1Navsahyadri Institute of Pharmacy, Naigaon ( nasrapur), Pune, Maharashtra, India
2Principal Of Navsahyadri Institute of Pharmacy, Naigaon, Nasrapur, Pune, Maharashtra, India
3Associate professor Of Navsahyadri Institute of Pharmacy, Naigaon, Nasrapur, Pune, Maharashtra, India
This paper aims to evaluate the use of artificial pancreas in patients with severe coronavirus disease 2019 (COVID-19) who have difficulty regulating their blood glucose (BG) levels. In this study, a total of 1048 arterial blood glucose readings were obtained, 181 under artificial and 867 under normal therapy. Compared to the usual therapy (16.021.5 mg/dl at 133 days), the artificial pancreas' daily HGI was reduced (p = 0.0387). There was no significant difference in the rate of parenteral sugar administration at each BG measurement point (p0.0001). Insulin delivery rate during normal therapy was substantially lower than that during BG measurements under the artificial Pancras (p=0.0001). The rate of enteral glucose administration at BG test points was significantly lower under artificial Pancreas compared to normal therapy (p 0.0001). Patients with severe COVD-19 may have a better prognosis with better BG control with the artificial pancreas. Further research is necessary to determine whether employing an artificial Pancreas to improve blood glucose control can improve the prognoses of critically ill patients.
The artificial pancreas system, or APS, is a cutting-edge new treatment for type 1 diabetic mellitus (T1DM). It is an example of how technological advancements are used to create novel approaches to treating human diseases. (1) People with type 1 diabetes today have access to a variety of treatment choices thanks to the fast advancement of diabetes technologies. (2)One of the top 10 causes of death for people is diabetes, a condition that is now a global health concern. Globally, 463 million persons (9.3%) had diabetes as of late. It is predicted that this figure will increase to 700 million people, or 10.9% of the global population. The cost of diabetes grew not only in terms of healthcare but also in terms of global health spending, which by 2019 had reached over USD 760 billion. By 2030, healthcare spending is expected to reach USD 825 billion, and by 2045, it will rise to USD 845 billion.(3) The main cause of type 1 diabetes mellitus (T1DM), a chronic autoimmune disease, is the death of the pancreatic beta cells that produce insulin. Because of low insulin levels brought on by this damage, hyperglycaemia persists, which is a hallmark of diabetes.(4) In children and teenagers, type 1 diabetes mellitus develops quickly and progresses quickly. It can cause numerous potentially fatal consequences, including diabetic ketoacidosis. Individuals diagnosed with Type 1 diabetes mellitus may experience symptoms such as excessive hunger, increased thirst, frequent urination, bedwetting in children, spontaneous weight loss, blurred vision, irritability or mood swings, fatigue, and weakness.(5) The Food and Drug Administration (FDA) authorized the first automated insulin delivery (AID) systems in 2016, which are referred to as automated closed-loop control (CLC) systems.(6) This device has a closed-loop mechanism made up of a wirelessly connected glucose sensor, an insulin pump, and a mobile device with the control algorithm. An artificial pancreas can replicate the endocrine pancreas' physiological response and return insulin secretion, absorption, and sensitivity to glucose levels to normal.(7)Research has documented the safety and efficacy of continuous glucose monitoring (CGM) in adult COVID-19 patients with severe coronavirus illness.(8) As an underlying condition, patients with coronavirus disease 2019 (COVID-19) frequently have poor glucose tolerance and trouble controlling blood glucose (BG). Patients who need mechanical ventilation for severe COVID-19 pneumonia may experience hyperglycemia as a result of illness stress. Angiotensin-converting enzyme 2 (ACE2) expression is decreased by COVID-19 infection. Moreover, ACE2 expression in cells is impacted by hyperglycemia, and this could influence viral infection.(9)
OTHER NAMES FOR THE ARTIFICIAL PANCREAS SYSTEMS:
1. Automated Insulin-Delivery System
2. Closed-Loop System
3. Bionic Pancreas
HISTORY OF THE ARTIFICIAL PANCREAS SYSTEMS:
