Introduction:
Anaemia is a prevalent complication in patients with end-stage kidney disease (ESKD) and optimisation of the treatment relies on clinical and haematological parameters. Current Haemoglobin (Hb) level monitoring practice often follows the KDIGO, Renal Association, European Renal Association, or National Institute for Health and Care Excellence, which recommend Hb monitoring every 2–4 weeks for patients starting ESA therapy and every 1–3 months for those who are stable on an ESA or intravenous iron therapy. This warrants regular blood tests and visits to the hospital and often repeat investigations to ascertain the response of the treatment offered.
The Alio SmartPatch™ represents a breakthrough in anaemia management in Chronic Kidney disease and it's a FDA- cleared device, the SmartPatch offers improvement over existing technologies by enabling real- time, non-invasive, and remote monitoring of multiple health parameters, including the world’s first non-invasive monitoring of abnormal potassium via a functioning AV access.
In this study, we aims to evaluate the accuracy of the SmartPatch in measuring Hb and haemltocrit (Hct) levels in haemodialysis patients by comparing its results with those of standard-of-care, blood- based laboratory methods used in different clinical sites
Methods:
This was a prospective, open-labeled, non-randomized study comparing the results from the SmartPatch with the laboratory results during the same haemodialysis session. Adults with capacity to consent and receiving haemodialysis via an arteriovenous fistula (AVF) or arteriovenous graft (AVG) in either upper limb were enrolled in this study.
The SmartPatch was placed over the AV access site prior to the pre-dialysis blood draw. The SmartPatch was programmed to take readings at intervals which coincided with the pre- and post-dialysis session blood draws. After the pre-dialysis blood draw was performed, the subject underwent their routine dialysis session. A post-dialysis blood draw was performed once dialysis was completed and just before the SmartPatch was removed. Each blood sample was analysed for Hb and Hct values by the hospital’s own and CLIA (Clinical Laboratory Improvement Amendments) certified laboratory.
Skin color was classified using the Massey scale: light tones (1) to very dark tones (10). For this analysis, the pigmentation scale was divided into three categories: light (types 1–2), medium (types 3–5), and dark (types 5–10)
Results:
The total study sample was 122 patients, of which 116 dialysed using native AVF and six using AVG. Due to the small sample size of patients (4.91%) with AVG, they were excluded from the final analysis. Each patient was allowed to participate in up to four dialysis sessions, which yielded 282 SmartPatch readings that were used for analysis,
All 116 patients used a native AVF for dialysis access. The left upper arm was the site most often used for access in 65 subjects (56.0%), followed by the left lower arm in 28 (24.1%), the right upper arm in 16 (13.8%), and right lower arm in 7 (6.0%).
Haemoglobin measured by the reference device ranged from 7.28 g/dL to 15.4 g/dL, with a mean of 10.49 g/dL and median of 10.4 g/dL. A pooled RMSD of 1.0 g/dL was achieved across all participants. The 95% confidence interval LoA was -2.00 to 1.86 g/dL. Pearson r correlation of 0.64 was statically significant (p<0.00001) (Figure 3A). The second image in Figure 3B shows the Bland–Altman mean difference and Pearson r plots of all subjects. Bias as shown on the Bland–Altman plot corresponds to a Hb mean difference of -0.04 g/dL
Haematocrit measured by the reference device ranged from 22.4 to 46.7%, with a mean of 32.77% and median of 32.6%. A pooled RMSD of 2.91 percentage points of haematocrit (% Hct) was achieved across all participants. The 95% confidence interval LoA was -5.84 to 5.53 % Hct. Pearson r correlation of 0.66 was statically significant (p<0.00001). Bias as shown on the Bland–Altman plot corresponds to a Hct mean difference of -0.14 %.
The SmartPatch Hb and Hct measurements achieved the primary endpoint of the study, which was a performance goal to measure Hb and Hct to within 95% confidence interval LoA of -6 to 6 percentage points of Hct and -2 to 2 g/dL for Hb at 80% power
Conclusions:
Thee results of this study shows that the FDA-cleared SmartPatch offers ability to remotely manage anaemia management in patients with ESKD, by providing longitudinal, non-invasive, and accurate monitoring of Hb & Hct. The SmartPatch enables a nuanced and responsive approach to managing anaemia. This capability not only promises to improve individual patient outcomes but also supports a shift toward a more patient-centred healthcare model.
As CKD prevalence increases, technologies like the SmartPatch will be important in effectively managing complex chronic conditions, ensuring that patients receive the best possible care.
There are limitations to this evaluation. This analysis was conducted assessing the SmartPatch’s accuracy against the gold standard invasive blood draws, over a finite collection time period during a patient’s dialysis session. Further studies are bring pursued to assess the impact of frequent measurement of Hb & Hct along with vital signs over extended periods and measurement of these metrics from other anatomical sites.
I have potential conflict of interest to disclose.
I act as an advisor to the team developed the smartpatch
I did not use generative AI and AI-assisted technologies in the writing process.