Usefulness of a Novel Vascular Access Management Method Using a Laser Blood Flowmeter

Vascular access (VA) is an essential lifeline for hemodialysis patients, and carrying out maintenance dialysis (mHD) is difficult without VA. Currently, autologous arteriovenous fistula (AVF) accounts for 90% of all VA types used in Japan. Therefore, VA and AVF are essentially synonymous in Japan [1, 2].

In April 2016, the Japan Association for Clinical Engineers issued “Guidelines for the Daily Management of Vascular Access for Clinical Engineers” [3]. The publication of these guidelines also served to highlight the extreme importance of VA management to other healthcare professionals involved in dialysis treatment. Various methods for VA management continue to be studied and implemented.

With the aging of the patient population receiving dialysis and the growing number of patients with diabetic nephropathy [1, 2], the importance of patient monitoring during dialysis treatment is increasing. Therefore, various monitoring devices have been developed and released.

Hemodialysis has a significant impact on blood pressure reduction during hemodialysis [4, 5]. Thus, obtaining biological information on microcirculation is becoming increasingly important, as evidenced by recent reports demonstrating the usefulness of microcirculation monitoring for detecting hypotension during dialysis and foot care [6‒9]. One such newly launched device is the Pocket LDF laser blood flowmeter (JMS, Tokyo, Japan; hereinafter, simply “LDF”), which allows for the noninvasive and real-time monitoring of microvascular blood flow. The LDF uses the Doppler effect, in which an 850 nm near-infrared light is irradiated from a semiconductor laser to tissue below the surface of the skin, and microvascular blood flow in the irradiated area is measured by receiving the backscattered light. The LDF sensors can be attached to peripheral sites, such as the fingertips and earlobes, using dedicated clips.

AVF reduces arterial blood flow to the periphery. Therefore, it is considered that the peripheral circulation flow in the AVF limb is reduced. Theoretically, it can be inferred that the ratio of peripheral circulation flow in the AVF limb to peripheral circulation flow in the non-shunt limb is ≤1 less. On the other hand, if there is stenosis in the VA vein, it is thought that the blood flow will be in stasis and the peripheral circulation flow will increase in the AVF limb. As a result, it can be inferred that the peripheral circulation flow ratio will be greater than 1. In this study, we investigated the usefulness of a new, simple, and quantitative VA management method using the LDF for monitoring mHD patients having an AVF.

This study involved 82 patients (43 men, 39 women; mean age 70.4 ± 9.2 years) receiving mHD via an AVF at our hospital. Patient information is summarized in Table 1.

The measurement procedure was as follows. In the dialysis room, the patient lay on the bed in the supine position, and the surface temperature of both hands was measured. Temperature correction was performed if the difference in surface temperature between the hands was >0.5°C or if the temperature could not be measured. Then, blood pressure was measured, an LDF sensor was attached to the third finger of each hand, and peripheral circulation flow was measured for 4 min to calculate mean blood flow (Fig. 1). The shunt symmetry index (SSI) was calculated for microcirculation evaluation using the following formula:

SSI = Peripheral circulation flow in the AVF limb/Peripheral circulation flow in the non-AVF limb.

We divided patients into those with a forearm AVF and those with an elbow AVF in order to compare SSI by AVF site, using the unpaired t test. According to usual procedure for AVF management at our hospital, we have determined which patients undergo vascular access interventional therapy (VAIVT) as follows. A total of 82 patients are undergoing maintenance hemodialysis at our hospital. Of these, 44 patients who had been regularly checked or who were suspected of having AVF stenosis based on auscultatory findings, high venous pressure, prolonged hemostasis time after dialysis session, or swelling in the AVF limb, underwent ultrasound evaluation (Doppler ultrasonography). As a result, we determined 10 patients as having indication for VAIVT due to AVF stenosis using ultrasound.