The current CGM is operated by attaching a sensor to the subcutaneous tissue and measuring glucose in the interstitial fluid at 1- to 5-minute intervals, despite the development of numerous non-invasive blood glucose monitoring systems. The CGM was connected to an insulin pump prior to the development of portable electronics like cell phones. In the late 2000s, sensor augmented pumps (SAPs) were created that connect an insulin pump to a continuous glucose monitor (CGM) and show glucose data. Clinical trials were conducted to sequentially develop and validate two SAPs: one with a low glucose suspension (LGS) function that stops insulin infusion in hypoglycaemia cases, and another with a predictive low glucose suspension (PLGS) function that predicts hypoglycaemia and stops infusions before it occurs. These were not closed-loop systems because the insulin dose had to be manually calculated up until that time. The first closed-loop system was created early in the new millennium and ran on a control algorithm based on a personal computer (PC). It was evolved into a portable form between the beginning and the middle of the 2010s utilizing a control algorithm that could be placed on a smartphone or the pump itself. a system that is hybrid closed-loop (HCL). Since the late 2010s, the HCL has been made available for purchase and put to use. The most recent HCL to be developed and brought to market was advanced HCL (single-hormone), which demonstrated better glycaemic control than the previous HCL. It also had an extra auto-correction bolus function that regularly corrects blood glucose levels that exceed the target.(10)
Fig.1.Timeline of development of the artificial pancreas system. IV, intravenous; SubQ, subcutaneous; CGM, continuous glucose monitoring system; SAP, sensor augmented pump; LGS, low glucose suspension; PLGS, predictive low glucose suspension; HCL, hybrid closed-loop; CE, Conformité Européenne; APS, artificial pancreas system; AHCL, advanced hybrid closed-loop.
HOW DO ARTIFICIAL PANCREAS SYSTEMS WORK:
Three devices make up an artificial pancreas system.
A small sensor is implanted under the skin to measure blood glucose levels every few minutes with a continuous glucose monitor (CGM). The information is wirelessly transmitted by the sensor to an insulin infusion pump or a smartphone app. When insulin must be given, the program determines how much is required and sends a signal to the insulin infusion pump. When your blood glucose is not within the desired range, the insulin infusion pump will dispense little amounts of insulin all day long. Different kinds of insulin pumps exist. One kind of pump is carried externally, either in a pocket or bag or fastened to a belt. Insulin is transferred from the pump to a smaller tube known as a catheter by means of a plastic tube. The catheter contains a needle that is put under the skin and remains there for few days. Another kind of pump delivers insulin through a catheter put beneath the skin and adheres to the skin directly with an adhesive pad. A pump of this type is changed every few days.(11)
Fig. 2. An artificial pancreas system uses a continuous glucose monitor, an insulin pump, and a program stored on the pump or a smartphone (top). The insulin pump can be worn on a belt, stored in a pocket, or attached directly to the skin (bottom)
WHAT ARE THE DIFFERENT TYPES OF ARTIFICIAL PANCREAS SYSTEMS:
There are several types of artificial pancreas systems. They include
DUAL HORMONE SYSTEMS:
A promising new treatment is bihormonal artificial pancreas devices, which can calculate and deliver insulin and glucagon infusions autonomously and detect glucose concentrations continually. Individuals with type 1 diabetes.(12)The reduction of clinically significant hyperglycaemic and hypoglycaemic episodes in T1D patients is a goal of dual-hormone artificial pancreas systems, which deliver both insulin and glucagon. These systems have been proposed as a viable alternative to single-hormone (insulin-alone) closed-loop systems in order to further improve glucose control.(13)The danger of hypoglycaemia is increased by large positive sensor deviations Since they could need to supply data to the control loop, people with T1D are both the plant to manage and the plant operator.om the true glucose value, while the risk of hyperglycaemia is increased by sensor underreadings due to insufficient insulin supply. Furthermore, in a trial using a bi-hormonal artificial pancreas, glucagon was less successful in preventing hypoglycaemia when delivery was postponed due to positive sensor deviations.(14)
