Un paso más al páncreas artificial 2016

Un paso más cerca al páncreas artificial…

Alex Ramírez Rincón

Fellow Endocrinología, Diabetes y Metabolismo del Adulto

Universidad Pontificia Bolivariana

Clínica Integral de Diabetes

Enero 2015IDEAS

Grupo de Investigación

Que es un páncreas artificial???

Que es un páncreas artificial???

Control de asa cerrada en niveles de glucemia, mediante un sistema que combina: sensor de glucosa, un algoritmo de control y un dispositivo de infusión de insulina.

Diabetes, Vol (60), Nov 2011: 2672

Historia

Calif Med. 1963 Jun;98:325-7

110

- - -100

90

U 80 %% b-

'Lx 70

~60~504040

9 30

20 DIABETIC

10 NORMAL

010 20 30 40 50 60

1 GM. TOLBUTAMIDE TIME IN MINUTES

Chart 3.-Representative blood sugar curves in two diabetic and two normal persons after injections of tolbu-tamide.

possibility of a versatile system for recordinga number of chemical variables simultaneously inpatients requiring intensive study and surveillance.Another possibility is employment of an analogcomputer for continuous analysis of the variationsin voltage taken directly from the recorder duringthe hook-up to experimental subjects. Such studiesof complex fluctuations and interactions may lay thebasis for mathematical models of physiologic func-tions. The procedure is ideally suited to intravenousglucose or tolbutamide tolerance tests.

435 North Roxbury Drive, Beverly Hills.

REFERENCES

1. Ferrari, A., and Kessler, G.: Continuous automaticchemical analysis in vivo, N. Y. Acad. Sc., 87:729-744, 1960.

2. Hoffman, W. S.: A rapid photoelectric method for thedetermination of glucose in blood and urine, J. BioL Chem.,120:51, 1937.

3. Kaplan, N. M.: Tolbutamide tolerance test in carbohy-drate metabolism evaluation, Arch Int. Med., 107:212, 1961.

4. Stewart, R., and Sanislow, C.: Silastic intravenouscatheter, N.E.J.M., 265:1283-1285, 1961.

5. Weller, C., Linder, M., Macaulay, A., Ferrari, A., andKessler, G.: Continuous in vivo determination of blood glu-cose in human subjects, Ann. N. Y. Acad. Sc., 87:658-668,1960.

6. Welsh, G. W., 3rd: Studies of abnormal glucose metab-olism in pregnancy, Diabetes, 9:466, 1960.

.4

VOL. 98, NO. 6 * JUNE 1963 327

Physiologic Monitoring of Blood GlucoseARNOLD HENRY KADISH. M.D., Beverly Hills

THE CONTINUOUS ANALYSIS of chemical fluctuationand interactions within live biologic systems willbecome increasingly important in medicine as timegoes on. Even now, continuous chemical analysis isa practical procedure when a dual lumen catheter isintroduced into the arm veins for continuous sam-pling. This paper deals with continuous blood glu-cose measurements in human volunteers.

Previous studies' 5 had the disadvantage that theyused moderately large amounts of blood. In thestudies here reported the amount needed was greatlyreduced. A double-lumen catheter was used, andthrough the outer lumen 0.6 cc of heparin waspumped to the distal end inside the vein. From thatpoint the mixture of heparin and blood was pumpedat a rate of 0.8 cc per minute through the innertube of the catheter for passage through the Auto-Analyzer. By this method the amount of blood usedwas 0.2 cc per minute or 12 cc per hour instead ofthe 28.8 cc per hour used in previous work. Usingthe smaller amount does not diminish accuracy.Indeed, with a simple manifold change, 0.1 cc ofblood per minute or 6 cc per hour gives accurateresults. Adding an extra dialyzing membrane to thedialyzer makes this possible. The dialyzer is com-posed of a matching pair of transparent plasticplates whose mated surfaces are mirror-grooved toprovide a continuous channel when the plates arebrought into contact. With a cellophane membranesandwiched between them, the plates are clampedtogether, leaving the continuous pathway separatedonly by the semi-permeable membrane. The singledialyzer path is 87 inches long, affording a largesurface presentation to the membrane. This is dou-bled with the use of the double dialyzing procedureand results in a considerable economy in the re-quirement for blood. A proportioning pump per-forms all the functions that ordinarily requiremanual pipetting, weighing and measuring. Itpumps the dilute heparin mixture through the outerlumen of the tube to its end within the vein where itmixes with blood and then quickly transmits it backthrough the inner lumen to the Auto-Analyzer sys-tem. Seven minutes from the time of withdrawal of

From Loma Linda University School of Medicine, 'Department ofInternal Medicine, Los Angeles 33.

Supkorted by Grants from John A. Hartford Foundation and Meta-bolic Dynamics Foundation.

Submitted December 19, 1962.

* The technique of continuous chemical analy-sis in vivo gives promise of real value in medi-cine. Applied to physiologic monitoring of bloodglucose, it provides an invaluable tool for thestudy of labile diabetes and metabolic problems.Its use during surgical procedures and in post-operative management of such patients and inteaching is to be encouraged.

blood through the catheter, the finished result isnoted on the chart paper of the recorder. That ishow long it takes for passage of the blood samplefrom the vein to the pump of the Auto-Analyzer,through the dialyzer membrane and the heating bathto the colorimeter and recorder. With a single dia-lyzing membrane the time is approximately fiveminutes.Types of Catheter Tubing

In the author's initial work, a dual polyethylenetube was used and inserted into an Intracath whichin turn was inserted through a No. 14 needle. TheIntracath tubing was used only once and the doublelumen polyethylene tube was merely inserted towithin one-sixteenth of an inch of the tip of theIntracath tubing. We had considerable success withthis method in continuous monitoring of blood glu-cose. In five cases there was clotting which forcedan interruption of the analytic procedure afterperiods of from 45 minutes to two hours. In studieson 15 volunteers, using a specially constructed dou-ble lumen catheter prepared from Silastic,* intro-duced directly through a No. 17 thin wall needle,no problem of clotting was encountered. Using thistubing, the process of continuous glucose analysiscould go on for many days without difficulty due toclotting.4 Advantages of Silastic over polyethylenetubing are that it is long-lived, is autoclavable andcan be clamped repeatedly. In addition, it seems tobe associated with a tendency to inhibit infectionon its surface, and it does not cause severe reactionphlebitis such as sometimes occurs with polyethylenetubing. Owing to the ease of autoclaving, there is nodanger of the transmission of infectious hepatitissuch as was possible in previous work. Gas steriliza-tion with ethylene oxide will prevent such infectionwith either type of tubing. The one disadvantage of

Prepared according to our specifications and supplied through thecourtesy of Ethel G. Mullison h.D., Dow Corning Center for Aidto Medical Research, Midland, Michigan.

VOL. 98. NO. 6 * JUNE 1963 325

Physiologic Monitoring of Blood GlucoseARNOLD HENRY KADISH. M.D., Beverly Hills

THE CONTINUOUS ANALYSIS of chemical fluctuationand interactions within live biologic systems willbecome increasingly important in medicine as timegoes on. Even now, continuous chemical analysis isa practical procedure when a dual lumen catheter isintroduced into the arm veins for continuous sam-pling. This paper deals with continuous blood glu-cose measurements in human volunteers.

