definitions for cure, treatment failure, and indeterminate results, but these outcomes were not reported. Although it was not a planned outcome, the researchers also described infection recurrence rates, with no reported differences between the interventions. Overall, Saleh 2017 did not report a difference in catheter infection‐free survival between catheter guidewire exchange and catheter removal with late replacement (Analysis 3.2 (678 participants): Hazard Ratio 0.88, 95% CI 0.43 to 1.79).
Primary outcomes
Cure (clinical resolution associated with negative blood cultures at least 72 hours after completion of systemic antibiotic treatment) was not correctly defined and evaluated by any of the included studies. Onder 2008 performed blood cultures, but these tests were performed during the use of antibiotics. Khosroshahi 2015 considered cure only as clinical improvement, without laboratory confirmation. The definition of cure considered by Saleh 2017 was in accordance with our review; however, the study did not report results for this outcome.
Stenosis and thrombosis of vascular access after catheter removal were not reported by any of the included studies.
Secondary outcomes
Death, catheter malfunction, duration of hospitalisation, development of antibiotic resistance, fatigue, cardiovascular disease, and adverse outcomes (e.g., bleeding or anaphylaxis) were not reported by any of the included studies.
Subgroup and sensitivity analysis
Subgroup and sensitivity analyses were not performed due to the limited available data.
Discussion
Summary of main results
The findings of this review suggest that it remains uncertain if systemic antibiotic in association with an ethanol lock solution is superior to systemic antibiotics alone for the treatment of CRBSIs in patients with kidney failure on maintenance HD. This uncertainty is mainly because of imprecision, directness, and a high risk of bias for many domains. This evidence was restricted to one single‐centre study (64 participants), with 45% of the considered CRBSIs presenting negative blood cultures and with cure defined only by a clinical improvement in signs and symptoms, without laboratory confirmation. Additionally, the safety of this treatment is also uncertain, as the study did not report other outcomes, such as adverse events, death, antibiotic resistance, or venous stenosis or thrombosis after catheter removal.
No differences were reported for catheter infection‐free survival when comparing catheter guidewire exchange to catheter removal and late replacement or when comparing two different locking treatments (which differed from anticoagulant or thrombolytic agent use only).
Overall completeness and applicability of evidence
This review aimed to assess the different treatments for bloodstream infections in patients with permanent CVCs receiving maintenance HD. Only three studies met our inclusion criteria, and each study compared different interventions: Onder 2008 compared the use of anticoagulants and thrombolytic for antibiotic lock solutions, Khosroshahi 2015 compared the inclusion of an ethanol lock solution with the use of systemic antibiotics alone, and Saleh 2017 compared different catheter removal manoeuvres. We found no other studies evaluating other comparisons. The secondary outcomes of this review, as well as the presence of venous stenosis or thrombosis, were not assessed by the included studies. Therefore, the data on the treatment of CRBSI remains incomplete, and other relevant questions regarding antibiotic resistance after antibiotic locking and the development of venous stenosis and thrombosis after catheter removal remain unanswered.
A critical feature observed in this review was the different definitions of cure used by each study. Lok 2019 highlighted the lack of consensus among associations for CRBSI definition impairs accurate reports and reliable comparisons between studies. Accordingly, none of the included studies’ definitions of cure fit our primary outcome. Therefore, we were unable to perform a combined analysis of this crucial outcome.
It is also important to note the significant clinical heterogeneity between the participants: one study only included children with kidney failure (Onder 2008), and two studies only included adults with kidney failure (Khosroshahi 2015; Saleh 2017). The mean age of the included participants was 15.1 ± 6.8 years for children and 57.5 ± 15.6 years for adults, and the primary aetiology of kidney failure differed significantly between these groups. In addition, although Saleh 2017 was a multicentre study, it included only Middle Eastern institutions, and the authors reported a substantially higher rate of CVC use for patients on prevalent HD in their centres than the corresponding rate observed in America and Europe.
