A novel, model-based insulin and nutrition delivery controller for glycemic regulation in critically ill patients

Parenteral glucose and glucose control in the critically ill: a kinetic appraisal

BACKGROUND

We investigated the influence of parenteral glucose infusion on insulin-driven tight glucose control (4.4-6.1 mmol/liter) in the critically ill by appraising kinetic characteristics of the glucoregulatory system.

METHODS

Turnover characteristics of the glucoregulatory system associated with constant 0, 1.2, and 2.4 mg/kg/min parenteral glucose infusion were obtained by literature review and mass-balance calculations.

RESULTS

Without parenteral glucose infusion, the achievement of tight glucose control is hampered by long time delays with an anticipated glucose equilibration half-time (T((1/2))) of 185 min. The constant parenteral glucose infusions of 1.2 and 2.4 mg/kg/min reduce T((1/2)) to 80 and 40 min, respectively. This follows on from the accelerated glucose turnover brought about by the insulin-modulated glucose uptake, which increases in response to increasing exogenous insulin required to achieve tight glucose control. However, large variations exist among glucose turnover characteristics in the critically ill.

CONCLUSIONS

The constant parenteral glucose infusion greater or equal to 2.4 mg/kg/min is expected to simplify the achievement of tight glucose control by reducing system delays and may facilitate the development of more intuitive, efficacious, and safer insulin-titration guidelines.

Overview of glycemic control in critical care: relating performance and clinical results

BACKGROUND

Hyperglycemia is prevalent in critical care and tight control can save lives. Current ad-hoc clinical protocols require significant clinical effort and produce highly variable results. Model-based methods can provide tight, patient specific control, while addressing practical clinical difficulties and dynamic patient evolution. However, tight control remains elusive as there is not enough understanding of the relationship between control performance and clinical outcome.

METHODS

The general problem and performance criteria are defined. The clinical studies performed to date using both ad-hoctitration and model-based methods are reviewed. Studies reporting mortality outcome are analysed in terms of standardized mortality ratio (SMR) and a 95(th) percentile (+/-2sigma) standard error (SE(95%)) to enable better comparison across cohorts.

RESULTS

Model-based control trials lower blood glucose into a 72-110 mg/dL band within 10 hours, have target accuracy over 90%, produce fewer hypoglycemic episodes, and require no additional clinical intervention. Plotting SMR versus SE(95%) shows potentially high correlation (r=0.84) between ICU mortality and tightness of control.

SUMMARY

Model-based methods provide tighter, more adaptable one method fits all solutions, using methods that enable patient-specific modeling and control. Correlation between tightness of control and clinical outcome suggests that performance metrics, such as time in a relevant glycemic band, may provide better guidelines. Overall, compared to the current one size fits all sliding scale and ad-hoc regimens, patient-specific pharmacodynamic and pharmacokinetic model-based, or one method fits all control, utilizing computational and emerging sensor technologies, offers improved treatment and better potential outcomes when treating hyperglycemia in the highly dynamic critically ill patient.

A pilot study of the SPRINT protocol for tight glycemic control in critically Ill patients

BACKGROUND

Stress-induced hyperglycemia is prevalent in critical care, even in patients with no history of diabetes. Increased counter-regulatory hormone response increases gluconeogenesis and effective insulin resistance, which can be exacerbated by drug therapy. Control of blood glucose levels to the 4.0-6.1 mmol/L range has been shown to reduce mortality and improve clinical outcomes. The Specialized Relative Insulin and Nutrition Tables (SPRINT) protocol is a simple alternative intensive care unit protocol for modulating insulin and nutritional input to gain tight blood glucose control in the 4.0-6.1 mmol/L target band. The look-up tables, implemented in a wheel-based format, are used by nurses to determine glycemic control actions based on hourly or 2-hourly blood glucose measurements and nutrition and insulin administration rates.

METHODS

An 11 patient pilot study was conducted comprising 2,152 hours of blood glucose level control using the SPRINT protocol. The patient cohort average Acute Physiology and Chronic Health Evaluation II score was 22, which was higher than previous intensive insulin clinical studies.

RESULTS

Overall, 64% of measurements were in the 4.0-6.1 mmol/L band, 89% in the 4.0-7.0 mmol/L band, and 96% of all measurements in the 4.0-7.75 mmol/L band. The average value was 5.8 +/- 0.9 mmol/L. Only 1.4% of all measurements were below 4 mmol/L, with a minimum of 3.2 mmol/L. The maximum value recorded was 11.8 mmol/L.

CONCLUSIONS

Control of blood glucose level was achieved using a protocol implemented by the nursing staff without the need for physician intervention or interpretation, where control is defined as maximizing time within a desired band. The results led to a high level of support for the SPRINT protocol among clinical staff and acceptance of the frequent measurement requirement for effective control. The ease-of-use of the protocol resulted in minimal noncompliance by clinical staff.

Transient and steady-state euglycemic clamp validation of a model for glycemic control and insulin sensitivity testing

BACKGROUND

There is an urgent need for a simple and accurate measure of insulin sensitivity to diagnose insulin resistance in the general population and quantify changes due to clinical intervention. A new physiological control model of glucose and insulin metabolism is validated with the euglycemic-hyperinsulinemic clamp during steady and transient states.

METHODS

The data consist of n = 60 (15 lean, 15 overweight, 15 obese, and 15 morbidly obese) euglycemic-hyperinsulinemic clamp trials performed on normoglycemic insulin-resistant individuals. The glucose and insulin model is fitted using an integral-based method. Correlations between clamp-derived insulin sensitivity index (ISI) and the model's insulin sensitivity parameter (SI) are obtained during steady and transient states. Results are compared with log-homeostasis model assessment (HOMA), a widely used fasting surrogate for insulin sensitivity.

RESULTS

Correlation between model-based insulin sensitivity, SI, and ISIG (ISI normalized by steady-state glucose) is r = 0.99 (n = 60) at steady state and r = 0.97 at transient state, respectively. Correlations did not significantly change across subgroups, with narrow 95% confidence intervals. Log-HOMA correlations are r=-0.72 to SI and r=-0.71 to ISIG for the overall population but are significantly lower in the subgroups, with wide 95% confidence intervals.

CONCLUSIONS

The model-based insulin sensitivity parameter, SI, highly correlates to ISIG in all subgroups, even when only considering a transient state. The high correlation of SI offers the potential for a short, simple yet highly correlated, model-based assessment of insulin sensitivity that is not currently available.