1 Provisional recommendations

1.1 The case for adopting the VeriQ system in the NHS for assessing graft flow during coronary artery bypass graft (CABG) surgery is supported by the evidence. The evidence suggests that intra-operative transit time flow measurement is effective in detecting imperfections that may be corrected by graft revision. This may reduce the incidence of graft occlusion and may reduce perioperative morbidity and mortality.

1.2 The VeriQ system is associated with a cost saving of £115 per patient compared with clinical assessment, when it is used routinely for assessing coronary artery bypass grafts during surgery.

2 The technology

Description of the technology

2.1 The VeriQ system (MediStim ASA) uses ultrasound for the non-invasive assessment of graft blood flow during coronary artery bypass graft (CABG) surgery. It is intended for use in patients with coronary artery disease who are having CABG surgery. It measures three parameters of transit time flow measurements (mean blood flow in ml/minute, pulsatility index and diastolic filling percentage) to assess graft blood flow and check patency. 

2.2 The VeriQ system measures transit time volume flow using specially designed probes. A microcomputer with a 19‑inch touch screen mounted on a moveable trolley is used to control the probes and store their outputs.

2.3 The VeriQ system uses two types of probes (the PS and PQ probes), which differ in the number of intended reuses and method of sterilisation. They deliver a bidirectional ultrasound beam across a target vessel and the system analyses the returning signal to calculate the blood flow through the vessel. A real-time flow curve is displayed together with the mean blood flow in ml/minute, pulsatility index and diastolic filling percentage. This information can be used to determine whether flow through the graft and its anastomoses is acceptable. If not, then the graft can be explored to detect imperfections and revised as necessary to achieve optimum blood flow.

2.4 The cost of the VeriQ system stated in the sponsor’s submission includes £32,000 for the VeriQ 2011 console, and £1582 for each PS probe.

2.5 The claimed benefits of the VeriQ system in the case for adoption presented by the manufacturer are:

• the VeriQ system will improve the outcomes of revascularisation procedures by reducing the risk of early graft failure and adverse events
• reduced hospital stay for some patients by reducing the incidence of complications during and after surgery
• reduced number of repeat procedures and treatments for post-operative complications.

Current management

2.6 Coronary artery disease is a common cause of symptoms, disability and death. It is caused by atherosclerosis, which leads to stenosis or occlusion of the coronary arteries. The draft NICE clinical guideline on the management of stable angina recommends that that revascularisation of the blocked coronary arteries using CABG or percutaneous coronary interventions should be considered in people whose symptoms are not satisfactorily controlled with optimal medical treatment.

2.7 CABG aims to bypass narrowed or blocked segments of the coronary arteries using grafts – usually lengths of the patient’s own long saphenous vein from the leg or their internal thoracic artery, although other vessels are also used.

2.8 Cardiac surgeons use a variety of techniques to avoid imperfections during CABG, but assessment of graft flow is usually subjective. Techniques used vary according to the graft constructed, the surgical technique, and the surgeon’s individual preference. They include the surgeon assessing resistance to flow after completion of an anastomosis (a join) and flushing the conduit, observing run-off into epicardial vessels (blood vessels beyond the bypass) during flushing and/or after establishing blood flow down the graft, and observing pulsation in the grafts and coronary arteries. There are a number of methods for the objective assessment of the technical results and of blood flow. Transoesophageal echocardiography evaluates end-organ function by assessing regional left ventricular wall motion abnormalities, which can be compared with preoperative regional left ventricular function. Perioperative graft flow can be visualised with conventional angiography in theatres or with indocyanine green fluorescence. NICE has produced interventional procedure guidance on one alternative technique, ‘Intraoperative fluorescence angiography in coronary artery bypass grafting’ (NICE interventional procedure guidance 98). The guidance states that ‘current evidence suggests that the procedure is safe enough for routine use in the evaluation of coronary artery bypass graft patency’.

3 Clinical evidence

Summary of clinical evidence

3.1 The key clinical outcomes for the VeriQ system presented in the decision problem were:

• incidence of graft failure
• time to graft failure
• peri- and post-operative clinical events associated with graft failure (including mortality)
• frequency of the need for graft revision and changes in VeriQ measurements afterwards 
• the need for repeat coronary revascularisation procedures
• long-term morbidity and mortality.  

3.2 The evidence for the clinical effectiveness of the VeriQ system was based on two retrospective observational studies that examined surgical outcomes, and one comparative study that compared parameter values from the VeriQ system against another flowmeter. The studies were conducted in hospitals in Europe and Canada (there were none in the UK). All patients in the studies were having CABG surgery.