Previous studies' 5 had the disadvantage that theyused moderately large amounts of blood. In thestudies here reported the amount needed was greatlyreduced. A double-lumen catheter was used, andthrough the outer lumen 0.6 cc of heparin waspumped to the distal end inside the vein. From thatpoint the mixture of heparin and blood was pumpedat a rate of 0.8 cc per minute through the innertube of the catheter for passage through the Auto-Analyzer. By this method the amount of blood usedwas 0.2 cc per minute or 12 cc per hour instead ofthe 28.8 cc per hour used in previous work. Usingthe smaller amount does not diminish accuracy.Indeed, with a simple manifold change, 0.1 cc ofblood per minute or 6 cc per hour gives accurateresults. Adding an extra dialyzing membrane to thedialyzer makes this possible. The dialyzer is com-posed of a matching pair of transparent plasticplates whose mated surfaces are mirror-grooved toprovide a continuous channel when the plates arebrought into contact. With a cellophane membranesandwiched between them, the plates are clampedtogether, leaving the continuous pathway separatedonly by the semi-permeable membrane. The singledialyzer path is 87 inches long, affording a largesurface presentation to the membrane. This is dou-bled with the use of the double dialyzing procedureand results in a considerable economy in the re-quirement for blood. A proportioning pump per-forms all the functions that ordinarily requiremanual pipetting, weighing and measuring. Itpumps the dilute heparin mixture through the outerlumen of the tube to its end within the vein where itmixes with blood and then quickly transmits it backthrough the inner lumen to the Auto-Analyzer sys-tem. Seven minutes from the time of withdrawal of

From Loma Linda University School of Medicine, 'Department ofInternal Medicine, Los Angeles 33.

Supkorted by Grants from John A. Hartford Foundation and Meta-bolic Dynamics Foundation.

Submitted December 19, 1962.

* The technique of continuous chemical analy-sis in vivo gives promise of real value in medi-cine. Applied to physiologic monitoring of bloodglucose, it provides an invaluable tool for thestudy of labile diabetes and metabolic problems.Its use during surgical procedures and in post-operative management of such patients and inteaching is to be encouraged.

blood through the catheter, the finished result isnoted on the chart paper of the recorder. That ishow long it takes for passage of the blood samplefrom the vein to the pump of the Auto-Analyzer,through the dialyzer membrane and the heating bathto the colorimeter and recorder. With a single dia-lyzing membrane the time is approximately fiveminutes.Types of Catheter Tubing

In the author's initial work, a dual polyethylenetube was used and inserted into an Intracath whichin turn was inserted through a No. 14 needle. TheIntracath tubing was used only once and the doublelumen polyethylene tube was merely inserted towithin one-sixteenth of an inch of the tip of theIntracath tubing. We had considerable success withthis method in continuous monitoring of blood glu-cose. In five cases there was clotting which forcedan interruption of the analytic procedure afterperiods of from 45 minutes to two hours. In studieson 15 volunteers, using a specially constructed dou-ble lumen catheter prepared from Silastic,* intro-duced directly through a No. 17 thin wall needle,no problem of clotting was encountered. Using thistubing, the process of continuous glucose analysiscould go on for many days without difficulty due toclotting.4 Advantages of Silastic over polyethylenetubing are that it is long-lived, is autoclavable andcan be clamped repeatedly. In addition, it seems tobe associated with a tendency to inhibit infectionon its surface, and it does not cause severe reactionphlebitis such as sometimes occurs with polyethylenetubing. Owing to the ease of autoclaving, there is nodanger of the transmission of infectious hepatitissuch as was possible in previous work. Gas steriliza-tion with ethylene oxide will prevent such infectionwith either type of tubing. The one disadvantage of

Prepared according to our specifications and supplied through thecourtesy of Ethel G. Mullison h.D., Dow Corning Center for Aidto Medical Research, Midland, Michigan.

VOL. 98. NO. 6 * JUNE 1963 325

Historia

Dr Arnold Kadish (1964), Los Angeles.

En pacientes con DT1.

Regulación externa: con medición de glucemia, e infusión de insulina y glucosa.

Am J Med Electron 1964;3:82–86

Historia

Inspiración (1974-1978).

Albisser, Pfeiffer, Mirouze, Kraegen y Shichiri.

1977 primer dispositivo “comercial”: Biostator.

Artif Organs 1978;2 (Suppl.):247–250

Historia

I977: Biostator.

Otros dispositivos IV.

Nikkiso STG-22 Blood Glucose Controller.

Horm Metab Res 1974;6:339–342

Crit Care Med 2011;39:575–578have documented its benefits (40–42) and charted guide-lines for its clinical use (43,44). Because CGM data are theinput to the AP control algorithm, understanding of thephysical, biochemical, and mathematical principles andlimitations of this technology is critical (45,46).

First, most of the commercial subcutaneous CGM devi-ces measure glucose concentration in a different thanblood compartment—the interstitium. However, duringrapidly changing conditions, e.g., after a meal or duringa hypoglycemic episode, interstitial glucose and bloodglucose can be markedly different (47–49). Thus, CGMdevices require calibration using one or more daily bloodglucose samples. The influence on CGM accuracy of thenumber and timing of calibration points was assessed byseveral studies (50,51). In particular, the DirecNet StudyGroup (52) analyzed changes in accuracy by modifying thecalibration retrospectively and showed that calibratingduring periods of relative glucose stability significantlyimproves CGM accuracy. Modern calibration procedureswere suggested, based on mathematical models of interstitialglucose kinetics (53,54).

Second, time lag exists because of blood-to-interstitialglucose transport and the sensor processing time (in-strument delay). Because such a time lag could greatlyinfluence the accuracy of CGM (55,56), a number ofstudies were dedicated to its investigation (57–61). In moststudies CGM readings lagged blood glucose (most of the

time) by 4–10 min, regardless of the direction of bloodglucose change, but the formulation of the push-pull phe-nomenon brought arguments for a more complex re-lationship than a constant time lag (60). For the purposeof closed-loop control, mitigation of the time lag was sug-gested based on near-term glucose forecast methods(51,61).

Third, errors from transient loss of sensitivity, and ran-dom noise confound CGM data (62–64). Thus, filtering,denoising, and artifact rejection in CGM data are importantfor closed-loop control. Algorithms performing these tasksare available in commercial CGM devices (65–67). Theprecise tuning of filter parameters in an automatic manneris, however, a difficult problem. Advanced methods thatcan be used to resolve this challenge have been recentlyreported (50,53,68–70).

Despite of their inherent limitations, CGM devices pro-duce rich frequently-sampled data sets (e.g., every 5–10min) allowing them to serve as AP-enabling technology(43,45). First steps from simple monitoring to control havealready been taken; modern CGM devices display trendsand blood glucose rate of change and are capable ofalerting the patient about upcoming hypo- or hyperglyce-mia (71–73). Studies of the utility of such alerts have beeninitiated (73–75), and the next logical step—prevention ofhypoglycemia via shutoff of the insulin pump—has beentaken (76).