For the patients who undergo VAIVT, we performed ultrasound evaluation of the AVF before and after VAIVT, and we measured pre-dialysis SSI on three different occasions before and after VAIVT to evaluate the change in SSI. We also examined the relationship between the vascular resistance index (RI), which was derived from periodic ultrasound evaluations of AVF, and SSI.

We used the unpaired t test to examine differences in SSI between patients who required VAIVT and those who did not. Receiver operating characteristic curve analysis was also performed to determine the cutoff value for distinguishing VA stenosis by SSI value. In all analyses, statistical significance was set at p < 0.05.

Of 82 patients receiving mHD at our hospital, 70 patients (85.4%) showed lower blood flow in the AVF limb than in the non-AVF limb. There was no significant difference in SSI by AVF site (p = 0.96; Fig. 2).

Changes in SSI values in patients who underwent VAIVT are shown in Figure 3. In all 10 patients, SSI values were >1.0 before VAIVT but decreased to <1.0 after VAIVT. In case 1, with the highest pre-VAIVT SSI, the three pre-VAIVT SSI values were 2.20, 1.44, and 1.43, and the corresponding post-VAIVT time points were 0.23, 0.63, and 0.41. On AVF ultrasound, RI changed from 0.7 before VAIVT to 0.57 after VAIVT, and brachial arterial flow volume changed from 337 to 474 mL/min (Table 2). In case 2, with the lowest pre-VAIVT SSI, the pre-VAIVT SSI values were 1.27, 1.06, and 1.20, and the post-VAIVT SSI values were 0.63, 0.69, and 0.75. On AVF ultrasound, RI changed from 0.56 before VAIVT to 0.51 after VAIVT, and flow volume changed from 612 to 858 mL/min (Table 2).

No correlation was found between SSI and RI, with a correlation coefficient of r = −0.09 (p = 0.54; Fig. 4).

The mean SSI of those who did and did not require VAIVT was 1.20 ± 0.49 and 0.65 ± 0.33, respectively, with a significant difference between the two groups (p < 0.001; Fig. 5).

The SSI cutoff determined by receiver operating characteristic analysis was 1.06 (sensitivity: 0.69, specificity: 0.93, area under the curve: 0.81; Fig. 6).

The finding that 70 of 82 patients (85.4%) on mHD had SSI <1.0 suggests that the peripheral circulation flow in the AVF limb was lower than that in the non-AVF limb, with the AVF reducing peripheral circulation flow. In contrast, in patients with SSI >1.0, stenosis of the proximal vein of the AVF caused stasis of blood flow, resulting in an increased volume of blood flowing to the periphery. Thus, patients with an SSI much lower than 1.0 are considered to be at risk of steal syndrome, while those with an SSI much higher than 1.0 are likely to have some disturbances in the proximal vein of the VA anastomosis.

The absence of a significant difference in SSI by AVF site suggests that the AVF site does not have an impact on peripheral circulation flow. In terms of VAIVT, all of the patients who underwent VAIVT showed a decrease in SSI to <1.0 after VAIVT, probably due to the improvement of stenosis after VAIVT and the subsequent decrease in peripheral circulation flow in the AVF limb. The significant difference in SSI between those who did and did not require VAIVT indicates that, as mentioned earlier, the SSI is significantly affected by the presence of stenosis in the proximal vein of the VA anastomosis.