3.3 In a retrospective case study in Canada, Kieser et al. (2010) evaluated transit time flow measurement with the VeriQ system to detect technical errors in CABGs intra-operatively and to predict postoperative major adverse cardiac events. They assessed 1000 arterial grafts in 336 consecutive patients. Three parameters of transit time flow (pulsatility index, flow and diastolic filling percentage) were measured in 990 (99%) of the grafts. A pulsatility index value of less than 5 was chosen as the principal measure of graft adequacy. In 82% of the patients (277 of 336), 93% of grafts (916 of 990) had pulsatility index less than or equal to 5. The remaining 74 (7%) grafts (in 59 patients, 18%) had pulsatility index greater than five, but grafts were revised only when an abnormally high pulsatility index was accompanied by other indications of graft malfunction (abnormal electrocardiogram [ECG] changes, regional wall motion abnormality on transoesophageal echocardiography or haemodynamic compromise). On this basis, 20 grafts (in 14 patients, 4%) that were suspected to be problematic were revised.

For analysis of the findings, patients were divided into two groups: the 277 (82%) with at least one graft with pulsatility index less than 5, and 59 (18%) with at least one graft with pulsatility index greater than 5. Major adverse cardiac events (recurrent angina, perioperative myocardial infarction, postoperative angioplasty, re-operation and/or perioperative death) occurred significantly more often in patients with a pulsatility index greater than 5 (10 of 59, 17%) when compared with patients with a pulsatility index less than 5 (15 of 277, 5.4%, p = 0.005). Mortality following non-emergency surgery was significantly higher in the patient group with a pulsatility index greater than 5 (5 of 54, 9%) than in the group with a pulsatility index less than 5 (5 of 250, 2%, p = 0.02).

3.4 Becit et al. (2007) evaluated the effect of detecting graft dysfunction by intraoperative transit time flow measurement on the surgical results of CABG using the VeriQ system in a case–control study in Turkey. A pulsatility index of greater than 5 and diastolic filling percentage of less than 50% were used as the indicators of inadequate flow. The study compared the surgical outcomes for two matched series of consecutive patients whose operations were performed by the same surgeons. The study group (n = 100) had transit time flow measurement during surgery and the control group (n = 100) did not. Three percent (9 of 303) of grafts in nine (9%) patients in the study group were revised on the basis of abnormal transit time measurements and after revision, all flow values and flow patterns improved. No information was presented about graft revision in the control group. The incidence of intra-aortic balloon pump insertion for low cardiac output was significantly lower in the study group compared with the control group (1 of 100 versus 7 of 100, p < 0.05). Also, perioperative myocardial infarction was lower in the study group compared with the control group (0 of 100 versus 5 of 100, p < 0.05). There was no statistically significant difference between the patient groups in intraoperative re-exploration for bleeding or deep sternal infection.

3.5 Nordgaard et al. (2010) investigated the variation in pulsatility index measurement between two different flowmeters (VeriQ and Transonic) and examined whether increasing filtering of the flowmeter signals influenced flow curves and pulsatility index. The VeriQ and Transonic flowmeters have default filter settings of 10 Hz and 20 Hz respectively. Flow patterns in 19 patients recorded simultaneously by both flowmeters during CABG surgery were analysed. This showed that the VeriQ system provided systematically higher pulsatility index values than the Transonic device (mean ± standard deviation [SD]: 2.7 ± 1.2 versus 1.8 ± 0.6 respectively, p < 0.001).

3.6 Twenty-six studies on predecessor devices of the VeriQ system were also included in the sponsor’s clinical evidence submission. The External Assessment Centre analysed these studies, the results of which suggested that transit time flow measurement can predict graft dysfunction following CABG surgery and may be easier to carry out than other methods.

Committee considerations

3.7 The Committee recognised that graft dysfunction is a major determinant of perioperative morbidity and mortality after CABG. It was advised that the majority of graft failures in the perioperative period are due to technical imperfections which, if recognised, might be corrected at the time of surgery.

3.8 The Committee noted that perioperative myocardial infarction resulting from graft failure may cause serious complications such as left ventricular dysfunction, ventricular arrhythmias and haemodynamic instability, which can necessitate prolonged intensive therapy unit stay. These complications may need interventions such as intra-aortic balloon pumping, coronary angiography and early reoperative CABG surgery. They may also lead to readmission to hospital.

3.9 The Committee considered that the available evidence supported the claim that transit time flow measured by the VeriQ system can identify grafts that have reduced flow caused by technical imperfections. 

