Therapy Adjustments Based on Trend Arrows Using Continuous Glucose Monitoring Systems

Abstract

Continuous glucose monitoring (CGM) systems use trend arrows to accurately display the anticipated glucose curve for the user. These are used for both “real-time” glucose monitoring and for intermittent scanning glucose monitoring. Trend arrow data are used by people with diabetes to make corrections to their glucose control. It is essential that they are correctly interpreted when adjusting insulin doses and to ensure that appropriate treatment decisions are made. The aim of this article is to provide general treatment guidance for diabetes teams and for people with diabetes using CGM in the context of trend arrows. This is based on previous recommendations for interpreting trend arrows without losing sight of the need for individual therapy adjustment.

Keywords

diabetes therapy diabetes self-management continuous glucose monitoring CGM rtCGM iCGM trend arrows

The use of continuous glucose monitoring (CGM) systems has greatly increased in people with diabetes, and they have become one of the key components of effective diabetes self-management.^1-5 CGM systems can be based on real-time CGM (rtCGM) or on CGM with intermittent scanning (iCGM) both of which measure glucose levels in the interstitial fluid (ISF). CGM systems also provide trend arrow information on the direction and rate of change (ROC) of glucose. These trend arrows help users to anticipate short-term future glucose levels. With rtCGM systems, trend arrow data can provide an instant insight into fluctuating glucose levels using an alarm function. For iCGM systems the sensor data are available to the user upon active scanning with a receiver. On average, glucose levels are scanned 9-15 times a day by a patient using iCGM.^6-8 It has been observed that an increased scanning frequency improves the awareness of the user for glucose self-management.⁷

Depending on the system, CGM provides between 288 (every 5 minutes) and 1440 glucose measurements (every minute) per day.⁹ Based on the ROC of the glucose readings stored over the last 15 minutes, the trend arrow displayed on the reader will indicate whether glucose levels are rising, falling or are stable.^10,11 This graphic information in the context of the current glucose level enables people with diabetes to take action to adjust their glucose levels in a timely manner, for example by taking insulin or intake of carbohydrates. It is important to note that trend arrows are based on retrospective data in accordance with the specifications of the Clinical Laboratory Standard Institution in the Performance Metrics for Continuous Interstitial Glucose Monitoring. Consequently, this can lead to a situation in which the glucose levels that are currently being measured are rising, while the retrospective trend arrow is pointing downward.¹² In such cases of doubt, CGM users should always make their treatment decision based on the current CGM data rather than the trend arrows.

The aim of this article and the recommendations it contains is to provide an overview of the methodology and manufacturer-dependent definitions of trend arrows. The article also aims to establish the relevance of trend arrows for immediate treatment decisions made by patients, and to outline the associated challenges. By reviewing recent publications on this topic, we also aim to provide on this basis recommendations for interpreting trend arrows and making subsequent changes to treatment that are generally applicable and practical but also individually adjustable and in the context of the separate systems available.^13-16

Trend Arrow Information and Its Evaluation

Comparison of Trend Arrow Information

Table 1 describes the characteristics of trend arrows that are displayed by CGM systems currently available in Germany (as of October 2018).

Table 1.

Comparison of the Rates of Change That Are Illustrated by the Trend Arrows of Different CGM Providers.

Abbott FreeStyle Libre¹⁷	Dexcom G4/G5/G6® Mobile¹⁸	Medtronic 640G¹⁹	Medtronic Veo	Roche Senseonics Eversense®²⁰
NA	NA	>3 mg/dl 0.2 mmol/l per minute	NA	NA
NA	>3 mg/dl >0.2 mmol/l per minute	2-3 mg/dl 0.1-0.2 mmol/l per minute	>2 mg/dl >0.1 mmol/l per minute	NA
>2 mg/dl >0.1 mmol/l per minute	2-3 mg/dl 0.1-0.2 mmol/l per minute	1-2 mg/dl 0.05-0.1 mmol/l per minute	1-2 mg/dl 0.05-0.1 mmol/l per minute	>2 mg/dl >0.1 mmol/l per minute
1-2 mg/dl 0.05-0.1 mmol/l per minute	1-2 mg/dl 0.05-0.1 mmol/l per minute	NA	NA	1-2 mg/dl 0.05-0.1 mmol/l per minute
<1 mg/dl <0.05 mmol/l per minute	<1 ml/dl <0.05 mmol/l per minute	NA	NA	0-1 mg/dl 0-0.05 mmol/l per minute
1-2 mg/dl 0.05-0.1 mmol/l per minute	1-2 mg/dl 0.05-0.1 mmol/l per minute	NA	NA	1-2 mg/dl 0.05-0.1 mmol/l per minute
>2 mg/dl >0.1 mmol/l per minute	2-3 mg/dl 0.1-0.2 mmol/l per minute	1-2 mg/dl 0.05-0.1 mmol/l per minute	1-2 mg/dl 0.05-0.1 mmol/l per minute	>2 mg/dl >0.1 mmol/l per minute
NA	>3 mg/dl >0.2 mmol/l per minute	2-3 mg/dl 0.1-0.2 mmol/l per minute	>2 mg/dl >0.1 mmol/l per minute	NA
NA	NA	>3 mg/dl >0.2 mmol/l per minute	NA	NA

