Description of Sensor Placement for Continuous Glucose Monitoring in a Pig Model of Diabetes Mellitus
Porcine models can potentially bridge the gap between basic studies and clinical trials in humans due to their homologous similarities; the pig model aids in understanding the pathophysiology of diabetes. Continuous glucose monitoring is crucial for precise insulin control in this disease. Although the use of glucose sensors has been described in the literature, details relating to placement of the sensor are not provided. Hence, we describe here, in detail, the insertion, placement, and fixation of the sensor for continuous glucose level monitoring and its use in diabetes mellitus studies in swine. Four male minipigs, weighing 15 to 22 kg and aged 10 mo, were used and underwent induction of diabetes using streptozotocin. The sensors were placed on the 4 pigs to enable continuous glucose monitoring throughout the study (35 d), with sensor changes every 7 d. The results obtained allow us to consistently track changes in glucose. In conclusion, this article highlights the methodology’s effectiveness and reinforces its practical applicability in obtaining reliable results for analyzing trends in interstitial glucose levels, thus further optimizing the swine model for diabetes research.
Introduction
Diabetes is the first noninfectious disease recognized by the U. S. National Institutes of Health as an epidemic of the 21st century.1 Animal models play an essential role in studying diabetes, not only because they increase our understanding of its etiologies but also because they provide a tool to identify and evaluate potential therapeutic strategies. The closer the model’s physiologic behavior is to humans, the better the concordance between preclinical and clinical studies, thereby facilitating the selection of properly validated therapeutic targets and saving time and costs. In this regard, porcine models serve as a valuable bridge between basic studies and clinical trials in human patients.2,3
The pig is a homologous model because it has shown similarities to humans, especially regarding the structure and function of the gastrointestinal tract, pancreas morphology, and overall metabolic state. These similarities make the pig useful for comparing human physiology and pathophysiology.4 However, diabetes mellitus does not commonly develop spontaneously in pigs, so the disease must be experimentally induced. There are 2 methods for this: a surgical method involving total or partial pancreatectomy and a chemical method involving the administration of a cytotoxic drug such as streptozotocin or alloxan which have a cytotoxic effect on pancreatic islet β-cells and act as a glucose analog entering the cell through the glucose transporter 2.5
After reliably inducing diabetes, establishing an optimal regimen for postoperative care is crucial, particularly for extended follow-up periods. In this regard, accurate monitoring for glycemia is critical. Recent advances in technology related to continuous glucose monitoring (CGM) systems have been highly successful in clinical practice and studies of this disease using large animal models, and allowing the animals to avoid repeated capillary punctures for glucose level measurements.6
Advances in technology, such as CGM and continuous subcutaneous insulin infusion, have been highly useful for maintaining metabolic control in patients with type 1 diabetes.7 The Guardian Connect CGM system8 (Medtronic MiniMed; Medtronic, Minneapolis, MN) acquires and stores signals from a subcutaneous sensor connected to a glucose monitor. Through the transmitter (MiniMed) communication station, stored signals can be downloaded to a personal computer and converted to glucose levels. The sensor’s function is based on the oxidase reaction with glucose; the signal is proportional to the glucose concentration in the interstitial fluid and handles a measurement range of 40 to 400 mg/dL. The system records glucose concentrations every 10 s and stores the mean level every 5 min.9 Using a sensor allows for the receipt of up to 288 glucose readings every 24 h, thus completing the information between blood glucose checks.
Sensor glucose and blood glucose.
In contrast to the blood glucose meter which measures glucose levels in the blood, the glucose sensor measures glucose in the fluid surrounding the tissue; that is, the interstitial fluid. Glucose readily moves between these 2 areas (blood and interstitial fluid). Because of how glucose moves, the blood glucose meter and sensor readings will be similar but not the same.8
When glucose levels rise or fall rapidly, a greater difference can be expected between blood glucose meter readings and sensor glucose readings. Therefore, to approximate the interstitial glucose value more closely to the blood glucose value, a capillary sample must be taken, and this value is used to calibrate the sensor (every 12 h). The transmitter transmits the sensor glucose values calculated by the algorithm in real time to a main display device, allowing the monitoring of the sensor glucose values.8
The Guardian Connect system provides real-time glucose values and trends through a Guardian Connect app installed on a compatible electronic mobile device and allows users to detect trends and track patterns in glucose concentrations. The Guardian Connect app alerts if a Guardian Sensor glucose level reaches, falls below, rises above, or is predicted to surpass set values.8
The Guardian Connect transmitter powers the sensor, collects and calculates sensor data, and sends the data via Bluetooth version 4.0 to the Guardian Connect app installed on a compatible mobile device. The app displays sensor glucose data and provides a user interface for sensor calibration, entering data such as exercise and meals, and uploading information to the CareLink personal website. The app can detect trends and track patterns in glucose concentrations.8
System description.
