Can BlueDop be used as an alternative for diagnosing peripheral arterial disease? A prospective comparison of traditional ABPI and TBPI with BlueDop

Abstract

Objectives

Ankle and toe brachial pressure index are recommended globally as first line non-invasive screening tests for peripheral arterial disease. However, they can be inaccurate and have limited utility in patients with incompressible arteries, lower limb wounds and/or lack of digits. This study aims to evaluate BlueDop, a novel device that purports to assess peripheral perfusion without the need for a lower limb tourniquet or patients lying flat.

Methods

Patients attending vascular and podiatry clinic at Aneurin Bevan University Health Board were recruited between May 2022 and May 2023. A BlueDop reading was recorded in addition to ankle and/or toe brachial pressure index. Satisfaction and preference of each method was recorded using an electronic proforma.

Results

176 patients were included. BlueDop ABPI was shown to be moderately correlated with cuff-ABPI (r_s = 0.42, p = .001) and weakly correlated with TBPI (r_s = 0.22, p = .04). BlueDop had moderate accuracy in diagnosing PAD (C statistic = 0.653) and was more accurate in diagnosing severe PAD (C statistic = 0.706). Diagnostic accuracy appeared to be improved when PAD was defined according to cuff-ABPI only. Patients were more satisfied with BlueDop (mean satisfaction score (MSS) = 4.33) compared to cuff-ABPI (MSS = 2.89, mean difference = 0.544, p < .001) and TBPI (MSS = 3.82, mean difference = 0.579, p < .001) while operator satisfaction was equal.

Conclusions

There is moderate correlation between BlueDop and cuff-ABPI. Patients prefer BlueDop to cuff-ABPI or TBPI. BlueDop appears to be an acceptable method of non-invasive assessment of PAD, although users must appreciate there is a discrepancy between it and cuff-ABPI/TBPI.

Keywords

Peripheral arterial disease chronic limb threatening ischaemia ankle/toe brachial pressure index BlueDop peripheral perfusion assessment

Introduction

The incidence of peripheral arterial disease (PAD) is increasing worldwide with an increasing physical and psychosocial burden on patients, and a financial burden on health services.¹ Patients with PAD can be asymptomatic; however, they can progress – often in a non-linear fashion² – to chronic limb threatening ischaemia (CLTI) with features of ischaemic rest pain and/or tissue loss.³ Early detection and management of CLTI is imperative as there is an estimated risk of limb loss of 20% at 1 year if left untreated.³

Ankle-brachial pressure index (ABPI) is currently recommended as a first line non-invasive method of assessment to aid diagnosis of PAD as per European (European Society Vascular Surgery),⁴ International (International Working Group on the Diabetic Foot)⁵ and Global guidelines.² Although ABPI is considered the gold standard in non-invasive screening for PAD, limitations of its accuracy and reproducibility have been documented.⁶ Best practice suggests that there should be a pre-ABPI rest time of up to 10-min and readings should be taken with the limbs in a horizontal plane.⁷ Other limitations include artificially elevated readings due to medial artery calcification (seen frequently in patients with diabetes mellitus and renal disease²) and inability to perform pressure measurements due to lower limb wounds or patient pain.⁸ An alternative pressure reading is the Toe Brachial Pressure Index (TBPI). TBPI utilises digital arteries that are less likely to develop medial sclerosis.⁹ It is recommended by several guidelines to assess the presence and severity of PAD.^2,10 While TBPI is useful, it requires training to use, and it cannot be undertaken in patients with toe wounds or digital amputations.

The BlueDop is a novel hand-held device that purports to evaluate limb perfusion without the need for a lower limb tourniquet or patients lying flat.¹¹

There are limited data on BlueDop value as a screening tool for PAD. Korszadeh et al studied 276 limbs in 166 patients and concluded that BlueDop is accurate and sensitive at detecting PAD when compared with Duplex sonography.¹² It is not, however, clearly determined how well BlueDop correlates with cuff-ABPI or TBPI undertaken in an unselected cohort of patients referred to the Vascular/Podiatry clinic, including those with diabetes.