FIG. 1. The Biostator (courtesy of William Clarke, University of Virginia).

C. COBELLI, E. RENARD, AND B. KOVATCHEV

diabetes.diabetesjournals.org DIABETES, VOL. 60, NOVEMBER 2011 2673

Historia

1978.

“Autosyringe”: 2da bomba comercial.

Vía IV.

Seguridad??? y Comodidad???.

Rechazo y resistencia, por parte de pacientes.


Historia

Que hacer con el paciente ambulatorio???.

Infusión subcutánea continua de insulina???.

Evolución tecnológica paralela.

La Necesidad…… Como artífice de la creatividad.

Historia

1979: Pickup y Tamborlane.

Utilidad de la vía subcutánea en pacientes con DT1.

Shichiri (1982): desarrolló el 1er prototipo de PA con infusor subcutáneo.

BMJ 1978;1: 204–207

N Engl J Med 1979;300:573–578

Historia

1980: Sistema Implantable.

Sensor de glucosa intravenoso.

Infusión de insulina intraperitoneal.

Aplicación clínica limitada.

Diabetes Care 1986; 9:124–128

Diabetes Care 2010;33:121–127

Sistemas de asa cerrada

Sistemas de asa cerrada: sensor intravenoso e infusor peritoneal.

Algorítmos de control: “controladores derivados de proporciones”.

Diferencias con la vía subcutánea.

Diabetes Care 1986; 9:124–128

Diabetes Care 2010;33:121–127

Modelos Predictivos de Control: Modelos Matemáticos del Metabolismo.

Valor de Glucosa +

Tasa de Cambio =

Dosis de Insulina

En la nueva era……

Sensor de glucosa subcutáneo mínimamente invasivo.

Fomentó el sistema SC - SC.

(2006) Viabilidad de ruta subcutánea en el control automático de glucosa en el paciente con DT1 (“Sistema de asa cerrada”).

Diabetes 2006;55:3344–3350

1999

Historia

LIMITATIONS OF INSULIN DELIVERY

Because of its pharmacokinetic and pharmacodynamicadvantages, the intravenous route of insulin delivery hasbeen tested for ambulatory use with implantable devicesfrom the 1970s to the early 1990s (77,78). However, despitetheir effectiveness, the limitations caused by recurrentcatheter complications due to blood clotting stopped thedevelopment of this route of insulin delivery (79). An al-ternative approach to staying close to physiology is to useintraperitoneal insulin delivery, e.g., insulin infusion viathe portal venous system (80–83). In view of closed-loopglucose control, the intraperitoneal infusion route hasseveral intriguing characteristics: reproducibility of insulinabsorption combined with quick time to peak and return tobaseline, close-to-physiological peripheral plasma insulinlevels, and restoration of glucagon response to hypogly-cemia and exercise (84–86). Although the experience ofimplantable programmable insulin pumps from the 1990shas highlighted their benefits including sustained im-provement in mean blood glucose and reductions in glu-cose variability and severe hypoglycemia (87–90), theclinical use of these devices has been limited because ofinsulin aggregation issues (91), increased production ofanti-insulin antibodies, which impair insulin action insome patients (92,93), and the cost associated with thistechnology. Still, the development of less invasive andcheaper implantable ports for intraperitoneal insulindelivery (e.g., DiaPort; Roche Diagnostics, Mannheim,Germany) may extend this option of insulin infusion inview of an AP (94).

Since the late 1990s, continuous subcutaneous insulininfusion (CSII) has become an accepted mode of insulinpump therapy (95). Improvements in safety, miniaturization,

refined tuning of insulin pumps allowing for fine adjust-ments of basal rate, the recently introduced “patch pumps”(96), and new insulin analogs (97,98), all lead to improvedpatient comfort and better glucose control. The key issue ofsubcutaneous insulin delivery remains the delay in actioncaused by the time needed for subcutaneous absorption,resulting in late insulin peaks up to 120 min after the in-jection of a subcutaneous bolus of regular insulin (95).Moreover, with subcutaneous insulin delivery the lostphysiological role of the liver in modulating peripheral in-sulin levels results in higher peripheral insulinemia (82).Whether such a reduced hepatic insulinization impairs thecontrol of hepatic glucose output significantly is still un-clear. In terms of safety, altered absorption at the sub-cutaneous delivery site has been pointed out as a risk forinsulin underdelivery—a major issue observed with CSII,which may result in ketoacidosis if undetected and notcorrected in time (77,99). This phenomenon has beenshown to occur even more rapidly with the use of fast-acting analogs (100).

In an AP setting, CSII combined with subcutaneousglucose sensing has been shown to be effective for “out-of-meal” periods, keeping blood glucose in the normal rangein the postabsorptive state (26,45). However, fully auto-mated closed-loop control has not been so successful inaddressing insulin needs at meals (25,26). Indeed, therapid rise of postmeal glucose is difficult to avert becauseof the inherent delays in subcutaneous insulin absorptionand action (101). As a result, all AP trials reported to dateshow a significant postprandial glucose peak above thenormal range. Moreover, delayed insulin action whilepostprandial blood glucose decreases may result in sec-ondary glucose lows a few hours after the meal (26).

FIG. 2. Key milestones in the timeline of AP progress. EU, Europe; IP, intraperitoneal; NIH, National Institutes of Health; SC, subcutaneous.

REVIEW OF THE ARTIFICIAL PANCREAS

2674 DIABETES, VOL. 60, NOVEMBER 2011 diabetes.diabetesjournals.org



Proyecto ADICOL (Comisión Europea).

Modelo predictivo no lineal del control de la concentración de glucosa en pacientes con DT1.

Estrategias avanzadas para el sistema de asa cerrada.

Physiol Meas 2004;25:905–920

2004


JDRF (2006): Juvenile Diabetes Research Foundation International. Artificial Pancreas Project.

University of Virginia and Padova.

University of California Santa Barbara and Sansum Diabetes Research Institute.

J Diabetes Sci Technol 2008;2:863-872

J Diabetes Sci Technol 2010;4:1374–1381

BMJ 2011;342: d1855

Estrategias de control automatizado

El presente……

Bomba de infusión subcutánea de insulina con monitoreo continuo de glucosa en tiempo real.

J Diabetes Sci Technol 2008;2:863-872

J Diabetes Sci Technol 2010;4:1374–1381

Sistema integrado MiniMed®

Sensor de glucosaBomba de insulina MiniMed® VeoTM

Barreras en el tiempo……

Romper paradigmas.

Creación de nuevos sistemas: teóricos y tecnológicos.

Rangos de normalidad.

Diferencias en tiempos reales vs sensor.

Necesidad de algoritmos de control inteligente

Factores de variabilidad en el sistema.

Regulación de ritmos.

Retroalimentación negativa al infusor.