The Guardian Connect system uses the Guardian Sensor, a glucose sensor positioned beneath the skin to monitor glucose levels in the interstitial fluid continuously. The Guardian Connect transmitter gathers these glucose readings and converts them into sensor glucose values. These values are then displayed on the Guardian Connect app, which can also issue alerts based on the detected glucose levels.8
Though the literature describes the use of CGM in minipigs, its placement for optimal functioning has not been precisely described. This study aims to describe in detail the insertion, placement, and fixation of the sensor for continuous glucose level monitoring in diabetes mellitus studies using porcine models.
Materials and Methods
Ethical statement.
The procedures performed were approved by the Institutional Research Committee and the Committee for the Care and Use of Laboratory Animals (IACUC-INP) under registration No. INP 2021/026, which is the main project where this procedure was carried out.
Animals.
Minipigs.
The minipigs used in this study were from a genetic line developed by a local breeder through the crossbreeding of small pigs from different skin pigmentation phenotypes, primarily those with white skin, and white with black or brown spots. Their average birth weight is 250 g, and at 2 y of age, they weigh between 25 and 35 kg. (The minipig vendor was RGS Research Global Solutions). Since the colonies began (2015), all animals were maintained in a conventional purpose-built and ventilated minipig facility accomplishing compliance with NOM 062 ZOO 1999 (https://www.gob.mx/cms/uploads/attachment/file/203498/NOM-062-ZOO-1999_220801.pdf). No other pig farm was within 20 km, and one person dedicated to this facility and who wore protective clothing provided care to the animals.
Animals were provided untreated tap water and were fed with a Purina diet (Finishing Family Farm swine diet; https://www.llabana.com/granja-familiar-finalizador), 200 g twice daily on the floor at 09:00 and 17:00. Health monitoring sampling was done annually on 10 wk old animals, and serum samples were taken for assay of virus and bacterial antibodies from 5 animals. Also, samples for parasitological examination were taken.
Animals were free of pseudorabies virus (PRV), circovirus, porcine epidemic diarrhea virus, PRRS virus, Actinobacillus pleuropneumoniae, Pasteurella spp., Bordetella bronchiseptica, Coccidia (Eimeria, Isospora), and helminths.
Diabetes was induced in four male minipigs (Sus scrofa domesticus), weighing 15 to 22 kg and aged 10 mo, using streptozotocin. They were acclimated for 1 month before the experimental phase in the animal facility, following the institute’s internal policies. The facility conditions comply with the requirements according to the Guide for the Care and Use of Laboratory Animals (National Research Council) in an AAALAC-accredited facility10 and NOM-062-ZOO-1999: pens with slotted concrete floors, nipple drinkers, temperature 22 to 26 °C, and relative humidity of 50% ± 20%. The pigs were housed in individual (1.12 m × 1.82 m) pens. All pigs were provided daily socialization and were in the standard enrichment program. The animals were monitored twice a day.
While on study, the animals were fed a commercial swine diet (Jamonina; Purina Mills) at a consumption rate of 1.5% of body weight per day as per nutritional recommendations, divided into 2 meals per day (09:00 and 17:00). Water consumption was provided ad libitum.
The sensors were placed on the 4 pigs to enable CGM throughout the study (35 d), with sensor changes every 7 d.
For sensor placement and administration of streptozotocin, the animals were anesthetized with an intramuscular mixture of 2.2 mg/kg ketamine (Anesket; Pisa) + 2.2 mg/kg xylazine (Rompun; Bayer) + 4.4 mg/kg tiletamine/zolazepam (Zoletil; Virbac) + 0.05 mg/kg atropine (Atropisa; Pisa).