INSULINE -ONLY SYSTEMS:
Subcutaneous insulin pumps, interstitial glucose monitoring, and ever-more-advanced algorithms are used in these contemporary closed-loop systems. The amount of data demonstrating the effectiveness of hybrid closed-loop systems that are now commercially available has increased along with their number. The creation of fully closed-loop systems that do not require human input for meal announcements or carb tracking is one of the upcoming problems in closed-loop technology.(15) Because they might need to offer information to the control loop, people with T1D are both the plant to control and the plant operator.(16)First commercially marketed hybrid closed-loop system was the Medtronic 670G. The 670G was followed by the 770G (FDA authorized, licensed for age 2 and above) and the 780G (CE marked, licensed for age 7-80 years), both cleared by the U.S. Food and Drug Administration (FDA) and marked for ages 7 and above by Conformitè Européenne (CE). The pump incorporates the PID algorithm developed by Medtronic.(17)
THRESHOLD SUSPEND AND PREDICTIVE SUSPEND SYSTEM:
Systems that mimic an artificial pancreas greatly lessen the daily strain of making decisions. These devices maximize blood glucose control with little effort from the user, extending the duration of time inside the desired range. For instance, the hybrid closed-loop technology will momentarily halt insulin supply in the event of a hypoglycaemia episode during sleep in order to reduce the amount of time spent below the target range. Furthermore, insulin supply is lowered or stopped momentarily by the sensor in the case that it senses a possible hypoglycaemia episode. The system may momentarily raise the basal insulin rate, give a tiny corrective bolus, or do both if the sensor determines that hyperglycaemia is about to occur. In order to replicate the normal functioning of the pancreas, bimodal insulin and glucagon regimens can offer tightly.(18)
METHODS:
Data collection:
The study included fifteen individuals with severe COVID-19 pneumonia who had trouble regulating their blood glucose levels. For the patients, Tokuhira et al. used an artificial pancreas. The individuals were given An intravenous fluid solution containing more than 12% glucose and amino acids was used to deliver parenteral nutrition. The artificial pancreas measures blood glucose levels and controls the rate at which glucose and insulin are infused. 200 mg/dl was established as the goal BG. BG exceeded 200 mg/dl. For severe COVID-19 patients in their ICU, Tokuhira et al. began routine therapy with intravenous insulin infusion in accordance with the BG control guideline. The 180–220 mg/dl target BG range was set. The hyperglycemic threshold was determined to be BG levels >300 mg/dl, whereas the hypoglycemia threshold was defined as BG levels <100>
Statistical analysis:
Tokuhira et al were Analysing the association between arterial blood glucose level and the artificial pancreas's blood glucose data using linear regression analysis. The Wilcoxon rank sum test was used to compare the artificial pancreas and usual therapy. The results were expressed as median and interquartile ranges (IQR). Using the Brown-Forsythe test, the standard deviation (SD) was used to evaluate the distribution of the data. Pearson's chi-square test was used to compare the values of regular therapy with an artificial pancreas. For all analysis, p<0>
RESULTS:
Table 1 displays the characteristics of the patients. The time span of 6.1±4.0 days was observed from the initiation of the artificial pancreas and the ICU hospitalization. The artificial pancreas was used for 51.9±16.8 hours in total. A total of 1048 arterial blood glucose readings were obtained, 181 under artificial pancreas and 867 under normal therapy.
TABLE 1: Patients' characteristics
Note: Numerical data are expressed as mean±SD. Abbreviation: ICU, intensive care unit; SOFA score, sequential organ failure assessment score.
Figure 1 shows the relationship between arterial BG and BG data from the artificial pancreas.