This study aims to evaluate the utility, acceptability and patient/operators experience of using BlueDop as a novel device for assessing PAD by comparing BlueDop to the gold standard of non-invasive screening methods, cuff-ABPI and TBPI.

Materials and methods

Study summary

This study is reported according to the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) guidelines.¹³ A cross-sectional prospective study was conducted in the Vascular and Podiatry outpatient department at Aneurin Bevan University Health Board, UK, between May 2022 and May 2023. Research and Development approval was sought prior to data collection (SE/1343/21). The approval included designation of the study as a ‘service evaluation’ not requiring formal ethical approval.

The BlueDop device

BlueDop is a Conformité Européene (CE) approved hand-held device that calculates ABPI, perfusion pressure and ‘vascular reserve’ (a percentage measure of arterial ‘redundancy’ within the limb) without the need for a tourniquet or patients lying flat.

It comprises a probe, used to record arterial waveforms of the arteries at the ankle, and a linked tablet which processes the input and provides results (Figure 1). The user inputs the patient’s systolic brachial blood pressure into the tablet, then uses the BlueDop probe, held still over the dorsalis pedis and posterior tibialis arteries, to analyse the doppler waveform. BlueDop uses ‘pressure from flow’ technology, that is, it evaluates the doppler waveform of the distal tibial arteries, and, after inputting the systolic blood pressure, gives an estimate of ABPI and other variables.¹¹ The device uses these data to calculate an ABPI and other outputs (see supplemental material 1 for details of these calculations). The precise methods used to calculate these results are not available in the public domain to the best of the author’s knowledge. The resultant ‘BlueDop-ABPI’ reading is from 0 to 1 with >0.8 implying no significant PAD, <0.8 implying PAD and <0.5 implying critical limb ischaemia. As the pressure in the ankle arteries is assumed to not be greater than the brachial systolic blood pressure, BlueDop does not produce ABPI values >1.

Figure 1.

BlueDop in use with linked tablet.

BlueDop requires simple training to use, and costs approximately £5000 (equivalent to €5,700 or $6300). Prior to commencing this study operators received device training from BlueDop Medical and those not in attendance received cascade training. Those undertaking ABPI/TBPIs were experts and were well versed in these skills as part of their clinical role. They were the same operators as those undertaking the BlueDop.

Participant identification

Participants suitable for inclusion were patients referred to the vascular and podiatry clinic with possible or suspected PAD undergoing investigation with traditional cuff-ABPI and/or TBPI. Recruitment was by convenience sampling on attendance to clinic. Patients without an indication for ABPI or TBPI were excluded.

Verbal consent was obtained prior to any recording of lower limb arterial supply (BlueDop, cuff-ABPI, TBPI), as per General Medical Council guidelines.¹⁴ Data were collected at a single time point. A BlueDop reading was taken in addition to cuff-ABPI and/or TBPI. In cases where neither a cuff-ABPI nor TBPI could be recorded, and a BlueDop reading was taken, these data were not used for comparative analysis. Patient and operator experience of BlueDop, cuff-ABPI and TBPI were evaluated using a simple satisfaction questionnaire.

Data capture

Data were collected using an electronic proforma, which was completed at the time of data capture by the operator. This comprised indication for investigation(s)/reason for referral, cuff-ABPI and/or TBPI, BlueDop ABPI, time taken for each investigation, patient/operator satisfaction and patient/operator preference for each investigation. Reason for inability to perform an investigation was documented (e.g. wounds or incompressible calf vessels). Patient presenting symptoms (if any) were recorded, that is, claudication, tissue loss and rest pain. Patient and user satisfaction was recorded on a 5-point Likert scale ranging from one (very unsatisfied) to five (very satisfied). Primary outcomes were cuff ABPI and/or TBPI and BlueDop ABPI value. Secondary outcomes were patient and operator experience and preference.