The methodological limitations, small samples, and mostly single‐centre settings of the included studies also lead to concerns regarding external validity.
Quality of the evidence
We classified the quality of the evidence as very low. We downgraded the quality of evidence due to serious limitations in the design (high risk of bias in critical domains such as randomisation, allocation, and other sources of bias), indirectness (lack of clear outcome definitions and use of catheter infection‐free survival as an indication of cure) and because of imprecision (small number of participants, only 760 in total, and just three included studies).
Potential biases in the review process
One study is awaiting classification (Khosroshahi 2006b). It was published as an abstract, the full report and results were not available, and doubts regarding participants' use of short‐ or long‐term catheters remained even after correspondence. We attempted to communicate with the authors responsible for the included studies and obtained two replies. The author of Khosroshahi 2006b and Khosroshahi 2015 clarified the allocation process of his last study but unfortunately did not eliminate many of our doubts once the raw data from both studies were unavailable. Onder 2008’s response presented conflicting information when compared to the article’s description, and further questions were not clarified. No response was obtained from Saleh 2017.
Agreements and disagreements with other studies or reviews
We identified other previous reviews that examined the treatments of CRBSIs (Allon 2004; Aslam 2014; Lok 2011).
Allon 2004 focused on observational, non‐randomised studies. The researchers reported clinical cure in only 22% to 37% of CRBSIs treated with systemic antibiotics alone and considered this method a suboptimal approach. The authors compared this finding with that derived from three other catheter strategies (including removal and delayed replacement, guidewire exchange, and maintenance with lock treatment) and reported better clinical cure with similar infection‐free survival rates between them.
Lok 2011 was a descriptive review of the literature. It addressed both CRBSI prevention and treatment and compared the use of systemic antibiotics alone to its association with different catheter approaches (lock treatment, catheter exchange over guidewire, and catheter removal with late replacement). Of the 12 included studies, only one was a quasi‐randomised study (Onder 2008), and the rest were either prospective, uncontrolled or observational studies. Overall, it concluded that catheter salvage by any means is associated more with failure rates (above 65%), and it should thus be avoided.
In the more recent of these reviews, Aslam 2014 conducted a systematic review and meta‐analysis that included 28 publications ranging from observational cohorts to chart reviews. Although the study described a thorough protocol, its findings were jeopardized by conducting a meta‐analysis with different study designs and different outcome definitions. The study focused on the role of antibiotic lock solutions and catheter guidewire exchange compared to systemic antibiotics alone. Decisions regarding whether the patient would undergo a catheter salvage attempt were made at the physician's discretion in many of the included studies, leading to an important selection bias. As in our review, the researchers reported poor cure proportions with the use of systemic antibiotics alone compared to antibiotic lock solution. They also reported a significant improvement in cure proportions when comparing systemic antibiotics alone with catheter guidewire exchange – which could not be evaluated in our review due to the lack of RCTs. Aslam 2014 found no significant differences in cure outcomes between antibiotic lock solution and catheter guidewire exchange. The authors also pooled the microbiological data and reported that infections by coagulase‐negative staphylococci exhibited higher cure proportions, followed by gram‐negative and S. aureus infections. In addition, they found that S. aureus CRBSI showed better resolutions when treated by catheter guidewire exchange than by antibiotic lock solution. Our review planned to analyse the outcomes by pathogen subgroup; however, we did not have enough data.
Study flow diagram.
Risk of bias graph: review authors' judgements about each risk of bias item presented as percentages across all included studies.
Risk of bias summary: review authors' judgements about each risk of bias item for each included study.