Rev Biomed Eng 2009;2: 54–96


NEED FOR “SMART” CONTROL ALGORITHMSDespite important developments in sensor and pump tech-nology, the AP must cope with the delays and inaccuraciesin both glucose sensing and insulin delivery described in theprevious sections. This is particularly difficult when a sys-tem disturbance, e.g., a meal, occurs and triggers a rapidglucose rise that is substantially faster than the time neededfor insulin absorption and action (Fig. 3).

The problem is that with inherent delays, any attempt tospeed up the responsiveness of the closed loop may resultin unstable system behavior and system oscillation. Thus,a sound controller design must consider a relatively slowresponse, giving time for the delays to wear off before thenext control action. However, a slow response cannotprovide good attenuation of postprandial glucose peaks.Hence the principal AP control dilemma: find a trade-offbetween slow-pace regulation well suited to mild controlactions applicable to quasi-steady state (e.g., overnight),and postprandial regulation calling for prompt and ener-getic corrections (102,103).

Historically, this problem was clearly demonstrated bythe first closed-loop experiments that used PID algorithms.Because PID is purely reactive, responding to changes inglucose concentration after they have occurred, it suffersmost from the problems described above. In particular,to avoid hypoglycemia after a meal, one has to design amoderately aggressive controller; however, such a “cau-tious” design would not react promptly and effectively tomeals. To improve PID performance, one possibility is toadd a feed-forward action (a regular premeal bolus), which

helps with meal compensation as demonstrated in clinicalstudies (27). To mitigate hypoglycemic events, an insulinnegative feedback on insulin delivery rate has also beenintroduced (104).

The new wave of control designs, MPC, is based onprediction of glucose dynamics using a model of the pa-tient metabolic system and, as a result, appears bettersuited for mitigation of time delays due to subcutaneousglucose sensing and insulin infusion. In addition, MPC isa better platform for incorporation of predictions of theeffects of meals and for introduction of constraints on in-sulin delivery rate and glucose values that safeguardagainst insulin overdose or extreme blood glucose fluctu-ations. In some sense, an MPC algorithm works as a chessstrategy (Fig. 4). On the basis of past game (glucose) his-tory, a several-moves-ahead strategy (insulin infusion rate)is planned, but only the first move (e.g., the next 15-mininsulin infusion) is implemented; after the response of theopponent, the strategy is reassessed, but only the secondmove (the 30-min insulin infusion rate) is implemented,and so on. In reality glucose prediction may be differentfrom the actual glucose measurement or an unexpectedevent may happen; with this strategy these events aretaken into account in the next plan.

Several successful clinical trials using MPC were re-cently published. Kovatchev et al. (31) reported a study inwhich the aggressiveness of the controller was indi-vidualized using patient parameters such as body weight,carbohydrate ratio, and insulin basal rate. Hovorka et al.(28) used a multimodel MPC approach, deciding in real

FIG. 3. Block diagram of closed-loop glucose control. Three major delays are indicated: insulin absorption (regular and ultrafast insulin), insulinaction on peripheral tissues and on the liver, and sensing in the interstitium.



NEED FOR “SMART” CONTROL ALGORITHMSDespite important developments in sensor and pump tech-nology, the AP must cope with the delays and inaccuraciesin both glucose sensing and insulin delivery described in theprevious sections. This is particularly difficult when a sys-tem disturbance, e.g., a meal, occurs and triggers a rapidglucose rise that is substantially faster than the time neededfor insulin absorption and action (Fig. 3).

The problem is that with inherent delays, any attempt tospeed up the responsiveness of the closed loop may resultin unstable system behavior and system oscillation. Thus,a sound controller design must consider a relatively slowresponse, giving time for the delays to wear off before thenext control action. However, a slow response cannotprovide good attenuation of postprandial glucose peaks.Hence the principal AP control dilemma: find a trade-offbetween slow-pace regulation well suited to mild controlactions applicable to quasi-steady state (e.g., overnight),and postprandial regulation calling for prompt and ener-getic corrections (102,103).

Historically, this problem was clearly demonstrated bythe first closed-loop experiments that used PID algorithms.Because PID is purely reactive, responding to changes inglucose concentration after they have occurred, it suffersmost from the problems described above. In particular,to avoid hypoglycemia after a meal, one has to design amoderately aggressive controller; however, such a “cau-tious” design would not react promptly and effectively tomeals. To improve PID performance, one possibility is toadd a feed-forward action (a regular premeal bolus), which

helps with meal compensation as demonstrated in clinicalstudies (27). To mitigate hypoglycemic events, an insulinnegative feedback on insulin delivery rate has also beenintroduced (104).

The new wave of control designs, MPC, is based onprediction of glucose dynamics using a model of the pa-tient metabolic system and, as a result, appears bettersuited for mitigation of time delays due to subcutaneousglucose sensing and insulin infusion. In addition, MPC isa better platform for incorporation of predictions of theeffects of meals and for introduction of constraints on in-sulin delivery rate and glucose values that safeguardagainst insulin overdose or extreme blood glucose fluctu-ations. In some sense, an MPC algorithm works as a chessstrategy (Fig. 4). On the basis of past game (glucose) his-tory, a several-moves-ahead strategy (insulin infusion rate)is planned, but only the first move (e.g., the next 15-mininsulin infusion) is implemented; after the response of theopponent, the strategy is reassessed, but only the secondmove (the 30-min insulin infusion rate) is implemented,and so on. In reality glucose prediction may be differentfrom the actual glucose measurement or an unexpectedevent may happen; with this strategy these events aretaken into account in the next plan.

Several successful clinical trials using MPC were re-cently published. Kovatchev et al. (31) reported a study inwhich the aggressiveness of the controller was indi-vidualized using patient parameters such as body weight,carbohydrate ratio, and insulin basal rate. Hovorka et al.(28) used a multimodel MPC approach, deciding in real

FIG. 3. Block diagram of closed-loop glucose control. Three major delays are indicated: insulin absorption (regular and ultrafast insulin), insulinaction on peripheral tissues and on the liver, and sensing in the interstitium.



Necesidad de algoritmos de control inteligente


Ann. N.Y. Acad. Sci. 1311 (2014) 102–123

time which patient model is best fitting the data in hand,while dual-hormone strategies adding glucagon adminis-tration to insulin delivery to avoid hypoglycemia wereused by El-Khatib et al. (30) and Castle et al. (105).

Additional difficulties that the control algorithm mustface arise from coping with inter- and intrapatient vari-ability. Fortunately, MPC allows for relatively straight-forward individualization using patient-specific modelparameters (106). Given the difficulty of identifyingaccurate individual models, a customizable controllerhas been proposed, individually tuned through a “con-trol aggressiveness” parameter calculated from a fewroutine biometric and clinical data of each individual

(107). In addition, MPC can have “learning” capabilities; ithas been shown that a class of algorithms (known as run-to-run control) can “learn” specifics of patients’ daily routine(e.g., timing of meals) and then optimize the response to asubsequent meal using this information (108,109), or ac-count for circadian fluctuation in insulin resistance, suchas the dawn phenomenon observed in some people (110).