The summary in Table 1 details the different underlying rates of change (ROC) in glucose levels indicated by the trend arrow displays on different CGM devices: the ROC is calculated using the glucose levels per minute over the previous 15 minutes and are stated in mg/dL or mmol/L per minute.¹¹ For example, ROC of 2 mg/dL (0.1 mmol/L) per minute means that under otherwise steady state conditions, the glucose level will change by about 60 mg/dL (3.3 mmol/L) in the next 30 minutes.

An ROC <1 mg/dL (0.05 mmol/L) per minute is considered stable and commonly represented by a horizontal trend arrow. An upward or downward trend greater than this is indicated by different CGM systems and quantified visually using different trend arrow graphics.

To be able to better evaluate the clinical relevance, a calculation of the change over a 30-minute period is useful (Table 2). This allows a glucose level to be estimated such that any necessary treatment decisions can be taken in this 30-minute window. It is important to note that the measured ISF glucose levels do not correspond to blood glucose levels at any time point but are associated with a time lag, particularly at times of rapidly changing glucose.²¹ To date, there is no evidence-based recommendation for how these differences can be adjusted or factored into safe treatment decisions.

Table 2.

Anticipated Glucose Change in 30 Minutes.

Abbott FreeStyle Libre¹⁷	Dexcom G4/G5/G6® Mobile¹⁸	Medtronic 640G¹⁹	Medtronic Veo	Roche Senseonics Eversense®²⁰
NA	NA	>90 mg/dl >5.0 mmol/l	NA	NA
NA	>90 mg/dl >5.0 mmol/l	60-90 mg/dl 3.3-5.0 mmol/l	>60 mg/dl >3.3 mmol/l	NA
>60 mg/dl >3.3 mmol/l	60-90 mg/dl 3.3-5.0 mmol/l	30-60 mg/dl 1.7-3.3 mmol/l	30-60 mg/dl 1.7-3.3 mmol/l	>60 mg/dl >3.3 mmol/l
30-60 mg/dl 1.7-3.3 mmol/l	30-60 mg/dl 1.7-3.3 mmol/l	NA	NA	30-60 mg/dl 1.7-3.3 mmol/l
<30 mg/dl <1.7 mmol/l	<30 mg/dl <1.7 mmol/l	NA	NA	<30 mg/dl <1.7 mmol/l
30-60 mg/dl 1.7-3.3 mmol/l	30-60 mg/dl 1.7-3.3 mmol/l	NA	NA	30-60 mg/dl 1.7-3.3 mmol/l
>60 mg/dl >3.3 mmol/l	60-90 mg/dl 3.3-5.0 mmol/l	30-60 mg/dl 1.7-3.3 mmol/l	30-60 mg/dl 1.7-3.3 mmol/l	>60 mg/dl >3.3 mmol/l
NA	>90 mg/dl >5.0 mmol/l	60-90 mg/dl 3.3-5.0 mmol/l	>60 mg/dl >3.3 mmol/l	NA
NA	NA	>90 mg/dl >5.0 mmol/l	NA	NA

Calculated from the retrospective data over the last 15 minutes, the trend arrows in the different systems indicate approximately how ISF glucose levels will rise or fall in the next 30 minutes if the trend remains steady.

Challenges Faced When Assessing CGM Information and Trend Arrows

The ROC classification of different trend arrows and what that means for anticipated glucose levels varies depending on the manufacturer (Tables 1 and 2). Although the CGM systems listed here can have visually comparable graphic displays to illustrate changes in glucose levels to the user, an underlying uniform pattern is missing.¹⁵

This lack of standardization can be problematic in routine clinical use. Thus, there is a need for diabetes health care professionals and people with diabetes who use CGM to be familiar with the specific principles underlying the trend arrow displays for the particular system being used. Also, it must always be noted that the arrows represent trends and do not provide definitive numerical values.²²

A study comparing the trend arrow data obtained from the ISF with continuously measured blood glucose values has revealed important discrepancies.²³ The predictive trend arrow indications matched the values from the blood glucose measurements in 61.1% (Dexcom G5) and 58.5% (FreeStyle Libre) of cases, respectively. As previously stated, the trend arrow information and the real-time glucose curve can differ,¹² however this does not mean that the system provides incorrect calculations for the trend arrows. In addition, one should be aware of the fact that different situations can influence the accuracy of CGM sensors, as compression or hot bath.

