Received date: December 01, 2011; Accepted date: January 22, 2012; Published date: January 24, 2012
Citation: Yatabe T, Hanazaki K, Yokoyama M (2012) Automating Blood Glucose Control. J Anesthe Clinic Res 3:186 doi: 10.4172/2155- 6148.1000186
Copyright: ©2012 Yatabe T et al,. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Visit for more related articles at Journal of Anesthesia & Clinical Research
Blood glucose control in ICUs and operation rooms has some issues. First, hypoglycemia worsens the neurological prognosis of ICU patients and should definitely be avoided. To avoid hypoglycemic attacks require extensive attention by hospital staff. The two biggest advantages of continuous monitoring are the avoidance of hypoglycemia and the reduced workload for ICU staff. However, to achievement of automating blood glucose control, closed loop continuous glucose monitoring is essential. In this review, topics of blood glucose management in ICUs and operation rooms are introduced and advantages and limitations of closed loop continuous glucose monitoring are discussed.
Intensive care; Insulin; Continuous glucose monitoring; Closed loop system; Glucose variability; Artificial pancreas
Since the 2001 report by Van den Berghe et al., there has been a dramatic increase in interest regarding glycemic control in intensive care unit (ICU) patients . At present, intensive insulin therapy (IIT) is widely used in ICUs to keep blood glucose levels between 80–110 mg/dl. However, compared to conventional glycemic control methods, IIT is associated with a higher frequency of hypoglycemic attacks [2,3]. Thus the benefits of IIT are nullified by the increased occurrence of hypoglycemic attacks, which lead to adverse neurological after effects and complications [4,5]. In 2009, the Normoglycaemia in Intensive Care Evaluation and Survival Using Glucose Algorithm Regulation (NICE-SUGAR) research group reported that keeping blood glucose between 80–108 mg/dl using IIT was associated with a poorer prognosis than that associated with blood glucose between 144–180 mg/dl . These results, however, do not discount the fact that glycemic control is important in ICU patients . Hyperglycemia is also harmful, but neither does this reduce the importance of glycemic control. Hypoglycemic attacks and glycemic control both require extensive attention by hospital staff [8,9]. Thus resolving these issues is vital for improving glycemic control in the ICU. In this review, we provide an overview of the solutions and strategies associated with continuous glucose monitoring.
Prior to the discussion of continuous blood glucose monitoring, we have compiled a brief explanation of glycemic control and target blood glucose levels. Diabetes is common in patients who undergo surgery or require care in the ICU . Regardless of whether the patient has diabetes, surgical stress, sepsis, or severe clinical conditions may increase insulin resistance or cause hyperglycemia due to decreased glucose uptake . Hyperglycemia, particularly in the ICU, can develop due to catecholamines, steroids, or nutritional therapy . Hyperglycemia is the main cause of complications such as vascular endothelial damage, impediment of neutrophil chemotactic activity, and increases in active oxygen levels [11,12]. Lastly, an increase in infectious disease has been shown to be particularly responsible for poor clinical outcomes in a significant number of patients. Therefore, the recommended target blood sugar level for glycemic control was set at 80–110 mg/dl; but since the NICE-SUGAR study the target has been 150–180 mg/dl [6,12,13]. However, there have been few articles on glycemic control during surgery compared the number of studies examining glycemic control in ICU patients. A clear consensus has not been achieved on target blood glucose levels or whether IIT should be used during surgery. Nevertheless, hyperglycemia is clearly linked to surgery-associated infections, and blood glucose levels above 180 mg/ dl should therefore be avoided.
Glycemic control in ICUs and operation rooms have some issues. First, hypoglycemia worsens the neurological prognosis of ICU patients and should definitely be avoided [13,14]. However, during surgery when the patient is sedated or under anesthesia, it is difficult to determine the presence of hypoglycemia using methods other than measuring blood glucose levels. Using IIT, the frequency of hypoglycemia increases 5–20% and it becomes difficult to avoid severe hypoglycemic states . In addition, the presence of infectious diseases such as sepsis further increases the frequency of hypoglycemia . Second, should we recommend tight glycemic control irrespective of whether the patient is diabetic? It is not understood well. At present, the efficacy of IIT in patients with a history of diabetes has not been recognized, and there have even been reports linking it to a dangerous increase in mortality . Therefore, we have to decide target blood glucose level in diabetes patient carefully. Third, blood glucose variability have received significant attention since Egi et al. reported in 2006 that keeping blood glucose variability to a minimum might contribute to improvement of patients outcome . The standard deviation of blood glucose levels was measured among 7049 ICU patients, who were divided into survival and non-survival groups. The levels were 30.6 ± 23.4 mg/dl in the survival group and 41.4 ± 28.8 mg/dl in the non-survival group, which was a significant difference . Acute glucose variability during glucose swings exhibited a more specific triggering effect on oxidative stress than chronic sustained hyperglycemia . These oxidative stresses leaded to several undesirable effects such as endothelium damages . Controlling blood glucose variability is now considered an important component of blood glucose management.