Statistical methods

The Spearman Rank test was used to evaluate correlation, with ‘r_s’ denoting the correlation coefficient (ranging −1 to +1). For this analysis, cuff-ABPI and TBPI values were truncated to have a maximum value of 1 since BlueDop gives a maximum value of 1 for ABPI. Bland-Altman analyses were used to evaluate agreement between different methods of measurement. Bland-Altman plots were produced to visually represent the data, and a regression line fitted to demonstrate any potential proportional bias (where the bias varies depending on the result (ABPI value)). The bias (the mean difference between the two methods values) was calculated along with 95% confidence intervals (CIs). The limits of agreement were defined as the bias +/−1.96 standard deviation values.

C-statistics were calculated to assess BlueDop ABPI’s accuracy in identifying patients with PAD (defined as cuff ABPI<0.8 or TBPI<0.7) and severe PAD (defined as cuff ABPI<0.5 or TBPI<0.3). A C-statistic of >0.9 was considered outstanding, 0.7 to 0.9 considered acceptable, 0.6 to 0.7 considered poor and 0.5 considered no better than chance. Corresponding sensitivity, specificity, positive and negative predictive values (PPV and NPV) were also calculated. These are represented at the ‘optimal threshold’ which was determined using the ‘Youden method’ which aims to identify the threshold which maximises both sensitivity and specificity and is commonly used to report the performance of a prediction model.

Subgroup analyses included evaluating BlueDop’s accuracy to predict PAD and severe PAD when these were defined according to cuff ABPI alone and cuff TBPI alone.

A mean satisfaction score (MSS) was calculated for each method using the value of satisfaction according to the Likert scale from 1 (very unsatisfied) to 5 (very satisfied) (see supplemental material 2). The mean difference was compared using the paired t test.

Results

176 limbs in 176 patients were included in analysis of which 122 (69.3%) were diabetic. The mean age was 68.6 years (standard deviation = 13.3), 117/176 (66.5%) of patients were male. 58 (32.6%) were asymptomatic, 22 (12.5%) had symptoms consistent with intermittent claudication and 95 (54.0%) had suspected CLTI with either tissue loss or symptoms consistent with ischaemic rest pain. Asymptomatic patients were referred internally from the podiatry department to the high-risk podiatry team. Indications for referral included a lack of distal pulses on examination requiring further assessment, as well as lower limb pain initially thought to be vascular in origin, but on further assessment identified to be secondary to other causes, such as neuropathy. 67 (38.1%), 95 (54.0%) and 159 (90.3%) had cuff-ABPI, TBPI and BlueDop ABPI documented readings, respectively. Operators attempted, but were unable, to record a cuff-ABPI in 92 (52%) patients, TBPI in 70 (40%) patients and a Bluedop reading in 6 (3%) patients. Reasons recorded for inability to record readings are documented in Table 1. Patients that did not have both a cuff-ABPI and/or TBPI, and a BlueDop ABPI, were excluded from comparative analysis. In total 58 records were used to compare cuff-ABPI with BlueDop ABPI, and 85 records were used to compare TBPI with BlueDop ABPI. 123 records were used to compare BlueDop ABPI to either cuff-ABPI or TBPI. Satisfaction data were obtained from patients and operators in 170 of the cases.

Table 1.

Documented reason for inability to record result.

	Patient discomfort and/or wounds	Lack of equipment	Equipment failure	Inability to obtain result	Calcified vessels	Other
ABPI	1	3	0	0	8	28
TBPI	4	0	0	5	0	19
BlueDop	0	0	1	0	0	1

Correlation of BlueDop with cuff-ABPI or TBPI

BlueDop ABPI was moderately correlated with cuff-ABPI (r_s = 0.42, p = .001) and weakly correlated with TBPI (r_s = 0.22, p = .04).

