The study protocol was approved by the Institutional Local Research and Ethical Committee (Universidad de Extremadura, Badajoz, Spain; record number 152/2019). In conducting this study, we complied with the ethical principles of the Declaration of Helsinki [24], including any emendations between 2000 and 2013. All participants provided signed informed consent prior to participating in this study.

We recruited patients with CKD who attended the Fresenius Medical Care dialysis clinic in Plasencia, Spain. A total of 33 patients who met the selection criteria participated. The inclusion criteria were adults 25 years of age or older that received hemodialysis. We sought at least 10 male participants and 10 female participants. The exclusion criteria, which were created according to the ESH-IP2 [21,22], were sustained arrhythmia, circulatory problems where use of the cuff is contraindicated, and pregnancy.

The professional validation team consisted of 2 nurses with senior experience (more than 6 years) in BP measurement. The measurement area was correctly conditioned to a suitable temperature, and factors that could affect the records, such as noise or distractions, were removed [21,22]. All measurements were made in the same room. The color of the room was white.

After dialysis, each patient first reported information regarding their sex, age, height, and dry weight. In addition, we calculated participants’ BMI using Quetelet´s index in kg/m². The circumference of the patient’s arm was measured to ensure that the cuff size was adequate.

Next, patients sat in the measurement room and BP measurements were started after a 10- to 15-minute rest period. Each patient was seated in a standard-size plastic chair with a backrest and armrests.

In total, 9 consecutive measurements were made on each participant with the Omron M3 Intellisense (5) and the iHealth Track (4) as follows [21,22]:

During the measurements, participants remained calm, quiet, seated, and still. Participants kept their backs straight and feet on the floor in a parallel position (ie, without crossing their legs). They rested their arms on a flat surface with their palms facing upwards and their elbows slightly flexed so that their fists were at the height of their hearts.

BP records were made at heart level on the right arm in 31 participants and on the left arm in 2 participants (because of an arteriovenous fistula on the right arm). The standard cuff size (22-32 cm) for the Omron M3 Intellisense was used for all men (20). For women, the standard cuff (22-32 cm) was used for 11 participants, and the large cuff (32-42 cm) was used for 2 participants. Since the iHealth Track has only one cuff size (22-42 cm), all measurements were taken with it. The interval between one measurement and the next was 30 to 60 seconds [22].

All measurements were made for each participant during their hemodialysis appointment, after the dialysis was complete. The relative values were then used to calculate the mean difference between the reference device readings and the test device readings. All participants were receiving hemodialysis in the Fresenius Medical Care dialysis clinic in Plasencia.

We analyzed the data with the software SPSS Statistics, version 19.0 (IBM, Armonk, New York). We reported the findings as mean (SD).

According to the ESH-IP2, the accuracy of a device is based on a comparison between the test device (iHealth Track) and the standard reference device (Omron M3 Intellisense) measurements. For each participant, we first compared measurements BP2, BP4, and BP6 with measurements BP1, BP3, and BP5, respectively, and then compared measurements BP3, BP5, and BP7 with each other. The most favorable comparisons were used.

In our comparisons, we classified differences for both SBP and DBP, separately, by whether paired values were within 5, 10, or 15 mmHg [22], and whether paired values for HR were within 3, 5, or 8 beats per minute (BPM). We determined whether the test device passed the ESH-IP validation protocol. Part 1 of the validation process concerns the number of differences allowed in the specified ranges of each measure (SBP, DBP, and HR) for comparisons of individual measurements between devices (99 measurements) [22]. Part 2 concerns the comparisons between devices of each measure for individual participants (33) [22].

Moreover, we produced Bland-Altman plots [27,28] to display the agreement between the two devices (the iHealth Track and the Omron M3 Intellisense). These plots show the difference between each pair of measurements on the y-axis against the mean of each pair of measurements on the x-axis (for SBP, DBP, and HR).

Of the 34 participants we recruited, 33 completed the study successfully (one was excluded for device failure). The 33 participants included 20 men and 13 women. Table 1 shows a summary of their biometric characteristics.

The iHealth Track BP device validation results were taken in accordance with the ESH-IP2. The mean differences between the reference standard and test devices were 3.27 (SD 2.99) mmHg for SBP and 3.59 (SD 5.28) mmHg for DBP. In 90 out of 99 SBP and 89 out of 99 DBP comparisons between the devices, measurement differences were within 5 mmHg, exceeding the ESH-IP thresholds (>72 comparisons for SBP and >64 comparisons for DBP). Additionally, in 95 out of 99 SBP and 94 out of 99 DBP comparisons between the devices, measurement differences were within 10 mmHg, also exceeding the ESH-IP thresholds (>86 comparisons for SBP and >80 comparisons for DBP). Moreover, in 98 out of 99 SBP and 94 out of 99 DBP comparisons between the devices, measurement differences were within 15 mmHg, which again surpasses the ESH-IP thresholds (>95 comparisons for SBP and >92 comparisons for DBP). Therefore, the iHealth Track passed part 1 of the validation protocol for BP monitoring.

Regarding part 2 of the ESH-IP2, 31 out of 33 participants had at least two of the three comparisons between devices with measurement differences within 5 mmHg for both SBP and DBP, exceeding the ESH-IP threshold (>23 participants). One participant had all three comparisons for both SBP and DBP with measurement differences greater than 5 mmHg, which is less than the ESH-IP maximum of 3 participants. Given these results, the iHealth Track also passed part 2 of the validation protocol for BP. Thus, because the iHealth Track passed parts 1 and 2 of the BP validation protocol, it passed part 3 of the protocol, overall validation.

The validation findings for the iHealth Track HR monitoring device were taken according to the ESH-IP2. The mean difference between the reference standard and test devices was 2.18 BPM (SD 2.75). In comparisons between devices, 81 out of 99 pairs of measurements were within 3 BPM, 91 out of 99 were within 5 BPM, and 96 out of 99 were within 8 BPM. These results indicate that the iHealth Track passed part 1 of the validation protocol for HR.

Regarding part 2 of the ESH-IP2, 28 out of 33 patients had at least two of three comparisons between devices with measurement differences within 3 BPM, which exceeds the ESH-IP threshold (>23 participants). Only 2 participants had all three HR comparisons with measurement differences greater than 3 BPM, which is less than the ESH-IP maximum of 3 participants. Therefore, the iHealth Track passed part 2 of the HR validation protocol and, consequently, part 3 (overall validation) as well.

The Bland-Altman graphs (Figures 1-3) give further information on the performance of the iHealth Track device. The graphs show that the measurement differences between the devices were fairly constant across the ranges of SBP, DBP, and HR.