Therefore, assessing the real-time CGM data and trend arrow information as the basis for both short-term and medium-term adjustments in treatment presents a challenge, both for diabetes teams and for CGM users. Out of the large volume of data available, reliable, easy-to-use treatment guidelines for interpreting these values and trends must be generated for routine clinical use.²⁴ This can be achieved with the recommendations in this article.

Previous Recommendations for Adjusting Doses Based on Trend Arrow Categories

For the reasons indicated above, general recommendations for treatment decisions should take into account both the trend arrows and the different ROC in glucose that they reflect. Ultimately, these recommendations must be individually adjusted to accommodate personal distinctions and different CGM devices. To date, there have been five recommendations published for how to respond to trend arrow information (Table 3).

Table 3.

Previously Published Recommendations on Responding to Trend Arrow Information.

Insulin adjustment	JDRF/DirecNet^25,26	Scheiner²⁶	Pettus/Edelman¹³	Klonoff/Kerr¹⁵	Aleppo/Laffel Adults¹¹		Laffel/Aleppo Children¹⁴
Increase	Increase by 20% of the total dose	Increase current sensor value by 50 mg/dL (2.8 mmol/L)	Increase current sensor value by 100 mg/dL (5.5 mmol/L)	Increase current insulin dose by 2 units	<25	+4.5 IU	<25	+4.0 IU
					25-<50	+3.5	25-<50	+3.0
					50-<75	+2.5	50-<75	+2.0
					≥75	+1.5	75-<12	+1.0
							≥125	+0.5
Increase	Increase by 20% of the total dose	Increase current sensor value by 25 mg/dL (1.4 mmol/L)	Increase current sensor value by 75 mg/dL (4.2 mmol/L)	Increase current insulin dose by 1.5 units	<25	+3.5	<25	+3.0
					25-<50	+2.5	25-<50	+2.0
					50-<75	+1.5	50-<75	+1.0
					≥75	+1.0	75-<125	+0.5
							≥125	No adjustment
Increase	Increase by 10% of the total dose	No adjustment	Increase current sensor value by 50 mg/dL (2.8 mmol/L)	Increase current insulin dose by 1 unit	<25	+2.5	<25	+2.0
					20-<50	+1.5	25-<50	+1.0
					50-<75	+1.0	50-<75	+0.5
					≥75	+0.5	75-<12	No adjustment
							≥125	No adjustment
No adjustment	0%			No adjustment	No adjustment		No adjustment
Reduce	Reduce by 10% of the total dose	No adjustment	Reduce current sensor value by 50 mg/dL (2.8 mmol/L)	Reduce current insulin dose by 1 unit	<25	−2.5	<25	−2.0
					25-<50	−1.5	25-<50	−1.0
					50-<75	−1.0	50-<75	−0.5
					≥75	−0.5	75-<12	No adjustment
							≥125	No adjustment
Reduce	Reduce by 20% of the total dose	Reduce current sensor value by 25 mg/dL (1.4 mmol/L)	Reduce current sensor value by 75 mg/dL (4.2 mmol/L)	Reduce current insulin dose by 1.5 units	<25	−3.5	<25	−3.0
					25-<50	−2.5	25-<50	−2.0
					50-<75	−1.5	50-<75	−1.0
					≥75	−1.0	75-<12	−0.5
							≥125	No adjustment
Reduce	Reduce by 20% of the total dose	Reduce current sensor value by 50 mg/dL (2.8 mmol/L)	Reduce current sensor value by 100 mg/dL (5.5 mmol/L)	Lower current insulin dose by 2 units	<25	−4.5	<25	−4.0
					25-<50	−3.5	25-<50	−3.0
					50-<75	−2.5	50-<75	−2.0
					≥75	−1.5	75-<12	−1.0
							≥125	−0.5

The recommendation from the Juvenile Diabetes Research Foundation (JDRF) and Diabetes Research in Children Network Study Group (DirecNet) includes a percentage adjustment of insulin doses in accordance with the trend arrows, up to a maximum of 20%.

Studies of different patient groups (type 1 diabetes [T1D], type 2 diabetes [T2D], intensive insulin therapy [MDI], or continuous subcutaneous insulin infusion [CSII]) show that patients guided by their own experience often make very significant insulin dose adjustments based on the visible trend arrows that go well beyond these recommendations. These studies, which are based on self-reported survey responses by people with diabetes, reveal that an upward trend arrow might prompt an insulin dose increase of more than 100%, whereas a downward trend arrow might result in an insulin dose reduction of only up to 50%.^25,26

Consequently, recommendations by Pettus/Edelman and Scheiner were developed in which a defined glucose concentration value was added or subtracted to adjust the current glucose level, depending on the trend arrow displayed. The individual correction rule is then applied to the calculated value (Table 3). This requires mathematical skills and some level of numeracy on the part of the patients and can create additional burden.^13,27 Calculating the right dose of insulin for boluses has been shown to be complex, burdensome and sometimes difficult for people with diabetes in studies regarding the use of bolus calculators.^28-30

Klonoff/Kerr developed a simpler model for using the ROC information represented by trend arrows: it assumed that glucose levels will rise or fall depending on the trend information by 67 mg/dL (3.7 mmol/L), 112 mg/dL (6.2 mmol/L), or at least 135 mg/dL (7.5 mmol/L) within the next 45 minutes. An easy to apply addition or subtraction formula for correcting the insulin bolus is then indicated by the model: either 1, 1.5, or 2 insulin units depending on the trend arrow information.¹⁴ This simplifies the calculation of the insulin dose but may only apply to certain groups of patients as it does not take into account patient-specific factors such as their individual insulin regimen, their insulin sensitivity or other clinical features.

