- CGM threshold alerts elevated time in goal glucose vary by 3.9% in a trial of inpatients with diabetes, however predictive alerts offered no additional advantage.
- Time spent in goal glucose vary modifications are thought of clinically significant at 3% in trials.
- Excessive alert frequency and hospital workflow obstacles could clarify why predictive alarms failed to enhance time in vary.
Steady glucose monitoring (CGM) threshold alerts modestly improved inpatient glycemic management, however including predictive alerts offered no additional profit, a three-arm trial confirmed.
Amongst 533 adults with kind 1 or kind 2 diabetes admitted to endocrinology wards, the common time spent in goal glucose vary (TIR; 70-180 mg/dL) was considerably increased amongst sufferers with threshold alerts turned on in contrast with sufferers who had all alerts turned off (adjusted imply distinction 3.9%, 95% CI 0.1%-7.8%, Holm-adjusted P=0.04).
Nevertheless, the imply TIR didn’t differ considerably between sufferers with each threshold and predictive alerts turned on versus these with all alerts off (3.1%, 95% CI -0.8% to six.9%, Holm-adjusted P=0.11).
The outcomes had been constant throughout sensitivity analyses, Jian Zhou, MD, PhD, of Shanghai Jiao Tong College Faculty of Drugs in China, and colleagues reported in JAMA Network Open.
Worldwide consensus on CGM metrics for scientific trials considers a minimal 3% change in imply TIR clinically significant for a remedy group distinction, the authors identified.
“Subsequently, the noticed 3.9% distinction … could characterize a significant change within the distribution of TIR, indicating that the brink alerts of the CGM system had a useful impact on glucose management within the hospital,” Zhou’s group wrote.
On the particular person degree, a distinction of 5% or extra is mostly thought of the brink for being clinically significant, “which has been related to clinically vital advantages in outpatient populations with kind 1 and sort 2 diabetes,” they added.
No consensus has been established for hospitalized sufferers, however the American Diabetes Affiliation recommends continuation of private CGM use throughout hospitalization amongst folks with diabetes. “No research have evaluated the impact of CGM alerts in inpatient settings, the place CGM methods are managed below scientific supervision somewhat than solely for affected person self-care,” Zhou’s group famous.
The researchers recommended a number of elements which may clarify why enabling threshold and predictive alerts concurrently failed to enhance time in vary. Present predictive algorithms could not absolutely account for exogenous insulin administration, they famous. Staffing constraints and excessive alert frequency in hospital settings additionally could have created obstacles to immediate scientific response. Alarm fatigue was not evaluated, however the threshold-plus-predictive group logged practically 2,000 extra alerts than the threshold-on/predictive-off group, they added.
In an accompanying commentary, Eileen Faulds, PhD, RN, of the Ohio State College Wexner Medical Heart in Columbus, emphasised that each the examine setting and affected person inhabitants “differ considerably from typical inpatient follow” and should be thought of when decoding generalizability.
“Not like pharmacologic interventions, the effectiveness of alarms relies upon not solely on the expertise itself but in addition on the scientific methods by means of which alarms are delivered, interpreted, and acted upon,” Faulds famous.
Whereas threshold alerts adopted a standardized response pathway with confirmatory point-of-care testing and protocol-driven remedy, predictive alerts had been as a substitute managed by scientific judgment.
“It stays unclear who acquired every alert, who held major duty for responding, what actions had been anticipated following a predictive alert, and the way alerts had been escalated or included into the scientific workflow,” added Faulds. “Future research must also routinely consider and report implementation constancy, that’s, whether or not alarms had been delivered, acknowledged, and acted upon as meant.”
The one-center, open-label SIGNAL trial randomized sufferers in China 1:1:1 to certainly one of three alert methods. All members had been admitted for suboptimal glucose management. Most (62.1%) had been male, common age was 62, and imply HbA1c was 9.2%. The trial ran from April to December 2024.
Two secondary outcomes favored the threshold-only group over the no-alert group: lowered time spent above vary (>180 mg/dL) and a decrease glycemia threat index. Different secondary outcomes — together with time spent under vary (<70 mg/dL and <54 mg/dL) — didn’t differ throughout teams.
Not one of the secondary outcomes confirmed vital variations when evaluating the threshold-plus-predictive group to both cohort.
In subgroup analyses, sufferers not on intensive insulin remedy, these 65 and older, and people with kind 2 diabetes achieved increased TIR within the threshold-only group in comparison with the alerts-off group. No vital remedy impact was noticed amongst sufferers on intensive insulin remedy, these below 65, people with kind 1 diabetes, or when stratified by baseline HbA1c (<9% vs ≥9%) or general insulin use.
Two extreme hypoglycemia occasions occurred within the no-alerts group, one within the threshold-plus-predictive group, and none within the threshold-only group. No circumstances of diabetic ketoacidosis, hyperosmolar hyperglycemic syndrome, or severe antagonistic occasions occurred.