Figure 2 demonstrates a Bland-Altman plot comparing BlueDop and cuff-ABPI values. The plot’s regression line demonstrates potential proportional bias (the difference between the two methods increases as the mean of their values increase). The bias was 0.062 (95%CI = 0.006 to 0.119). The limits of agreement were −0.358 to 0.482. This estimates that in 95% of cases, the difference between BlueDop and cuff-ABPI ranges from −0.358 and 0.428.

Figure 2.

Bland-Altman plot comparing BlueDop and cuff-ABPI values. Downward sloping values in the top right of figure two represent cases where cuff-ABPI values were truncated to = 1.

Figure 3 demonstrates a Bland-Altman plot comparing BlueDop and TBPI values. The regression line demonstrates potential proportional bias (increasing as the mean values increase). The bias was −0.145 (95%CI = −0.206 to −0.084). The limits of agreement were −0.697 to 0.303. This estimates that in 95% of cases, the difference between BlueDop and TBPI ranges from −0.697 and 0.303.

Figure 3.

Bland-Altman plot comparing BlueDop and TBPI values. Downward sloping values in the top right of figure two represent cases where cuff-TBPI values were truncated to = 1.

Ability of BlueDop to predict PAD and severe PAD

BlueDop had C-statistics of 0.600 and 0.748 in predicting PAD and severe PAD, respectively (Figure 4). Table 2 demonstrates the sensitivity, specificity, positive predictive value (PPV) and negative predictive value (NPV) of BlueDop in predicting PAD and severe PAD and when defined by cuff-ABPI and TBPI alone. C-statistics were better (0.713 and 0.942, respectively) when PAD/severe PAD was defined according to cuff-ABPI only and were less good (0.421 and 0.659, respectively) when PAD/severe PAD was defined by TBPI only. Corresponding predictive values were also worse.

Figure 4.

Receiver operating characteristic (ROC) curve demonstrating C statistics of BlueDop in predicting PAD and severe PAD.

Table 2.

Sensitivity, specificity, positive predictive value and negative predictive value of BlueDop in predicting PAD and severe PAD.

	Sensitivity	Specificity	PPV	NPV
PAD	0.328	0.936	0.833	0.586
Severe PAD	0.643	0.862	0.375	0.949
PAD defined by cuff ABPI	0.579	0.846	0.647	0.805
Severe PAD defined by cuff ABPI	1.00	0.885	0.500	1.000
PAD defined by TBPI	0.542	0.514	0.591	0.463
Severe PAD defined by TBPI	0.889	0.461	0.163	0.972

Satisfaction with BlueDop

Patients were more satisfied with BlueDop (MSS = 4.33) compared to ABPI (MSS = 2.89; mean difference = 0.544, p < .001), and when compared to TBPI (MSS = 3.82, mean difference = 0.579, p < .001).

Operator satisfaction was equivalent between BlueDop (MSS = 4.29) and ABPI (MSS = 2.83, mean difference = 0.250, p = .137) and TBPI (MSS = 1.94, mean difference = 0.300, p = .104).

Discussion

In this prospective observational study of 176 limbs, correlation of absolute values of BlueDop was moderate with cuff-ABPI and was weak with TBPI. BlueDop was poor at predicting PAD/severe PAD (defined using a combination of cuff-ABPI and TBPI or TBPI alone), but when PAD/severe PAD were defined with cuff-ABPI alone, BlueDop’s predictive abilities to diagnose PAD/severe PAD was poor-acceptable. Overall, patients reported greater satisfaction with BlueDop compared to ABPI and TBPI. Operator satisfaction with BlueDop compared to ABPI and TBPI was equivocal.