Laffel/Aleppo also suggest an adjustment based on insulin units, but also determined by different levels of insulin sensitivity, separated into children and adults. This can be more easily adjusted to the requirements of the patients’ individual insulin therapy.^11,14

Suggestion for Further Simplification of Trend Arrow Information

The authors are a German expert group and have developed an interpretation matrix for trend arrows with associated treatment-adjustment recommendations that are presented in the following tables (Tables 4 -6). Based on the trend arrow ROC definitions of the individual CGM systems, recommendations for action for the different trend arrow displays were developed, based on five different baseline glucose levels (in accordance with international recommendations)⁸ and on individual insulin-sensitivity (correction factor). This is important to ensure that valuable information from the trend arrows can be used by as many CGM users as possible, not only those who are most technically oriented. This can therefore help users to see even better the benefit from the core features of CGM systems compared to SMBG, dynamic versus static presentation.

Table 4.

Scorecard for Type 1 Diabetes.

			Glucose level
ScorecardType 1 diabetes			<70 mg/dL^a <3.9 mmol/L	70-180 mg/dL3.9-10 mmol/L				180-250 mg/dL10-13.9 mmol/L				>250 mg/dL>13.9 mmol/L				Ketone positive^b
Change				Correction factormg/dLmmol/L				Correction factormg/dLmmol/L				Correction factormg/dLmmol/L
Abbott/Roche	Dexcom	Medtronic		<25<1.4	25-<50 1.4-<2.8	50-<75 2.8-<4.2	>75>4.2	<25<1.4	25-<50 1.4-<2.8	50-<75 2.8-<4.2	>75>4.2	<25<1.4	25-<50 1.4-<2.8	50-<75 2.8-<4.2	>75>4.2
Abbott/Roche	Dexcom	Medtronic		Insulin units				Insulin units				Insulin units
			CU	+3.5	+2.5	+1.5	+1	+4.5	+3.5	+2.5	+1.5	+5	+4	+3	+2	Max corr^b
				+2.5	+2	+1	+0.5	+3.5	+2.5	+1.5	+1	+4	+3	+2	+1.5	Max corr^b
				+1.5	+1	+0.5	+0.5	+2 .5	+1.5	+1	+0.5	+3	+2	+1.5	+1	Max corr^b
			1 fast-acting CU	+0	+0	+0	+0	+0	+0	+0	+0	+0	+0	+0	+0	Max corr^b
			2 fast-acting CUs	−2.5	−1.5	−1	−0.5	−2	−1	−0.5	−0.5	−2.5	−1	−0.5	−0	Max corr^b
			2 fast-acting CUs	−3.5	−2.5	−1.5	−1	−3	−2	−1	−1	−3.5	−2	−1	−0.5	Max corr^b
			2 fast-acting CUs	−4.5	−3.5	−2.5	−1.5	−4	−3	−1.5	−1	−4	−2.5	−1	−0.5	Max corr^b

Status Oct. 2018. A further correction is recommended no earlier than 2 hours after the last insulin dose. These values are intended as a guide only and need to be adjusted to the insulin sensitivity and individual values of the patient.

With a baseline glucose value <50 mg/dL (<2.8 mmol/L), three fast-acting carbohydrate units are recommended.

If ketones are positive, the individually suggested maximal correction should be applied independent of the trend arrow information.

Table 5.

Scorecard for Insulin-Dependent Type 2 Diabetes.

			Glucose level
ScorecardInsulin-dependent type 2 diabetes			<70 mg/dL^a <3.9 mmol/L	70-180 mg/dL3.9-10 mmol/L			180-250 mg/dL10-13.9 mmol/L			>250 mg/dL>13.9 mmol/L
Change				Correction factormg/dLmmol/L			Correction factormg/dLmmol/L			Correction factormg/dLmmol/L
Abbott/Roche	Dexcom	Medtronic		≤10≤0.6	>10-<30>0.6-<1.7	>30->50>1.7-<2.8	≤10≤0.6	>10-<30>0.6-<1.7	>30->50>1.7-<2.8	≤10≤0.6	>10-<30>0.6-<1.7	>30->50>1.7-<2.8
Abbott/Roche	Dexcom	Medtronic		Insulin units			Insulin units			Insulin units
			CU	+5	+4	+3	+6	+5	+4	+7	+6	+5
				+4	+3	+2	+5	+4	+3	+6	+5	+4
				+3	+2	+1	+4	+3	+2	+5	+4	+3
			1 fast-acting CU	+0	+0	+0	+0	+0	+0	+0	+0	+0
			2 fast-acting CUs	−3	−2	−1	−3	−2	−1	−3	−2	−1
			2 fast-acting CUs	−4	−3	−2	−4	−3	−2	−4	−3	−2
			2 fast-acting CUs	−5	−4	−3	−5	−4	−3	−5	−4	−3