3.10 The Committee recognised limitations in the available evidence. The main studies were observational, with potential for bias. The study by Kieser et al. (2010) investigated the VeriQ system on arterial grafts only, whereas in the NHS the majority of CABGs are vein grafts. Nevertheless, it judged that there was sufficient additional evidence relating to predecessor devices and expert advice to support the expectation that routinely revising all grafts on the basis of VeriQ measurements would result in reduced perioperative graft occlusions and consequent complications.

3.11 The Committee was advised that cardiac surgeons use a variety of methods to minimise and detect technical imperfections during CABG surgery but these may have limitations. On the basis of the evidence, it judged that the routine use of VeriQ, either alone or as an adjunct to other methods of assessment, would be likely to detect technical problems in grafts that appear to be satisfactory on clinical assessment alone.

3.12 The Committee noted that recent joint guidelines on myocardial revascularisation issued by the European Society of Cardiology (ESC) and the European Association for Cardio-Thoracic Surgery (EACTS) have recommended graft evaluation before leaving the operating theatre after CABG surgery. These guidelines refer to flow less than 20 ml/minute and a pulsatility index greater than 5 as predicting technically inadequate grafts that need revision before leaving the operating theatre.

3.13 The Committee recognised that the clinical outcomes of CABG surgery have improved in the UK in the past 20 years and that complication rates are now very low. However the expert adviser stated that there is still a perioperative graft occlusion rate of 1–3%. The Committee considered that the VeriQ system has potential to reduce this graft occlusion rate and so further reduce morbidity and mortality after CABG.

4 NHS considerations

System impact

4.1 Approximately 22,500 isolated CABG operations are performed in the UK each year. In addition, a substantial proportion of patients having other cardiac surgery (for example, valve replacement surgery) have concomitant CABG. Based on this large number of patients, any reduction in graft occlusion rates by the VeriQ system during CABG surgery potentially offers significant cost savings to the NHS.

4.2 The Committee was informed that the VeriQ system is easy to use and does not significantly increase operative time.

Committee considerations

4.3 As described in section 3, the Committee judged that reduction in graft occlusion rates by VeriQ assessment and appropriate revision at the time of surgery could decrease complication rates, and so reduce the likelihood of subsequent interventions, prolonged intensive therapy unit and hospital stay, and readmission. Each of these reductions would result in significant resource savings.

5 Cost considerations

Cost evidence

5.1 The economic evidence for the VeriQ system comprised a new cost analysis to assess the cost savings to the NHS of introducing the VeriQ system for assessing graft flow during CABG surgery, compared against clinical assessment.

5.2 In the base-case analysis, the equipment cost for the VeriQ system was about £111 per procedure and the additional time for measuring flow in three grafts was 2.35 minutes. The equipment costs were based on the VeriQ 2011 console (purchase cost £32,000 with an anticipated life span of 10 years) and an average use of 1.7 probes per procedure (£1582 per PS probe, which is recommended for up to 30 uses). The costs per patient were based on the purchase cost of a VeriQ system divided by 220 days’ use per year over its life expectancy, including annual maintenance costs. Annual maintenance costs are payable from the end of year two at £1800 per year. It was assumed that the annual maintenance costs for the remaining 8 years would be averaged over the 10-year life expectancy of the equipment. All time costs in the model were based on the salaries of a CABG team comprising two cardiothoracic surgeons, one anaesthetist, one cardiac perfusionist and two cardiac nurses.

5.3 The cost model evaluated the cost savings of using the VeriQ system compared with clinical assessment alone in assessing graft flow in all patients having CABG. The outcomes considered in the model are complications associated with the CABG surgery.

5.4 The consequences of using the VeriQ system were based on results from two studies (Kieser et al. [2010] and Becit et al. [2007]). In the base-case analysis, use of the VeriQ system was associated with an increase of 6.6% in the graft revision rate (a 2.3% increase in minor revisions and a 4.3% increase in major revisions). Costs were based on the time taken by the CABG team to perform the revisions. The cost of the time taken to perform a minor revision was estimated to be £11, and for major revisions this was £180.