Comparison 1: Tissue plasminogen activator‐antibiotic lock treatment versus heparin‐antibiotic lock treatment, Outcome 1: Short‐term success
Comparison 1: Tissue plasminogen activator‐antibiotic lock treatment versus heparin‐antibiotic lock treatment, Outcome 2: Long‐term success
Comparison 1: Tissue plasminogen activator‐antibiotic lock treatment versus heparin‐antibiotic lock treatment, Outcome 3: Infection‐free survival
Comparison 2: Systemic antibiotic alone versus systemic antibiotic plus antibiotic lock solution, Outcome 1: Successful eradication of infection
Comparison 3: Catheter guidewire exchange versus catheter removal and replacement, Outcome 1: Infection recurrence
Comparison 3: Catheter guidewire exchange versus catheter removal and replacement, Outcome 2: Infection‐free survival
| Systemic antibiotics plus antibiotic lock solutions plus TPA versus systemic antibiotics plus antibiotic lock solutions plus heparin | |||||
| Patient or population: patients on maintenance HD with CRBSI Settings: inpatient Intervention: systemic antibiotics plus antibiotic lock solution plus TPA Comparison: systemic antibiotics plus antibiotic lock solution plus heparin | |||||
| Outcomes | Anticipated absolute effects* (95% CI) | Relative effect | No. of Participants | Quality of the evidence | |
|---|---|---|---|---|---|
| Risk with heparin | Risk with TPA | ||||
| Cure: short‐term success Follow‐up: 2 weeks | 1000 per 1000 | 920 per 1000 (740 to 1150) | RR 0.92 (0.74 to 1.15) | 18; 24 catheters | ⊕⊝⊝⊝ |
| Cure: long‐term success Follow‐up: 6 weeks | 917 per 1000 | 587 per 1000 (348 to 972) | RR 064 (0.38 to 1.06) | 18; 24 catheters | ⊕⊝⊝⊝ very low 1,2 |
| Cure: infection‐free survival Follow‐up: between the final dose of antibiotics and the first positive blood culture obtained from the catheter | Infection‐free survival was 27.7 days less with TPA (88.68 days less to 33.28 days more) than with heparin | ‐ | 18; 24 catheters | ⊕⊝⊝⊝ very low 1,2 | |
| Stenosis or thrombosis of vascular access site after catheter removal | Not reported | Not reported | ‐ | ‐ | ‐ |
| Death | Not reported | Not reported | ‐ | ‐ | ‐ |
| Development of antibiotic resistance | Not reported | Not reported | ‐ | ‐ | ‐ |
| Adverse effects | Not reported | Not reported | ‐ | ‐ | ‐ |
| *The risk in the intervention group (and its 95% CI) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI). | |||||
| GRADE Working Group grades of evidence | |||||
| 1High risk of bias in random sequence generation and allocation concealment (selection bias), blinding of participants and personnel (performance bias), inconsistencies regarding the description of the data analysis and the tables, protocol break and also conflicting information between the author's reply and the article (other bias); downgraded by 2 levels (methodological limitation) 2Small number of participants and studies (doubt about the reproducibility of the data); downgraded by 1 level (imprecision) | |||||
| Systemic antibiotics alone versus systematic antibiotics plus antibiotic lock solution | |||||
| Patient or population: patients on maintenance HD with CRBSI Settings: inpatient Intervention: systemic antibiotics alone Comparison: systemic antibiotics and antibiotic lock solution (ethanol) | |||||
| Outcomes | Anticipated absolute effects* (95% CI) | Relative effect | No. of participants | Quality of the evidence | |
|---|---|---|---|---|---|
| Risk with systemic antibiotics alone | Risk with systemic antibiotic plus | ||||
| Cure: successful eradication of the infection Follow‐up: 2 days | 563 per 1000
| 906 per 1000 (653 to 1000) | RR 1.61 (1.16 to 2.23) | 64 (1) | ⊕⊝⊝⊝ 1,2,3 |
| Stenosis or thrombosis of vascular access site after catheter removal | Not reported | Not reported | ‐ | ‐ | ‐ |