Whilst ABPI and TBPI are considered the gold standard for screening and diagnosis of PAD, both are flawed.¹⁴ There are some data available to support individual non-invasive tests in the diagnosis of PAD; however, there is no perfect method.¹⁴ A normal ABPI cannot reliably exclude PAD, however, false positive rates are generally low and therefore it can aid in screening.¹⁴ Alternatively, a normal TBPI makes a diagnosis of PAD less likely.¹⁴ Variable diagnostic reliability and a lack of standardised normal values for TBPI could explain the weaker correlation between TBPI and BlueDop in this study.⁵ An additional test that could be considered for future comparison is pulse volume recording (PVR). BlueDop and PVR have similarities in utilising waveform analysis to estimate perfusion,¹⁵ whereas PVR measures the waveform response to inflation of a cuff. It is not routinely measured in the Aneurin Bevan University Health Board, therefore was not chosen for comparison in this study. As there is no perfect non-invasive screening test for PAD, there is no perfect comparison for BlueDop. This could explain the correlation findings in this study compared to Korszadeh et al, who document an accuracy of 92%, sensitivity of 95% and specificity of 90% when compared to Duplex sonography.¹² Moreover, all tests are operator dependent which can lead to variation in results. The sensitivity, specificity, NPV and PPV reported in this study vary in part due to our method of producing these values to optimise both sensitivity and specificity which is susceptible to change due to changes in the ROC curves’ shapes.

A greater patient satisfaction with BlueDop compared to ABPI and TBPI was demonstrated in this study. This is understandable and aligns in part to qualitative research that highlighted that patients with PAD and CLTI can have poor experience of cuff-ABPI due to factors such as tissue loss and pain.⁸ A method of evaluating limb perfusion that is viewed more favourably by patients could positively influence patient compliance with screening. Operator satisfaction of BlueDop compared to ABPI/TBPI was equivocal, however, the cohort of operators in this study are frequently using ABPI/TBPI in the vascular and podiatry clinic. Overall, operators find all methods acceptable.

A strength of this study is the prospective design. Furthermore, inclusivity in terms of patients with a variety of underlying pathologies, such as diabetes, reflects real work practice making this study clinically relevant. In addition, comparison of BlueDop to ABPI and TBPI makes this study applicable to current practice.

Limitations of this study include lack of blinding, and potential reporting bias as patients submitted responses directly to the operator, potentially impacting subjective outcomes. Patients who did not have both ABPI/TBPI and BlueDop could not be included in comparative analysis. The factors preventing assessment with ABPI and TBPI included wounds or lack of digits meaning a significant proportion of patients presenting to such services were excluded, limiting the applicability of the results. The sample size for this study was not determined by a power calculation. To the best of our knowledge, there are no publicly available data comparing BlueDop to ABPI or TBPI that would inform such calculations. We cannot determine BlueDop’s usefulness in comparison with other tests, such as Duplex sonography, CT or MR angiography, or the gold standard: digital subtraction angiography, which are accepted to have far greater sensitivity and specificity than ABPI or TBPI. Further comparison of BlueDop with these imaging modalities in addition to clinical outcomes would provide valuable information on its efficacy. Whilst operator satisfaction with BlueDop in this study was equivalent to other methods, we must acknowledge that they were all well versed in assessing limb perfusion with cuff-ABPI and TBPI; therefore, we cannot assert that non-specialist users would report the same.

BlueDop’s ability to better predict more severe PAD suggests that it could be best placed in the vascular clinic for initial assessment of patients with suspected CLTI, or perhaps in the podiatry clinic which frequently assesses patients with high risk lower limbs. It would need to achieve reasonable sensitivity in predicting patients with severe PAD as the corresponding specificity was low which is a limitation since the number of false positives will be relatively high. Ultimately, further evidence demonstrating a correlation of BlueDop with mild to moderate PAD would be required in order to support introduction into a non-specialist setting. Features of CLTI, such as significant rest pain and tissue loss, might also make the cuffless feature even more appealing compared to cuff-ABPI or TBPI for this patient group.

Supplemental Material

Supplemental Material - Can BlueDop be used as an alternative for diagnosing peripheral arterial disease? A prospective comparison of traditional ABPI and TBPI with BlueDop

Supplemental Material for Can BlueDop be used as an alternative for diagnosing peripheral arterial disease? A prospective comparison of traditional ABPI and TBPI with BlueDop by Lucy Fligelstone, Brenig Llwyd Gwilym, Melissa Blow, Julie Lee, Annie Clothier, Kristian Glover, Tracey Hutchings and David C Bosanquet in Vascular

Footnotes

Present address: Sunshine Coast University Hospital, 6 Doherty Street, Birtinya, QLD 4575

Author’s note

This work has been presented at the Vascular Societies’ Annual Scientific Meeting in Brighton (2022) and Dublin (2023).