Finally, an MPC system can also have certain feed-forward capabilities, i.e., the ability to use a combinationof feed-forward (e.g., patient-initiated) and feedback(controller-initiated) insulin delivery that can partiallysolve the dilemma posed by the need for trade-off betweenslow-pace regulation in quasi-steady state and prompt

FIG. 4. A: The concept of MPC. At each step, future glucose levels are predicted and insulin delivery strategy is mapped several steps ahead. Then,the first insulin delivery step is implemented, and the situation is reassessed with new glucose data. The process is very similar to a chess game inwhich several moves are planned ahead, and after the implementation of the first move the position is reassessed given the response of the op-ponent. B: The critical stage of the famous chess game between Leonid Stein (white) and Lajos Portisch (black), Stockholm, 1962 (courtesy ofLeon Fahri, University of Virginia).



time which patient model is best fitting the data in hand,while dual-hormone strategies adding glucagon adminis-tration to insulin delivery to avoid hypoglycemia wereused by El-Khatib et al. (30) and Castle et al. (105).

Additional difficulties that the control algorithm mustface arise from coping with inter- and intrapatient vari-ability. Fortunately, MPC allows for relatively straight-forward individualization using patient-specific modelparameters (106). Given the difficulty of identifyingaccurate individual models, a customizable controllerhas been proposed, individually tuned through a “con-trol aggressiveness” parameter calculated from a fewroutine biometric and clinical data of each individual

(107). In addition, MPC can have “learning” capabilities; ithas been shown that a class of algorithms (known as run-to-run control) can “learn” specifics of patients’ daily routine(e.g., timing of meals) and then optimize the response to asubsequent meal using this information (108,109), or ac-count for circadian fluctuation in insulin resistance, suchas the dawn phenomenon observed in some people (110).

Finally, an MPC system can also have certain feed-forward capabilities, i.e., the ability to use a combinationof feed-forward (e.g., patient-initiated) and feedback(controller-initiated) insulin delivery that can partiallysolve the dilemma posed by the need for trade-off betweenslow-pace regulation in quasi-steady state and prompt

FIG. 4. A: The concept of MPC. At each step, future glucose levels are predicted and insulin delivery strategy is mapped several steps ahead. Then,the first insulin delivery step is implemented, and the situation is reassessed with new glucose data. The process is very similar to a chess game inwhich several moves are planned ahead, and after the implementation of the first move the position is reassessed given the response of the op-ponent. B: The critical stage of the famous chess game between Leonid Stein (white) and Lajos Portisch (black), Stockholm, 1962 (courtesy ofLeon Fahri, University of Virginia).



Modelos Predictivos de la Dinámica de la Glucosa

No Insulina Basal Continua

A futuro……

55 ecuaciones diferenciales.

Arquitectura modular.

Uni vs Bihormonal.

Lancet Diabetes Endocrinol. 2015; 3(1):17-26

Nature Reviews Endocrinology 11, 63 (2015)

correction of meals. Associated with such a feed-forwardaction is a nominal glucose profile, which represents theexpected consequence of the conventional patient-initiatedtherapy. The algorithm bases its actions on the differencebetween the sensor signal and this nominal profile. If thedifference is zero, no closed-loop correction is applied andthe patient is subject to the conventional therapy alone. Inpractice the difference will always be nonzero, thus a feed-forward action would also prompt small-size feedbackcorrections adapting to unpredicted events, disturbancesand changes in patient’s dynamics. Clinical results obtainedby this type of control strategy are reported by Kovatchevet al. (31).

A STEP FORWARD: IN SILICO EXPERIMENTS REPLACINGANIMAL TRIALS

The future development of AP will be greatly acceleratedby using mathematical modeling and computer simulation.A number of simulation models have been proposed in thelast 4 decades and used to assess the performance ofcontrol algorithms and insulin infusion routes (111–117).However, all these models are “average,” meaning thatthey are only able to simulate average population dynam-ics but not the interindividual variability. The average-model approach is not sufficient for realistic in silicoexperimentation with control scenarios. For this purpose,it is necessary to have a simulator equipped with a cohortof in silico subjects that spans sufficiently well the ob-served interperson variability of key metabolic parametersin the type 1 diabetic population. The knowledge on inter-subject variability is indeed crucial to the design of robustcontrollers, providing valuable information about theirsafety and limitations.

Building on the large scale meal model developed inthe healthy state (116,117), we have developed a type 1diabetes simulator that, thanks to its ability to realistically

describe intersubject variability, has been accepted by theFood and Drug Administration as a substitute of pre-clinical animal trials for certain insulin treatments (32). Inthis simulator, a virtual human is described as a combinationof several glucose and insulin subsystems. To permitin silico experiments using CGM, the model includessubcutaneous glucose transport and sensor errors. Insummary, the model consists of 13 differential equationsand 35 parameters for each subject (116,117). The simu-lator is equipped with 100 virtual adults, 100 adolescents,and 100 children, spanning the variability of type 1 diabeticpopulation observed in vivo. Key “biometric” characteristicsof these virtual subjects are presented by Kovatchev et al.(32). Figure 5 illustrates the overall design of the simulationmodel.

With this technology, any meal and insulin deliveryscenario can be pilot-tested very efficiently in silico, priorto its clinical application. Because in silico experimentsproduce results at a fraction of time and cost of animaltrials, this simulator in now adopted by the JDRF Artifi-cial Pancreas Consortium and by others as a primary testbed for new closed-loop control algorithms. The capabilitiesof the simulator are now being expanded by incorporatingintraday variability of key fluxes (e.g., glucose production)and signals (e.g., insulin sensitivity [118]), and by includinghypoglycemia counterregulation (119).

MODULAR ARCHITECTURE FOR SEQUENTIAL APDEVELOPMENT

Today’s technological advancements open the possibilityfor ambulatory AP. To account for the multitude of avail-able possibilities, academic, and industrial developments,we have introduced the concept of modular approach toAP design, which allows technologies developed by dif-ferent entities to be seamlessly integrated in a functionalhierarchical system that can be sequentially deployed in

FIG. 5. Principal component of the type 1 diabetes simulator: a model of the glucose-insulin system, a model of the sensor, a model of the insulinpump and subcutaneous insulin kinetics, and the controller to be tested.



Autoregulación

Artificial pancreas review Peyser et al.

Figure 4. (Left) The DiAs smartphone used in recent UVA outpatient artificial pancreas studies with the closed-loop controlalgorithms operating on a smartphone rather than a laptop computer. (Right) A patient holding the DiAs smartphone next to theDexcom CGM and Insulet insulin pump on her abdomen.45

approaches, each comparing open-loop to closed-loop control.

Using the sCTR algorithm, a subsequent studyby the University of Virginia group demonstratedthe feasibility of outpatient closed-loop control us-ing the Dexcom SevenPlus continuous glucose-monitoring system and an Insulet Omnipod insulinpump communicating directly to a specially con-figured Android smartphone designated the DiAs,rather than to a laptop computer.45 A photographof the DiAs platform used in the study is shown inFigure 4.

