4 Committee discussion

Clinical-effectiveness overview

DyeVert Systems are effective in reducing contrast media volume but more evidence is needed on acute kidney injury incidence

4.1 There is consistent evidence that DyeVert Systems can reduce contrast volume received by the patient by up to 40%. Clinical experts said that using DyeVert Systems can maintain the pressure of the injection needed to preserve image quality, while reducing the overall amount of contrast given. The committee heard that the cause of acute kidney injury (AKI) is multifactorial and complex. It understood that contrast media is 1 risk factor for AKI, but it can be difficult to identify its direct cause (or causes), given other confounding factors such as comorbidities (which are often significant in people having angiography) and procedural complexities. The committee also heard conflicting views on the causal relationship between contrast agent and AKI. Two experts pointed out that clinical guidelines say that contrast volume is a risk factor for AKI, and that contrast volume is a modifiable parameter that could help to reduce AKI risk. However, 1 expert maintained that contrast-induced AKI has not been proven clinically, and that the link between contrast agent and AKI may only be an association. This was a key uncertainty, and as a result the committee was not confident that a reduction in contrast dye received would lead to reduced incidence of AKI. It felt that the strength of evidence on AKI incidence reduction using DyeVert Systems was not robust enough to address these concerns, and it could not be confident in the results of the meta-analyses (see section 4.4).

The evidence for coronary angiography is not generalisable to peripheral angiography

4.2 The evidence presented on peripheral angiography included 9 people in 2 studies. Clinical experts explained that it was unlikely that the evidence for coronary angiography was transferable to peripheral angiography procedures. Types of peripheral angiography procedure vary. Different volumes of contrast dye and different injection pressures are needed, making comparisons with coronary angiography difficult. The committee also heard that peripheral angiography can be done without using contrast media, for example when used for diagnosis.

Generalisability of the evidence for the PLUS EZ device to the Power XT device is uncertain

4.3 The evidence presented on the Power XT version of the DyeVert System included 2 studies on 35 people and was based on an older model of the system. The company said that the PLUS EZ and Power XT versions work in a similar way, with both devices responding to pressure going through the valve. Clinical experts thought that the power injector was more likely to be used in peripheral angiographies, especially when large volumes of contrast dye are needed to be given quickly. The committee was uncertain if the evidence on the PLUS EZ device was generalisable to the Power XT device. It felt that more evidence was needed before the devices could be considered comparable.

The meta-analyses are statistically robust, but the risk of bias is uncertain

4.4 The company submitted meta-analyses for several parameters, which the EAC said were statistically valid. However, some of the analyses, such as a reduced risk of AKI because of DyeVert Systems use, predominantly used data from abstracts and posters. The committee was concerned that it was difficult to assess the methodological quality of some studies, making it difficult to quantify the size and direction of bias in the meta-analysis. This made the reliability of the outcome measures uncertain.

Outcome measures

Better quality long-term follow up is needed to identify most cases of AKI

4.5 Most of the clinical evidence was limited by only having data from during the procedure. If AKI incidence was reported not all may have been identified because the studies were retrospective. Clinical experts said that most AKI events happen 4 days to 5 days after the contrast exposure, and sometimes as late as 7 days to 10 days afterwards. Incidence of AKI may be difficult to track because serum creatinine measurements may not be done routinely and there may not be consistent post-procedure renal monitoring. Overall, the committee felt that more long-term evidence on AKI incidence was needed.

Longer-term follow up is needed to collect data on secondary end points

4.6 Because the clinical evidence was limited by the follow-up evidence, the committee thought that further data collection was needed to capture secondary end points. This includes the need for temporary or end-stage dialysis and hospital stay.

NHS considerations overview

DyeVert Systems may be of most benefit to people with chronic kidney disease stage 4 and over

4.7 The clinical experts said that people with chronic kidney disease (CKD) stage 4 and over (estimated glomerular filtration rate [eGFR] less than 30 ml/min/1.73 m²) would most benefit from DyeVert Systems and are likely to be the target population in clinical practice. Evidence presented on DyeVert Systems was in people with CKD stages 2 and 3 (eGFR 30 ml/min/1.73 m² to 89 ml/min/1.73 m²), so the evidence on people with CKD stage 4 and over was limited. The company said that the population included in the studies had additional comorbidities, making them at greater risk of developing AKI after a contrast procedure.

Procedure type and anticipated contrast volume use should be considered when deciding whether to use DyeVert Systems

4.8 Clinical experts said that some procedures, such as diagnostic angiographies, are lower risk for AKI than more complex procedures, such as percutaneous coronary intervention. This is because the more complex procedures need larger volumes of contrast dye. The clinical experts felt that procedure type and anticipated volume of contrast dye needed should be taken into account when deciding whether to use DyeVert Systems.

Cost modelling overview

The cost model for DyeVert Systems is well constructed

4.9 The cost model for DyeVert Systems used a well-constructed decision tree and Markov model, with appropriate outcomes based on NICE clinical guidelines, and an appropriate time horizon. The clinical experts thought that the EAC's revised assumption of a relative risk of AKI of 8.74% was reasonable.

Because the link between reduced risk of contrast-induced AKI and using DyeVert Systems is not certain, the cost savings are not certain

4.10 The committee considered that the most robust evidence for DyeVert Systems was on surrogate markers, such as contrast volume reduction, rather than AKI incidence. According to clinical experts the causal relationship between contrast agent and AKI is not certain (see section 4.1). The results of the cost analysis were sensitive to the reduction in relative risk of AKI from using DyeVert Systems. Three of the 4 studies that reported that DyeVert Systems reduced the relative risk of AKI were abstracts or posters. These were included in the meta-analysis that provided the values used in the economic model. Including these studies meant the robustness of and risk of bias in the meta-analysis were uncertain. The reduction in relative risk of AKI for DyeVert was reported as 42% in the 1 full text published study and 41% in the meta-analysis. The EAC's sensitivity analysis found the break-even reduction in relative risk to be 38.5%. Because the economic model was very sensitive to this parameter, there was uncertainty around whether using DyeVert Systems would lead to a cost saving.

Further research

Further good-quality research is needed to address uncertainties about the clinical efficacy of DyeVert Systems

4.11 The committee concluded that further research is needed to address uncertainties in the clinical effectiveness of DyeVert Systems compared with standard care. It concluded that, although there is clear evidence that DyeVert Systems can reduce contrast volume received while maintaining image quality, the evidence around AKI incidence was weaker. The committee concluded that a randomised controlled trial was needed. It should collect data on AKI incidence, renal replacement therapy, hospital stay and biochemical markers of kidney injury, and should follow up for enough time to capture these outcomes. The population should include people having an elective procedure who have an eGFR less than 30 ml/min/1.73 m². Further evidence should be collected to resolve the uncertainties around the generalisability between the 2 DyeVert devices. The committee noted that collecting evidence for DyeVert in peripheral angiography will be more difficult because of the different ways the procedure is done. It concluded that for peripheral angiography it may be appropriate to trial the system in procedures that use a consistent volume of contrast, such as endovascular aneurysm repair.

Real-world evidence may be difficult to interpret because of confounding factors

4.12 The committee thought that confounding factors make interpreting real-world evidence challenging. Comorbidities and procedural factors affect the risk of developing AKI after a contrast procedure, and these may not be clearly reported. There may also be inconsistencies in routine post-procedure renal monitoring, including timing and frequency of serum creatinine measurements.

How are you taking part in this consultation?

Question on Document

Question on Document

Question on Document

Question on Document

DyeVert Systems for reducing the risk of acute kidney injury in coronary and peripheral angiography