Status Oct 2018. A further correction is recommended no earlier than 2 hours after the last insulin dose. These values are intended as a guide only and need to be adjusted to the insulin sensitivity and individual values of the patient.

With a baseline glucose value <50 mg/dL (<2.8 mmol/L), three fast-acting carbohydrate units are recommended.

Table 6.

Scorecard for Children/adolescents With Type 1 Diabetes.

			Glucose level
ScorecardChildren/adolescents with type 1 diabetes			<70 mg/dL^a <3.9 mmol/L	70-180 mg/dL3.9-10 mmol/L					180-250 mg/dL10-13.9 mmol/L					>250 mg/dL>13.9 mmol/L					ketone-positive^b
Change				Correction factormg/dLmmol/L					Correction factormg/dLmmol/L					Correction factormg/dLmmol/L
Abbott/Roche	Dexcom	Medtronic		25-<50 1.4-<2.8	50-<75 2.8-<4.2	75-<125 4.2-<6.9	125-<2006.9-<11.1	≥200 ≥11.1	25-<50 1.4-<2.8	50-<75 2.8-<4.2	75-<125 4.2-<6.9	125-<200 6.9-<11.1	≥200 ≥11.1	25-<50 1.4-<2.8	50-<75 2.8-<4.2	75-<125 4.2-<6.9	125-<2006.9-<11.1	≥200 ≥11.1
Abbott/Roche	Dexcom	Medtronic		Insulin units					Insulin units					Insulin units
			CU	+2	+1	+0.5	+0.5	+0.25	+3	+2	+1	+0.5	+0.3	+4	+3	+2	+1.5	+0.75	Max corr^b
				+1	+0.5	+0	+0	+0	+2	+1	+0.5	+0	+0	+3	+2	+1.5	+1	+0.5	Max corr^b
				+0.5	+0	+0	+0	+0	+1	+0.5	+0	+0	+0	+2	+1.5	+1	+0.5	+0.25	Max corr^b
			1 fast-acting CU	+0	+0	+0	+0	+0	+0	+0	+0	+0	+0	+0	+0	+0	+0	0	Max corr^b
			2 fast-acting CUs	−1	−0.5	−0.5	−0.5	−0.5	−1	−0.5	−0.5	−0.3	−0.2	−1	−0.5	−0.5	−0.5	−0.3	Max corr^b
			2 fast-acting CUs	−2	−1	−1	−1	−0.75	−2	−1	−0.7	−0.5	−0.3	−2	−1	−0.5	−0.5	−0.5	Max corr^b
			2 fast-acting CUs	−3	−2	−1.5	−1.5	−1	−3	−2	−1	−0.7	−0.5	−3	−1.5	−1	−0.5	−0.5	Max corr^b

With a baseline glucose value <50 mg/dL (<2.8 mmol/L), three fast-acting carbohydrate units are recommended.

If ketones are positive, the individually suggested maximal correction should be applied independent of the trend arrow information.

The authors have established this interpretation matrix based on the expert consensus by Laffel/Aleppo, as they recommend specific adjustments of insulin dosing with absolute values. Those are grouped according to insulin sensitivity, but independent of the glucose level.^11,14 In contrast, the present recommendations include differentiated adjustments with respect to different baseline glucose levels <70 mg/dl (<3.9 mmol/l), 70-180 mg/dl (3.9-10 mmol/l), 180-250 mg/dl (10-13.9 mmol/l), and >250 mg/dl (>13.9 mmol/l). Depending on the actual glucose level, recommendations for adjusting the insulin dose in response to a specific glucose trend arrow display will differ according to the glucose level. Consequently, the specific recommendation might differ depending on whether glucose levels are elevated or are within target range.

The interpretation matrix is presented in a color-coded scorecard format for three patient groups: T1D (Table 4), insulin-dependent T2D (Table 5), and children/adolescents with T1D (Table 6). Each scorecard includes the following elements:

Trend arrow symbols as displayed by different CGM systems.

Baseline glucose level: five ranges from <70 mg/dL (<3.9 mmol/L) to >240 mg/dL (>13.3 mmol/L) and ketone-positive.