Phillip et al. have used an FL system to emu-late the expert decision-making algorithms used bymedical personnel in providing treatment recom-mendations to patients.82–84 Mauseth et al. have alsodeveloped an independent FL system using expe-rience gleaned by the algorithm developers fromtheir previous work in the aviation industry.85,86 In2013, Phillip et al. published the results of a clini-cal study using their FL controller, MD-Logic, forovernight or nocturnal closed-loop control in a di-abetes camp.87 The components used in this studywere a Medtronic Enlite continuous glucose mon-itor and a Medtronic Paradigm Veo insulin pumpcommunicating wirelessly with a laptop computer.

The study found that the MD-Logic system wasable to lower the mean overnight glucose from140.4 mg/dL in the open-loop arm to 126.4 mg/dL inthe closed-loop arm while at the same time reducingthe incidence of hypoglycemic events threefold be-tween the two groups. This is an important result fortwo reasons: first, it demonstrated the successful useof an early-stage artificial pancreas device in a trueoutpatient setting, a diabetes camp; and second, itdemonstrated the potential of current artificial pan-creas technology to achieve overnight closed-loopcontrol of glycemia. Glycemic control in the twoarms of the MD-Logic overnight closed-loop con-trol camp study, along with both basal and bolusinsulin dosing, are shown in Figure 5.

There have been two prominent research groupsin recent years that have advocated and evaluatedthe bihormonal approach to closed-loop control:Ward et al. at Oregon Health Sciences University andDamiano et al. at Boston University and the Mas-sachusetts General Hospital. These two approachesdiffer slightly in the frequency and amount ofglucagon delivered. In the Oregon system, glucagonis given relatively infrequently—on average, onlyonce or twice daily—for purposes of reversingovert or impending hypoglycemia. In the Boston

112 Ann. N.Y. Acad. Sci. 1311 (2014) 102–123 C⃝ 2014 New York Academy of Sciences.

Y en Colombia…… En que punto estamos???

Bombas de insulina en el mundo

USA en 2005: 20 a 30% de pacientes con DT1 y < 1% en DT2.

375.000 para el 2007 y estimación de 515.000 para el 2014.

Latinoamerica: 9800, 1200 residen en Colombia. En Medellín - Antioquia 220.

Endocrine Practice Vol 20 N 5 May 2014: 463

Objetivos de manejo con Bomba de infusión

Buen control glucémico.

Reducir tasa de hipoglucemia.

Disminuir ganancia de peso.

Mejorar calidad de vida.

Diab Research and Clinical Practice 2011;93s:S109-S113

Disminución de complicaciones asociadas a la

Diabetes

Esquema Basal/Bolos

NEJM 2012; 366; 17: 1616 - 1624

T h e n e w e ngl a nd j o u r na l o f m e dic i n e

n engl j med 366;17 nejm.org april 26, 20121618

cated hemoglobin levels is accompanied by a 10 to 20% reduction in the dose of insulin.26-29 Among patients who switch from insulin injections to insulin-pump therapy, the most important deter-

minant of the benefit of pump therapy with re-spect to glycemic control is the baseline glycated hemoglobin level, with the greatest effect seen in patients with the worst control at baseline.11,28,30

A Multiple Daily Insulin Injections

B Insulin-Pump Therapy

0

8:00 12:00 16:00 20:00

Clock time

Clock time

Breakfast Lunch

Fastinghyperglycemia

Unpredictablehypoglycemia

Wide swings inblood glucose

Nocturnalhypoglycemia

Marked dawnphenomenon

High mean bloodglucose = high glycatedhemoglobin

Dinner

Bloo

d G

luco

se (

mm

ol/l

iter)

Insu

lin A

ctio

n

24:00 04:00 08:00

8:00 12:00 16:00 20:00 24:00 04:00 08:00

5

10

15

20

Rapid-actinginsulin injection

Long-actinginsulin injection

0

8:00 12:00 16:00 20:00

Clock time

Clock time

Breakfast Lunch

Decreased glycemic variabilityand hypoglycemia

Constant basal rate

Mealtimebolus Square-wave bolus

for fatty meal Increased basal rate toavoid dawn phenomenon

Lowered mean bloodglucose and glycatedhemoglobin

Dinner

Bloo

d G

luco

se (

mm

ol/l

iter)

Insu

lin D

eliv

ery

24:00 04:00 08:00

8:00 12:00 16:00 20:00 24:00 04:00 08:00

5

10

15

20

04/09/12

AUTHOR PLEASE NOTE:Figure has been redrawn and type has been reset

Please check carefully

AuthorFig #TitleMEDEArtist

Issue date

COLOR FIGURE

Rev9Dr. Pickup

--/--/2011

1

JarhoDaniel MullerThe New England Journal of Medicine

Downloaded from nejm.org on January 27, 2015. For personal use only. No other uses without permission. Copyright © 2012 Massachusetts Medical Society. All rights reserved.

Bomba de infusión

NEJM 2012; 366; 17: 1616 - 1624

T h e n e w e ngl a nd j o u r na l o f m e dic i n e

n engl j med 366;17 nejm.org april 26, 20121618

cated hemoglobin levels is accompanied by a 10 to 20% reduction in the dose of insulin.26-29 Among patients who switch from insulin injections to insulin-pump therapy, the most important deter-

minant of the benefit of pump therapy with re-spect to glycemic control is the baseline glycated hemoglobin level, with the greatest effect seen in patients with the worst control at baseline.11,28,30

A Multiple Daily Insulin Injections

B Insulin-Pump Therapy

0

8:00 12:00 16:00 20:00

Clock time

Clock time

Breakfast Lunch

Fastinghyperglycemia

Unpredictablehypoglycemia

Wide swings inblood glucose

Nocturnalhypoglycemia

Marked dawnphenomenon

High mean bloodglucose = high glycatedhemoglobin

Dinner

Bloo

d G

luco

se (

mm

ol/l

iter)

Insu

lin A

ctio

n

24:00 04:00 08:00

8:00 12:00 16:00 20:00 24:00 04:00 08:00

5

10

15

20

Rapid-actinginsulin injection

Long-actinginsulin injection

0

8:00 12:00 16:00 20:00

Clock time

Clock time

Breakfast Lunch

Decreased glycemic variabilityand hypoglycemia

Constant basal rate

Mealtimebolus Square-wave bolus

for fatty meal Increased basal rate toavoid dawn phenomenon

Lowered mean bloodglucose and glycatedhemoglobin

Dinner

Bloo

d G

luco

se (

mm

ol/l

iter)

Insu

lin D

eliv

ery

24:00 04:00 08:00

8:00 12:00 16:00 20:00 24:00 04:00 08:00

5

10

15

20

04/09/12

AUTHOR PLEASE NOTE:Figure has been redrawn and type has been reset

Please check carefully

AuthorFig #TitleMEDEArtist

Issue date

COLOR FIGURE

Rev9Dr. Pickup

--/--/2011

1

JarhoDaniel MullerThe New England Journal of Medicine

Downloaded from nejm.org on January 27, 2015. For personal use only. No other uses without permission. Copyright © 2012 Massachusetts Medical Society. All rights reserved.