Declaration of conflicting interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: Mr David Bosanquet has received funding from All Wales Diabetes Implementation Group (AWDIG) and the Royal College of England (Pump Priming Grant) to purchase the BlueDop equipment. Neither he nor any other member of the team have received any funding from BlueDop Medical at any point.

ORCID iDs

Lucy Fligelstone

David C Bosanquet

Supplemental Material

Supplemental material for this article is available online.

References

Peach

Griffin

Jones

, et al. Diagnosis and management of peripheral arterial disease. BMJ 2012; 345(aug14 1): e5208.

Conte

Bradbury

Kolh

, et al. Global vascular guidelines on the management of chronic limb-threatening ischemia. Eur J Vasc Endovasc Surg 2019; 58(1S): S1–S109.e33.

Abu Dabrh

Steffen

Undavalli

, et al. The natural history of untreated severe or critical limb ischemia. J Vasc Surg 2015; 62(6): 1642–1651.

Nordanstig

Behrendt

Baumgartner

, et al. Editor’s choice -- European society for vascular Surgery (ESVS) 2024 clinical practice guidelines on the management of asymptomatic lower limb peripheral arterial disease and intermittent claudication. Eur J Vasc Endovasc Surg 2024; 67(1): 9–96.

Fitridge

Chuter

Mills

, et al. The intersocietal IWGDF, ESVS, SVS guidelines on peripheral artery disease in people with diabetes mellitus and a foot ulcer. J Vasc Surg 2023; 78(5): 1101–1131.

Trevethan

. Subjecting the ankle-brachial index to timely scrutiny: is it time to say goodbye to the ABI? Scand J Clin Lab Invest 2018; 78(1–2): 94–101.

Davies

Kenkre

Williams

. Current utility of the ankle-brachial index (ABI) in general practice: implications for its use in cardiovascular disease screening. BMC Fam Pract 2014; 15(69): 69.

Scott

Lecouturier

Rousseau

, et al. Nurses’ and patients’ experiences and preferences of the ankle-brachial pressure index and multi-site photoplethysmography for the diagnosis of peripheral arterial disease: a qualitative study. PLoS One 2019; 14(11): e0224546.

Potier

Abi Khalil

Mohammedi

, et al. Use and utility of ankle brachial index in patients with diabetes. Eur J Vasc Endovasc Surg 2011; 41(1): 110–116.

10.

Hinchliffe

Brownrigg

JRW

Apelqvist

, et al. IWGDF guidance on the diagnosis, prognosis and management of peripheral artery disease in patients with foot ulcers in diabetes. Diabetes Metab Res Rev 2016; 32(S1): 37–44.

11.

BlueDop Medical . BlueDop: how it works. [cited 2023 Jul 6]. Available from: https://www.bluedop.com/how-it-works

12.

Kordzadeh

Hoff

Tokidis

, et al. Novel assessment (BlueDop) device for detection of lower limb arterial disease: a prospective comparative study. J Ultrasound Med 2018; 37(3): 763–768.

13.

von Elm

Altman

Egger

, et al. The Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) statement: guidelines for reporting observational studies. J Clin Epidemiol 2008; 61(4): 344–349.

14.

Forsythe

Apelqvist

Boyko

, et al. Effectiveness of bedside investigations to diagnose peripheral artery disease among people with diabetes mellitus: a systematic review. Diabetes Metab Res Rev 2020; 36(S1): e3277.

15.

Wang

Brooke

Black

. Clinical vascular examination. Image-Guided Interventions. Image-Guided Interventions: Expert Radiology Series, Third Edition: Elsevier, 2020, p. 16–23.e1.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.06 MB