The induction of diabetes mellitus was performed through chemical pancreatic ablation with streptozotocin at 150 mg/kg using the technique described by King.6
At the end of the study, pigs were euthanized with sodium pentobarbital (150 mg/kg IV) according to the protocol and AVMA guidelines.11
Description of the selection of the insertion site, placement, and fixation of the sensor in the pig for CGM.
The insertion site for the CGM sensor in the pig was selected based on several criteria to ensure accurate measurements and prevent sensor displacement. The region of the biceps femoris slightly distal to the ischial tuberosity, as shown by the red circled area in Figure 1, was preferred for its accessibility and the reduced likelihood of interference from the pig’s movements, whether on the left or right side.
Citation: Journal of the American Association for Laboratory Animal Science 64, 2; 10.30802/AALAS-JAALAS-24-096
Preparation of the insertion site.
The biceps femoris area was carefully cleaned and shaved to ensure a smooth and unobstructed surface for sensor placement, and an antiseptic was subsequently applied to minimize the risk of infection (Figure 2A and B).
Citation: Journal of the American Association for Laboratory Animal Science 64, 2; 10.30802/AALAS-JAALAS-24-096
Preparation of the sensor.
The sensor was first placed into the inserter device (serter) according to the manufacturer’s instructions. This involved ensuring the sensor was properly seated in the serter to facilitate smooth insertion into the skin (Figure 3A to D).
Citation: Journal of the American Association for Laboratory Animal Science 64, 2; 10.30802/AALAS-JAALAS-24-096
Insertion of the sensor.
The sensor filament was carefully inserted into the skin using the inserter device. The insertion angle and depth were adjusted according to the manufacturer’s recommendations to ensure proper sensor function and minimal discomfort to the animal (Figure 4).
Citation: Journal of the American Association for Laboratory Animal Science 64, 2; 10.30802/AALAS-JAALAS-24-096
Placement.
The sensor was positioned in the biceps femoris area, ensuring it was securely placed to minimize movement (Figure 4A and B).
The area was chosen to avoid high-motion regions that could cause dislodging.
Fixation.
The sensor was affixed with medical-grade adhesive (KolaLoca) to ensure it remained in place throughout the monitoring period (Figure 5A and B). This adhesive did not cause any skin irritation.
Citation: Journal of the American Association for Laboratory Animal Science 64, 2; 10.30802/AALAS-JAALAS-24-096
The transmitter was then connected to the sensor and secured with the same adhesive and a protective patch (Figure 6A).
Citation: Journal of the American Association for Laboratory Animal Science 64, 2; 10.30802/AALAS-JAALAS-24-096
A protective patch (Oval Tape; Medtronic, Minneapolis, MN) was applied over the sensor to further secure it and protect it from external factors such as rubbing against surfaces or other pigs (Figure 6B).
Once the sensor was inserted and the transmitter connected, the mobile device and transmitter were able to communicate. Within 2 h, the system was calibrated by entering blood glucose levels (Figure 6C).
This careful selection, placement, and fixation procedure ensured that the sensor remained securely in place, providing accurate and continuous glucose readings for the duration of the study.
Blood glucose measurement for equipment calibration.
A lancet was used to prick the auricular vein to obtain a drop of blood, which was then placed on a FreeStyle glucose test strip (Abbott Laboratories, Abbott Park, IL) and inserted into a FreeStyle glucometer (Abbott Laboratories, Abbott Park, IL) to measure the blood glucose.
Statistical analysis.
For this study, baseline glucose measurements were taken before the administration of streptozotocin, and the final measurements were taken 30 d postadministration of streptozotocin.
Glucose data were collected from the CareLink software for the 4 pigs to represent the changes in glucose levels throughout the day, both at baseline and at the end of the study. The mean glucose levels were estimated in 2-h intervals (0 to 2 h, 2 to 4 h, etc.) throughout the day. All statistical analyses were conducted using STATA software (version 17).