| Death | Not reported | Not reported | ‐ | ‐ | ‐ |
| Development of antibiotic resistance | Not reported | Not reported | ‐ | ‐ | ‐ |
| Adverse effects | Not reported | Not reported | ‐ | ‐ | ‐ |
| *The risk in the intervention group (and its 95% CI) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI). | |||||
| GRADE Working Group grades of evidence | |||||
| 1High risk of bias in selective reporting because of retrospective trial registration (reporting bias) and other bias once 45% of the included population have negative blood cultures; downgraded by 1 level (methodological limitation). 2Small number of participants and studies (doubt about the reproducibility of the data); downgraded by 1 level (imprecision). 3Outcome cure defined only by clinical improvement in signs and symptoms, without laboratory confirmation; downgraded by 1 level (indirectness). | |||||
| Catheter guidewire exchange versus catheter removal and replacement | |||||
| Patient or population: patients on maintenance HD with CRBSI Settings: inpatient Intervention: systemic antibiotics and catheter guidewire exchange Comparison: systemic antibiotics and catheter removal and replacement | |||||
| Outcomes | Anticipated absolute effects* (95% CI) | Relative effect | No. of participants | Quality of the evidence | |
|---|---|---|---|---|---|
| Risk with systemic catheter removal | Risk with guidewire exchange | ||||
| Cure: infection‐free survival Follow‐up: 45 days | 720 per 1000 | 749 per 1000 (555 to 868) | HR 0.881 (0.43 to 1.79) | 678 (1) | ⊕⊝⊝⊝ 2,3,4 |
| Stenosis or thrombosis of vascular access site after catheter removal | Not reported | Not reported | ‐ | ‐ | ‐ |
| Death | Not reported
| Not reported | ‐ | ‐ | ‐ |
| Development of antibiotic resistance | Not reported | Not reported | ‐ | ‐ | ‐ |
| Adverse effects | Not reported | Not reported | ‐ | ‐ | ‐ |
| *The risk in the intervention group (and its 95% CI) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI). | |||||
| GRADE Working Group grades of evidence | |||||
| 1Outcome cure was measured as a time‐to‐event outcome. Comparator group risk of event‐free survival at 90 days. Baseline risk line of 0.72 comes from the study's Kaplan‐Meier survival curves at 90 days time point, which represents the middle of the follow‐up 2High risk of bias in blinding of participants and personnel (performance bias), incomplete outcome data (attrition bias), selective reporting because of retrospective trial registration (reporting bias) and other bias; downgraded by 1 level (methodological limitation). 3Small number of participants and studies (doubt about the reproducibility of the data); downgraded by 1 level (imprecision) 4Use of catheter infection‐free survival as an indication of cure; downgraded by 1 level (indirectness) | |||||
| Outcome or subgroup title | No. of studies | No. of participants | Statistical method | Effect size |
| 1.1 Short‐term success Show forest plot | 1 | 24 | Risk Ratio (M‐H, Random, 95% CI) | 0.92 [0.74, 1.15] |
| 1.2 Long‐term success Show forest plot | 1 | 24 | Risk Ratio (M‐H, Random, 95% CI) | 0.64 [0.38, 1.06] |
| 1.3 Infection‐free survival Show forest plot | 1 | 24 | Mean Difference (IV, Random, 95% CI) | ‐27.70 [‐88.68, 33.28] |
| Outcome or subgroup title | No. of studies | No. of participants | Statistical method | Effect size |
| 2.1 Successful eradication of infection Show forest plot | 1 | 64 | Risk Ratio (M‐H, Random, 95% CI) | 1.61 [1.16, 2.23] |
| Outcome or subgroup title | No. of studies | No. of participants | Statistical method | Effect size |
| 3.1 Infection recurrence Show forest plot | 1 | 678 | Risk Ratio (M‐H, Random, 95% CI) | 1.50 [0.43, 5.27] |
| 3.2 Infection‐free survival Show forest plot | 1 | Hazard Ratio (IV, Random, 95% CI) | 0.88 [0.43, 1.79] | |