FIGURE 1. Relationship between data from the artificial pancreas and arterial blood glucose (BG). The BG data from the arterial pancreas and arterial BG showed a highly substantial association. With x representing the arterial BG and y representing the BG data from the artificial pancreas, the linear regression equation was y = 0.94x?2.9. 0.618 was the coefficient of determination (p<0>
Figure 2. The linear regression equation for this relationship was y = 0.20x?54.5, where y is the difference between BG data from the artificial pancreas and arterial BG and x is the average of BG data from the artificial pancreas and arterial BG
FIGURE 2: Arteriovenous blood glucose (BG) and BG data from the artificial pancreas are plotted in a Bland-Altman diagram. The disparity between arterial BG and BG data from the artificial pancreas was not significantly impacted by BG levels. The difference between the artery BG and BG data from the artificial pancreas is represented by y, and the average of the arterial BG and BG data from the artificial pancreas is represented by x in the linear regression equation At p = 0.0002, the coefficient of determination was 0.075. The gap between arterial blood glucose and the BG data from the artificial pancreas was marginally larger with higher BG.
Under the artificial pancreas, more points inside the target BG range were reached than under traditional therapy. Table 2
TABLE 2: Achievement of the target BG range
Note: Values were compared using Pearson's chi-square test. Abbreviation: BG, blood glucose. *p <0 xss=removed>
TABLE 3: Incidence of hyperglycemia and hypoglycemia
Note: The values were compared using Pearson's chi-square test. Abbreviation: BG, blood glucose. *p<0>
TABLE 4: Arterial BG, insulin administration rate, speed of parenteral sugar administration, and speed of total sugar administration at each BG measurement point
Note: The values were expressed as median and interquartile ranges (IQR). The medium of each group was compared using Wilcoxon's test. Tests for equality of variance were performed using the Brown–Forsythe test. Abbreviations: BG, blood glucose; SD, standard deviation. *p<0 xss=removed>
FIGURE. 3: Arterial blood glucose distribution (BG). The arterial BG (mg/dl) during conventional therapy is displayed in the upper plots. The arterial BG during the artificial pancreas is displayed in the lower graphs. The daily distribution of arterial BG in both groups is depicted in the left graphs. The whiskers on middle box plots, which display the median and interquartile ranges (IQR), are drawn to be 1.5 times the IQR. Between the two treatments, there was no discernible difference (p = 0.7097).
Figure 4 Displays the insulin delivery rate distribution at each BG measurement point. At every BG measurement point, the artificial pancreas administered insulin at a much higher rate than the usual therapy (p<0>
FIGURE 4: Distribution of insulin delivery rates at sites where blood glucose (BG) is measured. The insulin administration rate (units/h) at BG measurement sites during conventional therapy is displayed in the upper graphs. The artificial pancreas's insulin infusion rate at BG measurement points is displayed in the lower graphs. The daily distribution of insulin delivery rate for both regimens is displayed in the left graphs. Whiskers are drawn within 1.5 times the interquartile range (IQR) in middle box plots, which display the median and IQR. Insulin delivery rate during normal therapy was substantially lower than that during BG measurement points under the artificial pancreas (p<0>
Figure 5 Demonstrates the parenteral sugar injection rate distribution at each BG measurement point. There was no significant difference in the rate of parenteral sugar administration at each BG test point (p = 0.8559).
FIGURE 5: Parenteral sugar administration rate distribution at blood glucose (BG) measuring sites. The parenteral sugar infusion rate (g/h) at BG measurement sites during conventional treatment is displayed in the upper plots. The parenteral sugar infusion rate at BG measurement locations beneath the artificial pancreas is displayed in the lower graphs. Whiskers are drawn within 1.5 times the IQR in middle box plots, which display the median and IQR. The artificial pancreas and conventional therapy did not differ in the rates of parenteral sugar administration.
Figure 6 demonstrates the enteral sugar delivery rate distribution at each BG measurement point. Every BG measurement point under the artificial pancreas had a slightly lower total sugar delivery rate than under normal therapy(p = 0.0078).