El dispositivo

Procesador.

Reservorio de insulina.

Set de infusión.

Monitoreo continuo de glucosa.

Monitoreo Continuo de Glucosa

Glucosa - Oxidasa / Electrodo.

Glucosa => Peróxido de H y ácido glucorónico.

Rx Electrodo => Electrones => Corriente proporcional a los niveles de glucosa => Valor mg/dl.

Diferencia en tiempo y valor


Sistema de asa cerrada

Infusión continua (# Basales).

Bolos prandiales.

Monitoreo continuo de glucosa.

Suspensión por hipoglucemia.

Alarmas por hiperglucemia.

NEJM 2012; 366; 17: 1616 - 1624

Eficacia del dispositivoEn Diabetes Tipo 1.

Metanalisis: #5 y RCT #30.

HbA1c: diferencia de medias - 0.3%. HbA1c: 7.5 vs 8.1%.

HbA1c < 7% sin hipoglucemia severa o ganancia de peso: en Adultos 34% vs 12%.

Cochrane Database Syst Rev. 2010: CD005103

N Engl J Med. 2010; 363:311-320


Continuous subcutaneous insulin infusion (CSII) versusmultiple insulin injections for type 1 diabetes mellitus

(Review)

Misso ML, Egberts KJ, Page M, O’Connor D, Shaw J

This is a reprint of a Cochrane review, prepared and maintained by The Cochrane Collaboration and published in The Cochrane Library2010, Issue 1

http://www.thecochranelibrary.com

Continuous subcutaneous insulin infusion (CSII) versus multiple insulin injections for type 1 diabetes mellitus (Review)

Copyright © 2010 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.

Eficacia del dispositivoEn Diabetes Tipo 2.

4 RCT´s: estudios de 12 a 52 semanas.

No mejoría significativa en: HbA1c, pérdida de peso ni disminución en requerimientos de insulina.

Potencialmente: sin control metabólico, hipoglucemia severa, obesidad y ancianos.


VentajasMenor excursión glucémica.

Menor número de eventos de hipoglucemia severa.

Tratamiento del fenómeno del alba.

Control glucémico.


Suspensión por hipoglucemia.

Flexibilidad en el manejo.

Calidad de vida.

INDIVIDUALIZACION.

Utilidad en la práctica

clínica……

Caso N 1��7KHUDS\�0DQDJHPHQW�'DVKERDUG 3DJH��RI��

'DWD�6RXUFHV��3DUDGLJP�9HR��&RUUDOHV�-LPHQH]��'DQQ\�/HRQDUGR��

*HQHUDWHG��D�P�

�$YJ�)RRG�%ROXV��8�$YJ�&DUEV��J�3RVW�PHDO�6*�$YJ��PJ�G/�3UH�PHDO�6*�$YJ��PJ�G/�'LQQHU��S�P��S�P��

�$YJ�)RRG�%ROXV��8�$YJ�&DUEV��J�3RVW�PHDO�6*�$YJ��PJ�G/�3UH�PHDO�6*�$YJ��PJ�G/�/XQFK��D�P��S�P��

�$YJ�)RRG�%ROXV��8�$YJ�&DUEV��J�3RVW�PHDO�6*�$YJ��PJ�G/�3UH�PHDO�6*�$YJ��PJ�G/�%UHDNIDVW��D�P��D�P��

:DNH�XS��D�P��D�P�%HGWLPH��S�P��D�P�

%HGWLPH�WR�:DNH�XS

��SHU�GD\+LJK�6*�$ODUPV

��SHU�GD\/RZ�6*�$ODUPV

�G��K�SHU�ZHHN:HDU�'XUDWLRQ

��PJ�G/$YJ�6*

6HQVRU�8VH

(YHU\��GD\V5HV��6LWH�&KDQJH��P�SHU�GD\��7LPH��SHU�GD\/*6��(YHQWV��P�SHU�GD\6XVSHQG�'XUDWLRQ

��8��EROXVHV��2YHUULGH��8��EROXVHV��2YHUULGH��8��EROXVHV��&RUUHFWLRQ��8��EROXVHV��)RRG��8��EROXVHV��%ROXV�:L]DUG��8��EROXVHV��0DQXDO�%ROXVHV��%DVDO�%ROXV�5DWLR��8,QVXOLQ�7''3HU�'D\3XPS�8VH

��S�P��D�P�

��S�P��S�P�7LPH�3HULRG

��D�P��S�P�

+\SHUJO\FHPLF�3DWWHUQV��

��S�P��S�P��

��S�P��S�P��7LPH�3HULRG

��D�P��D�P��

+\SRJO\FHPLF�3DWWHUQV��

��J�SHU�GD\&DUEV�(QWHUHG

��SHU�GD\%*�5HDGLQJV

��(VWLPDWHG�$�&

��PJ�G/$YJ�%*

6WDWLVWLFV

%ROXV�,QVXOLQ��8

��DFWLYH��KU�

%DVDO�5DWH %ROXV�,QVXOLQ

��+RXU�$QDO\VLV�±�6HQVRU��,QVXOLQ��6HWWLQJV

$FWLRQ�3ODQ

�5HYLHZHG�%\

�'DWH��7LPH

Caso N 2ViEDGR��D�P�'HYLFH�6HWWLQJV�6QDSVKRW 0DUWtQH]��-XDQ�'DYLG

��

3DJH��RI��'DWD�6RXUFH��3DUDGLJP�9HR��


2Q%ORRG�*OXFRVH�5HPLQGHU

2II'XDO�6TXDUH��9DULDEOH�

��80D[LPXP�%ROXV

��,QVXOLQ�&RQFHQWUDWLRQ

��$FWLYH�,QVXOLQ�7LPH�K�PP�

J��PJ�G/8QLWV

2Q%ROXV�:L]DUG

%ROXV

��8��+RXU7RWDO ��8��+RXU

7RWDO ��8��+RXU7RWDO

��

��

��

��

��

��

8�KU7,0(

��

��

��

��

��

5DWLR7,0(

��

��

6HQVLWLYLW\7,0(

��

+LJK/RZ7,0(

PJ�G/%*�8QLWV

��7UDQVPLWWHU�,'