Individual correction factors, based on: four ranges for T1D (<25 to >75 mg/dl (<1.4 to >4.2 mmol/l)); four ranges for T2D (≤10 to <50 mg/dl (≤0.6 to <2.8 mmol/l)); five ranges for children/adolescents with T1D (25-<50 to ≥200 mg/dl (1.4-2.8 to ≥11.1 mmol/l))

Recommendations for treatment adjustments: either an increase or a decrease of insulin units

Recommendations for the intake of an individual carbohydrate unit (CU); a fast-acting CU; two fast-acting CU. CUs are based on the international benchmark for carbohydrate exchange units (CEU) but are not identical, however. CU, fast-acting CU and two fast-acting CU are defined as the quantity of carbohydrates that a patient needs, based on their own experience, to prevent hypoglycemia. For example, if one requires 10 g fast-acting CU at a glucose level between 50 and 70 mg/dl (2.8 and 3.9 mmol/l), it means that 2 fast-acting CUs represent 20g. If 5 g fast-acting CU are required between 50 and 70 mg/dl (2.8 and 3.9 mmol/l), 2 fast-acting CUs are defined as 10 g.

Illustrative examples:

The baseline glucose level of a T1D (Table 4) user is 100 mg/dL (5.6 mmol/L). The CGM device (Dexcom G4/G5/G6) shows two vertical upward trend arrows, meaning that the anticipated rise in the glucose level is <3 mg/dL (0.2 mmol/L) per minute. The correction factor set for the patient is 40 mg/dl (2.2 mmol/l-column 25-<50). The treatment recommendation indicated by the scorecard: injection of 2.5 insulin units.

The baseline glucose level for an insulin-dependent T2D user (Table 5) is 100 mg/dL (5.6 mmol/L) and is thus in the range of 70-180 mg/dL (3.9-10 mmol/L). The CGM device (Abbott FreeStyle Libre) shows a vertical downward trend arrow, meaning that the anticipated fall in the glucose level is >2 mg/dL (>0.1 mmol/L) per minute. The correction factor set for the patient is 20 mg/dl (1.1 mmol/l-column >10-<30 mg/dl (>0.6-<1.7 mmol/l). The treatment recommendation indicated by the scorecard: reduction in the insulin dose by 3 insulin units.

Regardless of the trend arrow information displayed, it must be understood that another correction should not be performed within 90 minutes and if possible not within two hours following the previous insulin injection. Any further corrections to glucose values should take into account the individual glucose variability that each user experiences and their tendency toward hypoglycemia, as well as their frequency of glucose monitoring after a correction has been performed.

What Else Also Has to Be Noted?

In case of a strong falling or rising trend, the expected glucose value after 30 minutes may be considerably lower or higher than the glucose value which has been used as a basis for insulin calculation. Accordingly, it would make sense to incorporate this anticipated trend into the calculated mealtime insulin dose or a correction dose in addition to the observed current glucose value. This dynamic adjustment of insulin doses for mealtimes and for correction doses approximately 2 hours postprandial are applicable for patients on CSII and on MDI. CSII users have different options for modifying the amount of insulin they need according to the trend arrows (temporary basal rate and different types of bolus). If the CSII system has a predictive low glucose suspend feature, patients can use the proposed scorecard with a rising trend, whereas they should primarily rely on the automatic pump suspend to avoid hypoglycemia with a falling trend.

When adjusting the treatment, there are a number of other factors that must always be taken into account in addition to the rate of glucose change, including: individual variability in insulin sensitivity, meal composition (carbohydrate, protein, and fat) and portion sizes, possible physical activity, medications, stress levels, and other preexisting medical conditions must also be considered. All these factors can affect glucose levels and thus complicate the interpretation of trend arrow information.¹³

If ketones have been confirmed, the proposed maximal correction factor should be applied, independent of the displayed trend arrow information.

Correct interpretation of the CGM information displayed and an appropriate response to this interpretation by the user are the foundation for making the right treatment decisions. Structured training is an essential prerequisite to ensuring that patients get the most benefit from their CGM systems and in Germany this is mostly a requirement before starting on CGM. In order to make sure that training is available in a standardized and quality-assured format, there are now a number of different training programs available (eg, SPECTRUM, Flash).^31,32 In addition to such a training, the recommendations made here are intended to enable CGM users to interpret CGM data and trend arrows in a way that is relevant for their own treatment. Previous recommendations for treatment adjustment that are guided by trend arrows do not always seem to be practical or they require a certain level of numeracy. The solutions offered by these prior recommendations do not take into account either the current baseline glucose levels or the individual’s insulin sensitivity. In addition, applying previously proposed recommendations involved making calculations that are too complex for many patients, or the increments between the suggested treatment options are too broad.

With the treatment adjustments based on trend arrows that we propose here, we think it will be possible to make more rapid, easier, safe, and practical use of the information provided by CGM, which also takes into account the individual requirements of insulin therapy at different glucose levels for people with diabetes. This might better enable CGM users to avoid short-term instability in their glucose levels and helps them to better manage their long-term glucose control. It should be noted that trend arrows displayed in non–medically approved applications are out of scope for this discussion.