As can be seen from the above, proper ICU glycemic control should involve completely avoiding hypoglycemic events, restricting blood glucose fluctuations, and maintaining blood glucose levels under 180 mg/dl. However, to avoid hypoglycemia and carry out proper glycemic control, frequent blood glucose measurements are necessary. Exactly how frequent these measurements should be taken is not clear, but Surviving Sepsis Campaign Guidelines 2008 (SSCG 2008) recommended that until blood glucose levels and insulin infusion rates are stabilized, measurements should be performed every 1–2 hours, after which they should be taken at 4-hour intervals . Thus patients could require as many as 24 measurements in one day, significantly increasing the workload of ICU staff. The two biggest advantages of continuous monitoring are the avoidance of hypoglycemia and the reduced workload for ICU staff.
Continuous glucose monitoring systems are broadly classified into 2 categories based on where measurements are taken: systems in which sensors are placed in subcutaneous tissue, and those in which measurements are taken form arterial or venous blood. In patients with type 1 diabetes, several different devices are being used to prevent hypoglycemia and improve the accuracy of blood glucose monitoring. These devices measure glucose concentration in the interstitial fluid using sensors placed in the subcutaneous tissue . However, these devices measure glucose levels in interstitial fluid, not blood, and thus involve problems such as a time lag of about 20 minutes, and the need to wait several hours for stabilization before measurements can begin . Moreover, in the ICU or during surgery, there is often instability of peripheral circulation, and edema due to fluid imbalance is common. These situations are even more pronounced in critically ill patients for whom glycemic control is most important. Under these circumstances, declines in measurement accuracy are cause for concern . Therefore, in such situations the most current technique is to use a special catheter to continuously draw blood. Several devices that can be placed directly in blood vessels or draw blood with a catheter to continuously measure glucose levels are currently in development . Continuous glucose monitoring devices include those that take measurements every few minutes. Researchers have questioned whether truly continuous glucose monitoring is necessary, but reports have suggested that delays of even 12 minutes in glucose measurement are a factor in hypoglycemia . Moreover, in hepatic ischemia reperfusion during hepatectomy or major vascular surgery involving hepatic circulatory arrest, blood glucose levels can increase 50–100 mg/dl within 5 minutes of hepatic ischemic reperfusion [23-25]. Therefore, real-time continuous glucose monitoring is considered the best option. Below we introduce a fully automated glucose control device, the STG-22, and its next generation model STG-55 (Nikkiso, Tokyo, Japan) (Figure 1).
The artificial pancreas STG-22 continuously draws small amounts of blood at a rate of about 2 ml per hour. Blood glucose is measured via a reaction between the blood and a glucose oxidase membrane. The STG-22 can also carry out automatic glycemic control using a built The artificial pancreas STG-22 continuously draws small amounts of blood at a rate of about 2 ml per hour. Blood glucose is measured via a reaction between the blood and a glucose oxidase membrane. The STG-22 can also carry out automatic glycemic control using a builtin algorithm if blood glucose levels for the start of insulin or glucose administration are set. The STG-22 determines the required dosage of insulin or glucose based on current glucose levels, difference from the target level, and blood glucose fluctuations every minute. In accordance with the infusion rate, insulin or glucose can be automatically injected via the device’s pump. For example, if insulin injection is set to begin at 110 mg/dl, and glucose injection is set to begin at 80 mg/dl, insulin injection starts if blood glucose exceeds 110 mg/dl and is stopped if it drops below 110 mg/dl; neither insulin nor glucose is injected when blood glucose is between 80-110 mg/dl; and glucose is automatically injected at levels below 80 mg/dl. There are no restrictions on glucose administration for total parenteral nutrition or enteral nutrition. In other words, flexible limits can be set for blood glucose, such as the 80–110 mg/dl target recommended by Van den Berghe et al. or the 144–180 mg/dl NICE-SUGAR standard. In systems that continually draw blood from the veins, clot formation may cause catheter occlusion, but the STG-22 contains a fixed amount of heparin on the tip of the catheter, thereby preventing thrombotic occlusion that would interrupt glucose measurements.
The benefits of the STG-22 in glycemic control are as follows:
High accuracy of blood glucose measurement
Blood glucose measurements obtained by the STG-22 were compared to those taken with a blood gas analyzer ABL800FLEX (Radiometer, Copenhagen, Denmark) in scenarios easily affected by dilution from transfusion or bleeding, such as hepatectomy or cardiovascular surgery; and the perioperative period when there are large fluctuations in peripheral circulation  (Figure 2). No influence from bleeding or dilution was detected, and the results of both methods showed a strong correlation. Further, highly accurate results were obtained in ICU patients .