2Q6HQVRU

��*UDSK�7LPHRXW��K�PP�

��0LVVHG�'DWD�:HDN�6LJQDO�K�PP�

��$8&�/LPLW��/RZ�_�+LJK

�PJ�G/�

��&DOLEUDWLRQ��$OHUW��5HSHDW�K�PP�

��&DOLEUDWLRQ�5HPLQGHU�K�PP�

2II$XWR�&DOLEUDWLRQ

3DWWHUQ�$��DFWLYH� 3DWWHUQ�%

8WLOLWLHV

6HQVRU

%ORRG�*OXFRVH7DUJHW��PJ�G/�

,QVXOLQ�6HQVLWLYLW\�PJ�G/�SHU�8�

&DUERK\GUDWH�5DWLR�J�8�

��

��

��

��

��

��

8�KU7,0(

��

��

��

��

8�KU7,0(

��$PRXQW

7LPH/RZ�5HVHUYRLU�:DUQLQJ

��/RZ�6XVSHQG��PJ�G/�

9LEUDWH$OHUW�7\SH

��8(QWU\��6WHS�

2II(DV\��$XGLR��%ROXV

3HUFHQW�RI�%DVDO7HPS�%DVDO�7\SH

��8�KU0D[LPXP�%DVDO�5DWH

%DVDO

6WDQGDUG

��

(QG�K�PP�

6WDUW�K�PP�

2II

0LVVHG%ROXV

5HPLQGHU

2II2II5DWH�$OHUW��)DOO�_�5LVH

�PJ�G/�PLQ�

��/RZ�_�+LJK��PLQV�

2II3UHGLFWLYH�$OHUW

��$OHUW�5HSHDW

��

+LJK�PJ�G/�

/RZ�PJ�G/�7,0(

2Q*OXFRVH$OHUWV

1RWHV

ViEDGR��D�P�'HYLFH�6HWWLQJV�6QDSVKRW 0DUWtQH]��-XDQ�'DYLG

��

3DJH��RI��'DWD�6RXUFH��3DUDGLJP�9HR��


2Q%ORRG�*OXFRVH�5HPLQGHU

2II'XDO�6TXDUH��9DULDEOH�

��80D[LPXP�%ROXV

��,QVXOLQ�&RQFHQWUDWLRQ

��$FWLYH�,QVXOLQ�7LPH�K�PP�

J��PJ�G/8QLWV

2Q%ROXV�:L]DUG

%ROXV

��8��+RXU7RWDO ��8��+RXU

7RWDO ��8��+RXU7RWDO

��

��

��

��

��

��

8�KU7,0(

��

��

��

��

��

5DWLR7,0(

��

��

6HQVLWLYLW\7,0(

��

+LJK/RZ7,0(

PJ�G/%*�8QLWV

��7UDQVPLWWHU�,'

2Q6HQVRU

��*UDSK�7LPHRXW��K�PP�

��0LVVHG�'DWD�:HDN�6LJQDO�K�PP�

��$8&�/LPLW��/RZ�_�+LJK

�PJ�G/�

��&DOLEUDWLRQ��$OHUW��5HSHDW�K�PP�

��&DOLEUDWLRQ�5HPLQGHU�K�PP�

2II$XWR�&DOLEUDWLRQ

3DWWHUQ�$��DFWLYH� 3DWWHUQ�%

8WLOLWLHV

6HQVRU

%ORRG�*OXFRVH7DUJHW��PJ�G/�

,QVXOLQ�6HQVLWLYLW\�PJ�G/�SHU�8�

&DUERK\GUDWH�5DWLR�J�8�

��

��

��

��

��

��

8�KU7,0(

��

��

��

��

8�KU7,0(

��$PRXQW

7LPH/RZ�5HVHUYRLU�:DUQLQJ

��/RZ�6XVSHQG��PJ�G/�

9LEUDWH$OHUW�7\SH

��8(QWU\��6WHS�

2II(DV\��$XGLR��%ROXV

3HUFHQW�RI�%DVDO7HPS�%DVDO�7\SH

��8�KU0D[LPXP�%DVDO�5DWH

%DVDO

6WDQGDUG

��

(QG�K�PP�

6WDUW�K�PP�

2II

0LVVHG%ROXV

5HPLQGHU

2II2II5DWH�$OHUW��)DOO�_�5LVH

�PJ�G/�PLQ�

��/RZ�_�+LJK��PLQV�

2II3UHGLFWLYH�$OHUW

��$OHUW�5HSHDW

��

+LJK�PJ�G/�

/RZ�PJ�G/�7,0(

2Q*OXFRVH$OHUWV

1RWHV

Caso N 3��7KHUDS\�0DQDJHPHQW�'DVKERDUG 3DJH��RI��

'DWD�6RXUFHV��3DUDGLJP�9HR��$UDQJR��6HEDVWLDQ��










��PJ�G/$YJ�6*

6HQVRU�8VH

(YHU\��GD\V5HV��6LWH�&KDQJH�P�SHU�GD\��7LPH��SHU�GD\/*6��(YHQWV�P�SHU�GD\6XVSHQG�'XUDWLRQ


��S�P��D�P�

��D�P��S�P�7LPH�3HULRG

��S�P��S�P�


��D�P��D�P��

��D�P��S�P��7LPH�3HULRG

��S�P��S�P��





��PJ�G/$YJ�%*

6WDWLVWLFV




�2QO\�KLJKHVW�SULRULW\�VKRZQ� �0RVW�UHFHQW�SXPS�VHWWLQJV�DUH�GLVSOD\HG


$FWLRQ�3ODQ

�5HYLHZHG�%\

�'DWH��7LPH

��7KHUDS\�0DQDJHPHQW�'DVKERDUG 3DJH��RI��

'DWD�6RXUFHV��3DUDGLJP�9HR��$UDQJR��6HEDVWLDQ��










��PJ�G/$YJ�6*

6HQVRU�8VH

(YHU\��GD\V5HV��6LWH�&KDQJH��P�SHU�GD\��7LPH��SHU�GD\/*6��(YHQWV��P�SHU�GD\6XVSHQG�'XUDWLRQ


7LPH�3HULRG

��D�P��D�P�


��S�P��D�P��

��D�P��D�P��7LPH�3HULRG

��D�P��S�P��





��PJ�G/$YJ�%*

6WDWLVWLFV




�2QO\�KLJKHVW�SULRULW\�VKRZQ� �0RVW�UHFHQW�SXPS�VHWWLQJV�DUH�GLVSOD\HG


$FWLRQ�3ODQ

�5HYLHZHG�%\

�'DWH��7LPH

Caracterización epidemiológica de los pacientes diabéticos en terapia con infusión subcutánea continua de insulina y monitoreo continuo de glucosa en tiempo real en un centro especializado de Medellín

IDEAS


Población

1ero de febrero de 2010 a 31 de mayo de 2014.

Inclusión consecutiva de DT1 y DT2: 174 => 141.

DT1 90.1%. Edad 33.6 para DT2 54 años. Mujeres 70%.

Mediana de duración de diabetes: 13 años.

Al menos 1 episodio de hipoglucemia: 39%.

Grupo de Investigación IDEAS

HbA1c 8.6% +/- 1.46

!

8.53% en DT1 9.12% en DT2

IndicacionesMal control metabólico.

Hipoglucemias frecuentes, nocturnas o asociadas al ejercicio y asintomáticas.

Hipoglucemia severa.

Fenómeno de alba.


Resultados HbA1c


HbA1c Inicial: 8.6% !

HbA1c al final del primer año: 7.5%

IDEAS


[email protected]

Dirección: Cra 43a # 18 Sur 135 Centro Comercial Sao Paulo

Local 537 Teléfono: +57 (4) 604 9564

IDEAS


mailto:[email protected]

Un paso más al páncreas artificial 2016

Health & Medicine

Transcript of Un paso más al páncreas artificial 2016