Table 5.

Summary of findings table

Outcomes	Impact	№ of participants (studies)	Quality of the evidence (GRADE)
Vascular Access interventions to prolong filter life in CRRT
Temporary Vascular Access Site	Optimal vascular access site ranked by association with longer filter life is: tunneled semi-permanent, femoral or internal jugular, subclavian site.a	(9 observational studies)	VERY LOW a,b,c
Tunneled Semi-permanent Catheters vs Temporary Catheters	Tunneled semi-permanent vascular access devices were consistently associated with longer filter life. A significant confounder is that these devices were often larger internal diameter than temporary devices however on the basis of current literature they should be considered in any cases expected to have prolonged CRRT requirement	(4 observational studies)	LOW b,c
Side of Vascular Access Catheter	Overall there is insufficient data and possibility of significant confounding by order of catheter choice such that optimal side of vascular access cannot be determined	(3 observational studies)	LOW a
Catheter length at thoracic vein sites	Favours longer catheter length with atrial placement when thoracic veins utilised. Single study only however unlike most filter life studies this was randomized. No increased in arrhythmias with longer catheter length however underpowered to detect complications.	(1 RCT)	MODERATE d
Catheter lumen size	Only one small study directly measured filter life with catheter size however indirect measures (increased renal dose) in RENAL study supports catheter size as important. Possible benefit from tunneled access may be due to catheter size	(3 observational studies)	VERY LOW a,c,e
Number of vascular access related alarms	Number of vascular access alarms is likely a significant contributor to poor filter life however data is limited	(1 observational study)	VERY LOW g
Access Catheter Type	No significant difference between brands of catheters though trend existed. Tunneled catheters were superior to temporary catheters	(4 observational studies)	LOW b,e,g
Circuit Management Interventions to Prolong Filter Life in CRRT
Haemofilter Membrane Characteristics	Hollow fibre membranes appear superior to flat plate membranes. It is unclear if an advantage exists for polyacrylonitrile membranes compared to polysulfone or cellulose membranes in regards to filter life. Membrane area was not associated with increased filter life in a single study.	(8 observational studies)	VERY LOW h
CRRT Modality	CVVH is associated with worse filter life in published studies	(7 observational studies)	LOW i
Pre vs Post Dilution in CVVH	One small RCT favoured pre-dilution. Overall affect from all studies trended toward pre-dilution but did not reach significance.	(4 observational studies)	VERY LOW i
Blood Flow Rate	Majority of trials suggest a higher blood flow rate increases filter life however it is unclear over what range this applies. Studies directly comparing low and high blood flow are required.	(9 observational studies)	VERY LOW i
Saline Flushes to Blood Path	There is no evidence to support intermittent saline flushing of the circuit to prolong filter life	(4 observational studies)	VERY LOW i
Education and Alarm Management	Limited evidence suggests focused training to recognise and respond to filter warnings prolongs filter life	(4 observational studies)	VERY LOW i
Patient Factors associated with prolong filter life in CRRT
Factors with a positive association with filter life	Increasing age (NS), Presence of vaso-active drugs (NS) f, lower pH (sig), Higher APTT (sig), Higher ATIII level (NS), Correction of ATIII deficiency (sig), Mobilization (sig) f, Number of position changes (sig) f		VERY LOW b,j,c
Factors with a negative association with filter life	Being male (NS), Mechanical Ventilation (sig) f, Increased temperature (NS) f, Liver failure with bilirubin > 3 mg/dL (NS) f, Presence of Sepsis (NS) f, Higher SOFA score (sig) f, Higher LOD score (sig) f, Unit increase in ionized calcium (sig), Higher platelet count (sig), RBC transfusion (sig) f, Platelet transfusion (NS) f, FFP transfusion (NS) f, PF4 antibodies f, Elevated fibrinogen (sig) f, Prothrombin fragment (F1 + 2) elevation f		VERY LOW b,j,c
GRADE Working Group grades of evidence: High quality: We are very confident that the true effect lies close to that of the estimate of the effect Moderate quality: We are moderately confident in the effect estimate: The true effect is likely to be close to the estimate of the effect, but there is a possibility that it is substantially different Low quality: Our confidence in the effect estimate is limited: The true effect may be substantially different from the estimate of the effect Very low quality: We have very little confidence in the effect estimate: The true effect is likely to be substantially different from the estimate of effect

^aSignificant heterogeneity exists and potential for confounders

^bTiming of catheters during admission has not been studied. Other factors such as choice of catheter length, insertion technique/operator experience at different sites and catheter size at different sites may bias results

^cHeterogeneity across small observational studies

^dUnblinded (however unavoidable) however unlikely to affect results

^eOne direct study, strong suggestion that the benefit of tunneled access could be due to catheter size, large RCT post hoc suggests larger size important

^fSingle study, low numbers

^gBefore - after studies with significant risk of other practice changes

^hMixed composite of varying quality and study designs with no direct comparison between groups

ⁱSome studies used composite interventions

^jMostly small observational studies with high risk of bias