Conclusion

Users of CGM devices receive continuous feedback about their current glucose levels, alongside retrospective information provided by trend arrows that helps them to assess both the direction and the rate of change of glucose at any given time. This gives them the opportunity to closely monitor changes in their glucose control and to proactively influence them. However, across different CGM systems there is a lack of a standardized protocol for interpreting their current glucose readings in the context of the information provided by the trend arrows. As well as harmonizing and standardizing CGM systems, an important goal is also to provide all CGM users with an easily implementable method for interpreting trend arrows as genuine tools for making treatment adjustment decisions. By presenting our own recommendations for action in a clear, color-coded, tabular format, we aim to provide CGM users with a personalized and simplified system of scorecards for daily use. Both the framework for the contents and the practical application of this matrix for interpreting trend arrows should, however, be confirmed in scientific studies, as is planned in an investigator-initiated study.

Footnotes

Acknowledgements

The authors thank Dr. Rob Brines, Bite Medical Consulting for supporting the preparation of the article.

Abbreviations

AGDT, Arbeitsgemeinschaft Diabetes und Technik der Deutschen Diabetes Gesellschaft [Diabetes and Technology Association of the German Diabetes Society]; CEU, carbohydrate exchange unit; CGM, continuous glucose monitoring; CLSI, Clinical Laboratory and Standards Institute; CSII, continuous subcutaneous insulin infusion; CU, individual carbohydrate unit; DirecNet, Diabetes Research in Children Network Study Group; iCGM, continuous glucose monitoring with intermittent scanning; JDRF, Juvenile Diabetes Research Foundation; MDI, intensive insulin therapy (multiple daily injections); NA, not applicable; POCT-05A, Performance Metrics for Continuous Interstitial Glucose Monitoring; ROC, rate of change; rtCGM, real-time glucose monitoring; SMBG, self-monitoring of blood glucose; T1D, type 1 diabetes; T2D, type 2 diabetes.

Declaration of Conflicting Interests

The author(s) declared the following potential conflicts of interest with respect to the research, authorship, and/or publication of this article: The authors are members of the national advisory board of Abbott Diabetes Care.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: The writing of the article was supported by Abbott Diabetes Care

References

IDF Clinical Guidelines Task Force. Global guideline for type 2 diabetes: recommendations for standard, comprehensive, and minimal care. Diabet Med. 2006;23:579-593.

Ceriello

Colagiuri

International Diabetes Federation guideline for management of postmeal glucose: a review of recommendations. Diabet Med. 2008;25:1151-1156.

American Diabetes Association. Standards of medical care in diabetes 2016. Diabetes Care. 2016;39(suppl 1):S1-S112.

Grunberger

Bailey

Camacho

et al . Proceedings from the American Association of Clinical Endocrinologists and American College of Endocrinology consensus conference on glucose monitoring. Endocr Pract. 2015;21:522-533.

National Collaborating Centre for Women’s and Children’s Health. National Institute for Health and Care Excellence: Clinical Guidelines. In Diabetes (Type 1 and Type 2) in Children and Young People: Diagnosis and Management. London, UK: National Institute for Health and Care Excellence; 2015.

Heinemann

Freckmann

CGM versus FGM; or, continuous glucose monitoring is not flash glucose monitoring. J Diabetes Sci Technol. 2015;9:947-950.

Bolinder

Antuna

Geelhoed-Duijvestijn

et al . Novel glucose-sensing technology and hypoglycaemia in type 1 diabetes: a multicentre, non-masked, randomised controlled trial. Lancet. 2016;388:2254-2263.

Danne

Nimri

Battelino

et al . International consensus on use of continuous glucose monitoring. Diabetes Care. 2017;40:1631-1640.

DDG AdTeVd. Kontinuierliche Glukosemessung (CGM) in der Gewebeflüssigkeit. Wissenschaftliche Bewertung von CGM und medizinische Beurteilung des Nutzens für die Diabetestherapie AGDT-DDG-GKV_CGM-Positionspapier 2010. Available at: https://de.scribd.com/document/252338406/DDG-Kontinuierliche-Glukosemessung.

10.

Horwitz

Klonoff

DC.

New Technologies for Glucose Monitoring and Insulin Administration. Chichester, UK: John Wiley; 2017.

11.

Lori

Grazia

Bruce

et al . A practical approach to using trend arrows on the Dexcom G5 CGM system for the management of adults with diabetes. J Endocrine Soc. 2017;20:1445-1460.

12.

Clinical and Laboratory Standards Institute. Performance metrics for continuous interstitial glucose monitoring; approved guideline. POCT05-A. Wayne, PA: Clinical and Laboratory Standards Institute; 2008. Available at: http://www.clsi.org/source/orders/free/poct05-a.pdf.

13.