Currently available VA management methods include shunt trouble scoring (STS) and AVF ultrasound [10]. STS is a check sheet devised by Ikeda et al. [10] that has check items before, during, and after dialysis, scores the presence or absence of each, and is used according to the VA management of our hospital. It is done by a nurse once a month. The items are the presence or absence of stenotic sound by auscultation, palpation, increase in venous pressure, etc., and AVF ultrasound is performed and reported to the doctor in case of with 3 points or more. The average time taken for STS assessment at our hospital is 3.5 min. Although STS can be performed easily without using special equipment, the results are greatly influenced by the examiner’s subjective judgment, and we therefore do not use the results for confirmatory diagnosis at our hospital. AVF ultrasound can directly evaluate how the patient’s VA is functioning and is therefore used in many institutions for VA evaluation. In particular, ultrasound evaluation of hemodynamic function is recommended in the guidelines published by the Japanese Society for Dialysis Therapy. The average time taken by a clinical engineer to perform AVF ultrasound at our hospital is 23.5 min (23.5 ± 6.1 min) and is dependent on the examiner’s skill and the condition of the VA being evaluated. For AVF ultrasound to be completed in a short time like STS, various considerations must be met, such as considerable practice and procedural standardization, to avoid individual examiner skills influencing the examination results. On the other hand, the results of peripheral circulation flow using the LDF do not depend on the examiner’s skill and are consistent across examiners, provided that the established measurement procedure is followed. The average time taken for such measurement at our hospital is 6 min, which is much shorter than the average 23.5 min for AVF ultrasound. At our hospital, both STS and AVF ultrasound are done before starting a hemodialysis session, so taking considerably longer for VA assessment may delay the session. Naturally, the shorter the interval between hospital arrival and the start of hemodialysis, the more beneficial it is for the patient. It may also help address concerns that delayed initiation may stress patients.

The RI value obtained by AVF ultrasound is widely used as a measure of resistance to blood flow through the AVF [11], and it is also often used as a guide to determine the need for VAIVT. Although the brachial artery is specifically used to calculate RI, the condition of the site of ultrasound waveform is known to affect RI. In particular, when there is calcification, RI can be used as reference information only, and the actual physical findings need to be analyzed in detail. The same applies to atherosclerosis; there are reports that RI changes in response to external stress experienced by patients [12, 13]. The SSI, on the other hand, is not affected by the condition of the blood vessels because it is calculated from peripheral circulation flow. SSI measures whether there is a stenosis in an AVF blood vessel. We found no relationship between RI and SSI in this study, probably because different sites were used for calculation of the two indices. From the above, SSI is a different index than RI. We believe that it is important for VA management to make comprehensive judgments from all perspectives, including RI, SSI, and physical findings, rather than making a single judgment.

AVF ultrasound is indispensable for VA evaluation because it allows us to observe not only the actual physical state of the AVF but also the vessel properties and stenosis in detail [14]. However, the procedure requires a certain level of proficiency. By comparison, SSI assessment is simple, does not require such proficiency, and takes much less time to perform. These benefits suggest that SSI assessment can be useful for VA evaluation. The LDF is beneficial for hemodialysis patients because it can measure vital information continuously and in real time. Additionally, the device’s compact size and ease of operation enable such assessments to be performed by medical personnel.

This study identified the SSI cutoff value of 1.06 for the presence or absence of stenosis. A theoretical cutoff value of 1.0 was considered, but there were patients with values between 1.0 and 1.06. This is thought to be caused by measurement error or the condition of the patient. Some patients had SSI values between 1.0 and 1.06 despite no apparent VA stenosis. The fact that blood flow was directed from the AVF to the deep veins may have influenced the increased vascular resistance in the AVF limb. Given the above, it may be possible to perform VA screening by periodically measuring SSI using the LDF, continuously recording the changes in SSI, and performing AVF ultrasound if SSI changes remarkably or exceeds 1.06.

Our results suggest that screening for the required AVF ultrasound using objective numerical measurements can be performed by simply attaching the LDF device to the fingers of each hand at the bedside.

This clinical study has been approved by the Clinical Research Review Board of Tokai University School of Medicine (approval number: 21R153). All patients provided written consent for inclusion in this study.

The authors have no conflicts of interest to declare.

This research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.

Moe Kojima: research planning, data collection and analysis, and drafting of the manuscript. Naoya Tanabe: date collection and analysis. Yu Kojima: analysis. Koichi Tamura, Hiroo Takahashi, and Jun Ito: revision of the manuscript.

The data that support the findings of this study are not publicly available due to their containing information that could compromise the privacy of research participants but are available from the corresponding author (M.K.) upon reasonable request.