Prevention of hypoglycemia
Hypoglycemia is the biggest problem in blood glucose control. Our institution began using the STG-22 in 2006 and used the closedloop method of glycemic control in more than 200 patients, with zero cases of hypoglycemia . Thus the STG-22 can be used to avoid hypoglycemia. In addition, the device constantly displays blood glucose readings, which eliminates one source of worry for ICU staff.
Reduced workload for ICU staff
Conventional glucose control methods such as sliding scale require frequent blood glucose measurement, which is a burden for ICU staff. In cases involving blood glucose fluctuations, measurements are required every 30 minutes, and IIT requires measurements at intervals of 1–2 h. After measuring blood glucose, insulin often needs to be injected; or in continuous insulin administration, more-frequent adjustments to the infusion rate are required. Nurses dealing with the care of critically ill patients in the ICU are constantly under tension and stress . Under these situations, a lot of stress is caused to ICU nurses because of the frequent measurements of blood glucose levels and the repetitive handling of insulin, which is likely to cause incidents . Although computer-assisted insulin infusion protocols have been used increasingly in recent years, changes in insulin and monitoring of blood glucose levels remains an essential task. In this respect, the STG-22 is better able to reduce staff workloads than the conventional sliding-scale method of glucose control . By using an artificial pancreas, the amount of labor spent per patient was reduced by about 20 min, and stress was substantially relieved. Because this allows nurses to devote more time and concentration on other care-giving tasks, it may contribute to the improvement of the overall quality of care and to the reduction of the risk of occurrence of other incidents .
Glucose control with minimal glucose variability
Blood glucose variability have been linked to prognosis [17,31]. The STG-22 is capable of reducing blood glucose bariablity, allowing a standard deviation of 19.9 ± 10.9 mg/dl . In addition, this system does not require blood glucose measurement by the ICU staff. In other studies [17,32], the standard deviation of blood glucose in survivors and non-survivors were reported to be 30.6–33.3 mg/dl and 41.4–42.1 mg/dl, respectively. Thus, the blood glucose variability associated with the use of STG-22 was lower than the above-mentioned values. Notably, STG-22 achieved SD reduction in patients without causing hypoglycemia .
The STG-22 system has the following disadvantages:
Interruption of glycemic control due to poor blood removal
The STG-22 measures blood glucose levels by drawing blood via a peripheral venous catheter. Therefore, blood sampling may be interrupted due to certain blood vessel conditions, and there are cases where the device cannot be used due to inability to secure peripheral veins . To address this issue, a device capable of taking measurements from a central venous catheter is needed.
Problems of cost and device preparation
The STG-22 device is expensive, and there additional costs involving consumables and maintenance . However, previous study described that the total hospital actual costs, including the costs of using the STG- 22 system, for the original surgical admission of patients in the STG- 22 group ($16,407) were significantly lower than those for patients in the sliding scale group ($21,879; P= 0.047) . The main reason of this result was that incidence of surgical site infection in the STG-22 group was significantly lower than that in the sliding scale group . In addition, we have reported that the cost simulations of our hospital’s 5-bed ICU suggested that this device may allow cost cuts of 14,000 US dollars per year in personnel expenses related to nurses . The nextgeneration STG-55 has a more affordable main device and is smaller in size. However, at present this system can only be used in Japan .
This has been a brief summary of how the STG series can be used to automate glycemic control in the ICU and during surgery. Van den Berghe has pointed out the need for a closed loop continuous glucose monitoring system . Automating glycemic control makes it possible to avoid hypoglycemia, reduce labor, and reduce blood glucose variability. The STG series introduced here allows for vastly improved, although not perfect, glycemic control (Table 1). Further innovations will allow glycemic control to begin as soon as the switch is turned on and without interruption of blood sampling.
|Sample blood from...||peripheral vein and/or central venous|
|(hybrid type is best)|
|First calibration||within a few minutes|
|Daily calibration||minimum for guaranteeing accuracy
(within 4 times per day)
|Frequency of measurement||continuously, real time|
|Closed-loop control||necessary for automation|
not so expensive
Table 1: Optimal continuous glucose monitor in perioperative period.
Recently, we can use continuous blood glucose monitoring devices in ICU and operation room (Figure 3). Furthermore, device with closed loop system such as the STG series provided us automating glucose control. However, these technologies are not perfect. Therefore, it is hoped that manufacturers will develop a novel optimal closed loop continuous glucose monitoring in the future.