Pettus

Edelman

SV.

Recommendations for using real-time continuous glucose monitoring (rtCGM) data for insulin adjustments in type 1 diabetes. J Diabetes Sci Technol. 2017;11:138-147.

14.

Laffel

Aleppo

Buckingham

et al . A practical approach to using trend arrows on the Dexcom G5 CGM system to manage children and adolescents with diabetes. J Endocrine Soc. 2017;20:1461-1476.

15.

Klonoff

Kerr

A simplified approach using rate of change arrows to adjust insulin with real-time continuous glucose monitoring. J Diabetes Sci Technol. 2017;11:1063-1069.

16.

JDRF CGM Study Group. JDRF randomized clinical trial to assess the efficacy of real-time continuous glucose monitoring in the management of type 1 diabetes: research design and methods. Diabetes Technol Ther. 2008;10:310-321.

17.

Abbott. FreeStyle Libre manufacturer’s guidance. www.freestylediabetes.co.uk/images/uploads/documents/FreeStyle_Libre_Manual.pdf. Accessed September 2018.

18.

Dexcom. G5 Mobile manufacturer’s guidance. https://s3-us-west-2.amazonaws.com/dexcompdf/G5-Mobile-Users-Guide-Touchscreen-Receiver.pdf. Accessed September 2018.

19.

Medtronic. Guardian Connect manufacturer’s guidance. https://www.medtronicdiabetes.com/sites/default/files/library/download-library/user-guides/GuardianT%20Connect%20CGM%20System%20User%20Guide.PDF. Accessed September 2018.

20.

Eversense manufacturer’s guidance. https://www.eversensediabetes.com/wp-content/uploads/2018/08/LBL-1602-01-001-Rev-D_Eversense-User-Guide_mgdL_R1-2.pdf. Accessed September 2018.

21.

AGDT. Stellungnahme der AGDT zum Ersatz von Blutglukosemessungen durch Messungen mit Systemen zum kontinuierlichen real-time Glukosemonitoring (rtCGM) oder CGM mit intermittierendem Scannen (iscCGM). Available at: http://www.deutsche-diabetes-gesellschaft.de/fileadmin/Redakteur/Dokumente_Startseite/Teaser-Box/AGDT_Stellungnahme_Replacement_2016_11_11.pdf

22.

Schmelzeisen-Redeker

Schoemaker

Kirchsteiger

et al . Time delay of CGM sensors: relevance, causes, and countermeasures. J Diabetes Sci Technol. 2015;9:1006-1015.

23.

Pleus

Kamecke

Link

et al . Prediction quality of glucose trend indicators in current tissue glucose monitoring systems for use in therapeutic decisions. EASD congress Lisbon 2017. Available at: https://www.easd.org/virtualmeeting/home.html#!resources/prediction-quality-of-glucose-trend-indicators-in-current-tissue-glucose-monitoring-systems-for-use-in-therapeutic-decisions.

24.

Stellungnahme der AGDT zum Ersatz von Blutglukosemessungen durch Messungen mit Systemen zum kontinuierlichen Glukosemonitoring (CGM) oder Flash-Glukosemonitoring (FGM), 2016. Available at: http://www.deutsche-diabetes-gesellschaft.de/fileadmin/Redakteur/Stellungnahmen/2016/AGDT_Stellungnahme_Replacement_20160130-L.pdf.

25.

Diabetes Research in Children Network Study Group, Buckingham

Xing

et al . Use of the DirecNet Applied Treatment Algorithm (DATA) for diabetes management with a real-time continuous glucose monitor (the FreeStyle Navigator). Pediatr Diabetes. 2008;9:142-147.

26.

Scheiner

Practical CGM: Improving Patient Outcomes Through Continuous Glucose Monitoring. 4th ed. Alexandria, VA: American Diabetes Association; 2015.

27.

Ziegler

Cavan

Cranston

et al . Use of an insulin bolus advisor improves glycemic control in multiple daily insulin injection (MDI) therapy patients with suboptimal glycemic control: first results from the ABACUS trial. Diabetes Care. 2013;36:3613-3619.

28.

Barnard

Parkin

Young

et al . Use of an automated bolus calculator reduces fear of hypoglycemia and improves confidence in dosage accuracy in patients with type 1 diabetes mellitus treated with multiple daily insulin injections. J Diabetes Sci Technol. 2012;6:144-149.

29.

Sussman

Taylor

Patel

et al . Performance of a glucose meter with a built-in automated bolus calculator versus manual bolus calculation in insulin-using subjects. J Diabetes Sci Technol. 2012;6:339-344.

30.

Ziegler

Cavan

Cranston

31.

AGDT. SPECTRUM CGM-Schulungs- und Behandlungsprogramm. Mainz, Germany: Kirchheim Verlag; 2015.

32.

Kulzer

Flash Schulungsset. Mainz, Germany: Kirchheim Verlag; 2017.