FIGURE 6: At blood glucose (BG) measurement sites, the distribution of enteral sugar administration rate (g/h). The enteral sugar delivery rate under standard treatment is displayed at BG measurement points in the upper plots. The enteral sugar delivery rate at BG measurement locations beneath the artificial pancreas is displayed in the lower graphs. The whiskers on middle box plots, which show the median and IQR, are drawn to within 1.5 times the IQR. The artificial pancreas considerably reduced the rate of enteral sugar administration at BG measurement points compared to normal therapy (p<0>
DISCUSSION:
When utilizing the artificial pancreas as opposed to conventional therapy with intravenous insulin infusion, the blood glucose range was higher. The artificial pancreas rarely caused severe hypo- or hyperglycemia, and there was little fluctuation in BG readings(19) The BG data of the artificial pancreas and arterial BG in this investigation demonstrated a strong connection. The artificial pancreas uses peripheral venous blood rather than arterial blood. The study's findings showed no issues with the clinical application of the artificial pancreas in the intensive care unit.(20) For individuals who are very sick, the target BG control range is often 140–180 mg/dl.(21) BG values greater than 200 mg/dl in COVID-19 patients ought to initiate insulin infusion therapy.(22) Diabetes mellitus is a serious side effect for COVID-19 patients, although it might be challenging to ascertain the patient's history and blood glucose level prior to the illness. In our ICU, many patients with severe COVID-19 pneumonia required intubation. The majority of their family members were quarantined after contracting COVID-19. As a result, it's possible that these patients' precise underlying illnesses are unknown. Many COVID-19 patients have high blood glucose levels and struggle to maintain blood glucose control.(23) Need to focus more on preventing hypoglycemia. As a result, they established the target BG control range for conventional insulin infusion as 180–220 mg/dl. According to a number of standards, COVID-19 patients' permissible BG upper limit was 216 or 220.(24) Patients with severe COVID-19 may have a better prognosis with better BG control with the artificial pancreas. They discovered that improved BG management was likely to be attained even if the artificial pancreas' BG target was set lower.(25) Lessening the workload for ICU staff might be accomplished more successfully by using an artificial pancreas. Critically sick patients experience hyperglycemia as a result of BG controls. It has been proposed that both hyperglycemia and hypoglycemia impair these patients' prognoses, highlighting the significance of BG management.(26) According to this study, critically ill patients' BG control may be enhanced by an artificial pancreas. Patients who are in critical condition may also have a better prognosis if an artificial pancreas is used. Patients who need to be administered glucocorticoids, in particular, frequently struggle to control their blood sugar.(27) This was an observational research conducted at a single centre. The attending physicians made the final decisions about the attachment of the device, the artificial pancreas' parameters, and the pace of insulin infusion. Patients with COVID-19 pneumonia had varying degrees of impaired glucose tolerance, glucocorticoid dose, sugar dose, and other factors. Each patient was treated individually based on the attending physician's assessment. RCTs are required in order to more precisely assess the artificial pancreas's utility in the intensive care unit. Further research is necessary to determine whether employing an artificial pancreas to improve blood glucose control can improve the prognosis of critically ill patients.(9)
CONCLUSION
In patients with severe COVID-19 pneumonia and difficult BG management, the rate of achieving the target BG range was greater with the artificial pancreas compared with usual therapy using intravenous insulin infusion. The insulin delivery speed varied significantly when utilizing the artificial pancreas, indicating that it could be precisely adjusted based on the blood glucose level. The findings of Tokuhira et al. imply that the artificial pancreas aids in BG control in the intensive care unit.
REFERENCES
Swati Dilipsingh Thakur , Kishor V. Otari , Ajay Y. Kale , Improvement Of Blood Glucose Control With Artificial Pancreas In Coronavirus Disease Covid-19 , Int. J. of Pharm. Sci., 2024, Vol 2, Issue 9, 893-905. https://doi.org/10.5281/zenodo.13786284
10.5281/zenodo.13786284