5.5 The outcomes considered were incidence of postoperative myocardial infection and associated rehabilitation costs, use of intra-aortic balloon pumping, incidence of intraoperative re-exploration of bleeding and deep sternal infection. The rates of these outcomes for CABG with and without the VeriQ system were based on Becit et al. (2007). The base-case analysis compared a 0% postoperative myocardial infarction rate for patients assessed clinically and with VeriQ versus a 5% rate for patients who had clinical assessment alone. The cost of postoperative myocardial infarction and associated rehabilitation costs were estimated to be £1667 per patient. The base-case analysis compared a 1% rate for intra-aortic balloon pumping for patients assessed clinically and with VeriQ versus a 7% rate for patients who had clinical assessment alone. An intra-aortic balloon pumping cost was estimated to be £2657. There was no difference in the rate of intraoperative re-exploration of bleeding and deep sternal infection rate between the arms. No adverse event costs were included in the model because no adverse events have been reported as a result of using the VeriQ system.  

5.6 The cost saving associated with the VeriQ system in the base case was £115 per patient based on purchase of a VeriQ 2011 console (£32,000), using a PS probe (£1582 for 30 uses), and annual maintenance costs (£1800) payable from year three.

5.7 The sensitivity analysis based on the parameters and ranges identified by the manufacturer showed that estimates of cost saving for the VeriQ system are robust. The key drivers of the cost saving were the reduction in the rate of postoperative myocardial infarction and the reduction in use of intra-aortic balloon pumping associated with the use of the VeriQ system. The highest cost saving obtained in the sensitivity analysis was £323 per patient and was associated with 0% use of intra-aortic balloon pumping in patients assessed with the VeriQ system compared with a usage rate of 14% in those patients assessed without the VeriQ system. The lowest cost saving, of £38 per patient, was obtained for a 2.5% postoperative myocardial infarction rate. The only case in which use of the VeriQ system was not cost saving (when the cost per patient was £45) was when there was no change in the usage rate of intra-aortic balloon pumping in either arm (3.5%). The External Assessment Centre advised that this is an unnecessarily pessimistic view and that the VeriQ system is likely to be cost saving when used appropriately.

Committee considerations

5.8 The Committee considered that the assumptions made in the cost model were realistic and that the range of savings calculated for the use of VeriQ was likely to be realised in practice.

5.9 The Committee noted that the manufacturer’s cost model did not include potential cost savings from reductions in intensive therapy unit stay and reduced readmission rates. The cost savings associated with the VeriQ system may therefore have been underestimated with regard to these parameters.

5.10 The Committee also noted that the manufacturer’s estimated usage of the VeriQ system at one patient per day for 220 days per year was likely to be conservative. The Committee was advised that on average three to four CABG operations are performed per day in a cardiac operating theatre in the UK. Increased annual use of a VeriQ system is expected to reduce the estimated equipment cost per procedure because the capital cost of the VeriQ system will be divided across an increased number of procedures

5.11 The Committee considered that the reductions in perioperative myocardial infarction rate to zero and of intra-aortic balloon pump use from 7% to 1% by using the VeriQ system compared with clinical assessment alone in the base case were likely to be overestimates. This would tend to reduce the estimated cost savings of the VeriQ system. However, the Committee noted that sensitivity analysis showed that if using the VeriQ system had no impact on the postoperative myocardial infarction rate or led to only a small change in intra-aortic balloon pumping rates (of less than 2%), the VeriQ system remained cost saving compared with clinical assessment alone, resulting in a saving to the NHS.

6 Conclusions

6.1 The Committee concluded that the available clinical and cost evidence supported the case for adopting the VeriQ system in the NHS for routine intraoperative graft flow assessment in patients having CABG surgery.

7 Implementation

7.1 NICE has developed tools to help organisations put this guidance into practice (listed below). These are available on our website (www.nice.org.uk/guidance/TAXXX). [NICE to amend list as needed at time of publication]

• Slides highlighting key messages for local discussion.
• Costing template and report to estimate the national and local savings and costs associated with implementation.
• Implementation advice on how to put the guidance into practice and national initiatives that support this locally.
• A costing statement explaining the resource impact of this guidance.
• Audit support for monitoring local practice.

Related NICE guidance

Published

• Prevention of cardiovascular disease at population level. NICE public health guidance 25 (2010). Available from www.nice.org.uk/guidance/PH25
• Endoscopic saphenous vein harvest for coronary artery bypass grafting. NICE interventional procedure guidance 343 (2010). Available from www.nice.org.uk/guidance/IPG343
• Totally endoscopic robotically assisted coronary artery bypass grafting. NICE interventional procedure guidance 128 (2005). Available from www.nice.org.uk/guidance/IPG128
• Intraoperative fluorescence angiography for the evaluation of coronary artery bypass graft patency. NICE interventional procedure guidance 98 (2004). Available from www.nice.org.uk/guidance/IPG98

Bruce Campbell
Chairman, Medical Technologies Advisory Committee
July 2011