Results
For the 4 pigs evaluated, we obtained 14 d of glucose measurements before the streptozotocin administration (baseline data). Following streptozotocin administration, 9 d of measurements were collected as the final data; and the mean and SE glucose levels were calculated every 2 h for both time periods (baseline and final). The results can be seen in the graph in Figure 7.
Citation: Journal of the American Association for Laboratory Animal Science 64, 2; 10.30802/AALAS-JAALAS-24-096
Discussion
Accurately placing a CGM device in pigs allows a more precise evaluation of glucose level behavior in a diabetes model, enabling the administration of insulin when needed and monitoring episodes of hyperglycemia or hypoglycemia. An accurate evaluation of glucose levels is crucial for preclinical diabetes studies, providing reliable and repeatable results. Using CGM devices in animal models of diabetes is an excellent option, as it reduces the stress associated with continuous blood sampling.12 For example, in this study, the mean glucose levels every 2 h illustrate the behavior of glucose levels before the administration of streptozotocin (baseline data) and through 30 d after its administration (final data).
We chose the CGM device evaluated here because it had been previoulsy evaluated in pigs; however, its insertion, placement, and fixation were not precisely described by Strauss and colleagues.9
The complications observed with the sensors were that, despite proper placement and fixation, the animals could still dislodge them by rubbing against surfaces, requiring the insertion of a new sensor. Regarding streptozotocin administration, the most critical period was the hypoglycemia phase postadministration, which could be detected using the sensor. This justifies the use of the CGM system in the porcine diabetes model. The Medtronic CGM used in humans is believed to be appropriate for these studies, because pigs have skin characteristics similar to humans, including extracellular matrix and vascularization.
Although animal models have been crucial in advancing diabetes research, it is important to acknowledge that none can completely replicate the conditions of human type 1 and type 2 diabetes, and each has its limitations that require careful consideration. This highlights the importance of ongoing research to enhance these animal models. In addition, ethical considerations are paramount in animal research. Researchers must follow stringent guidelines to ensure ethical treatment of animals, minimize discomfort, and use alternative methods whenever possible to reduce the use of animals in research.13
Unlike other studies that used different devices,14 this device allowed us to insert the filament with the inserter into the pig’s skin. Placing the device in the biceps femoris area prevented the animal from removing it. Fixing the sensor with cyanoacrylate adhesive provided better stabilization, making it easy to attach the transmitter and secure it to the skin with a patch. This ensured the correct transmission of interstitial glucose values to the pump, and results were subsequently obtained through the software for trend analysis of interstitial glucose levels.
Limitations of the study.
One of the limitations of this study was that pigs have a mean baseline interstitial glucose level of 40 mg/dL. Since the sensor’s range is 40 to 400 mg/dL, CGM at 40 mg/dL triggered hypoglycemia alerts. Due to this, it was necessary to perform more frequent calibrations and, at times, more automated requests to replace the sensor, as it was mistakenly detected as damaged. However, this issue did not occur when the pigs were diabetic and had glucose levels above 140 mg/dL.
Conclusion.
The methodology employed in this study was effective and practically applicable for obtaining reliable results for trend analysis of interstitial glucose levels. This reinforces the significant contribution of the study to optimizing diabetes research using the pig as an animal model.
(A) Entire area where the sensor can be placed (biceps femoris), slightly distal to the ischial tuberosity, as shown by the red circled area. (B) Mobile device connecting to the transmitter to receive and sensor data. (C and D) The sensor can be positioned on either the left or right side.
(A) Biceps femoris area (red arrow), where the sensor will be placed. (B) Cleanliness of the area.
(A) Glucose sensor assembly ([1] pedestal, [2] needle housing, and [3] sensor base). (B, C, and D) Placement of sensor into the one-press serter.
(A) Placement of the sensor in the biceps femoris area. (B) Needle housing withdrawal.
(A) Sensor secured with its adhesive and protective patch. (B) Sensor fixation.
(A) Correct connection to the transmitter because it emits a green light. (B) Sensor fixation with a patch. (C) Sensor ready to start data acquisition.
Represents the 2-h interval mean (SE) glucose levels before and 30 d after streptozotocin administration. The baseline data are green, and the final data are pink.
Contributor Notes