Quick Tips

Healthcare Provider Q & A

Click on a discussion question to view feedback provided by content experts.

A physician affiliated with our hospital has a pregnant patient who requires an oral glucose tolerance test. This patient reports an anaphylactic allergy to corn and, refuses to complete the testing with our stocked D-glucose solution (Simply Pure; Azer Scientific) because the D-glucose is derived from corn. The only alternative agents that we have been able to identify on the Canadian market, Trutol 75 g Glucose Tolerance Test Beverage (Fisher Diagnostics) and Jamp-Glucose 75 by (JAMP Pharma Corporation) also include D-glucose derived from corn. We were wondering if any of the members of the CDA Clinical Practice Guidelines Expert Committee have encountered this scenario before and, if so, how the situation was handled. We welcome any practical, practice-based tips for managing this scenario.

I have not encountered this issue before. A couple of things, you say the person requires an OGTT, is this just standard screening or is there some stronger indication (e.g. previous IGT, big baby, very strong family history). How many weeks pregnant is your patient?

If this person is a high risk individual I think it would be reasonable to simply start them monitoring their glucose, four times a day (fasting and post meals, one or two hours) for a couple of weeks, if all is perfect then could cut back to less frequent and if still fine could maybe discontinue.

If this is not a high risk situation then I would check a fasting and a 2 hr pp after a large meal and if okay would feel reassured. Alternatively is there a reason you could not mix 75 gm sucrose in water and use this as the test load, not perfect maybe but reasonable.

Edmond Ryan MD FRCPC

My colleagues and I recently opened a discussion in our department on the subject of "rosiglitazone" being identified as having a CV "neutral" effect in the "Cardiovascular Outcome Trial", while still requiring therapeutic considerations "identified cardiovascular controversy" (Figure 1: Pharmacological Management of Type 2 Diabetes: 2016 Interim Update"). Would it be possible to clarify?

While there remains a "cardiovascular controversy" because of the flawed Nissen meta-analysis, the definitive CV outcome trial for rosiglitazone, RECORD, was neutral for the primary CV outcome.

Ronald M. Goldenberg, MD, FRCPC, FACE
LMC Diabetes and Endocrinology

(Pharmacological Management of Type 2 Diabetes: 2016 Interim Update)

Since the new changes in the Guidelines about SGLT2 inhibitor agent use in T2DM management, I wonder if SGLT2i should be continued or, stopped when a person starts insulin? Other than metformin, what other anti-hyperglycemic agents can be combined with insulin?

SGLT2i are usually continued when insulin is initiated and, all three SGLT2i in Canada are approved for use with insulin. Generally, all anti-hyperglycemic agents are continued when initiating basal insulin except for thiazolidinediones (TZDs) and, sulfonylureas (SUs); some clinicians stop SUs, but another approach is to only stop the SU when prandial insulin is started.

Ronald M. Goldenberg, MD, FRCPC, FACE
LMC Diabetes and Endocrinology

Why are the values for the diagnosis of GDM different between the preferred approach vs the alternative approach following the 75 OGTT?

Both approaches take their numbers from the HAPO study where fetal/neonatal outcomes were associated with levels of glucose on the OGTT. The preferred approach takes the glucose values that are associated with a two-fold risk of neonatal outcomes, whereas the alternative approach takes the glucose values that are associated with a 1.5-fold risk of neonatal outcomes. There was dissent among the group as to which threshold to use, and because the risk is continuous, there is no definite threshold to take. Therefore, both thresholds were quoted in the guidelines.

Denice Feig, MD, MSc, FRCPC

What is an SGLT2 inhibitor?

Every day, about 180 grams of glucose is filtered through our kidneys. However, glucose is a critical fuel so there are important mechanisms to bring the glucose back into the body so that it is not entirely excreted in urine. The "doorways" in the nephron through which glucose is reabsorbed are the sodium glucose linked transporters 2 (SGLT2). The SGLT2 inhibitors block these "doorways", thereby allowing for glucose to be excreted in the urine. There are three SGLT2 inhibitors approved in Canada - canagliflozin, dapagliflozin, and empagliflozin. The recent interim update to the Pharmacologic Management of Type 2 Diabetes chapter incorporates this class of medications into the algorithm (http://guidelines.diabetes.ca/browse/chapter13_2015).

Alice Cheng MD, FRCPC & Chair CDA CPG

When you state mild to moderate hypoglycemia, what should the capillary blood glucose ranges be? Your literature states less than 4 mmol/L. Our thinking is 3.1 – 3.9 mmol/L. Can you confirm this please? Severe hypoglycemia is also stated as less than 4 mmol/L. If we were to put into our algorithm “equal to or less than 3mmol/” would you consider this acceptable?

As you know, the definitions of mild, moderate, and severe are based on clinical presentation (need for assistance) and less so the actual glucose levels. Table 2 of the Hypoglycemia chapter gives a value of <2.8 as the typical level at which severe Hypoglycemia occurs.

For your institution, you will need to decide what makes the most sense based on your patient population and staff logistics etc.

FYI, a sample template hypoglycemia clinical protocol will be made available on the guidelines website in the near future. Perhaps that will be of assistance to your group!

Alice Cheng MD, FRCPC & Chair CDA CPG

What are the best practice recommendations for diabetic foot screens?

In people with diabetes, foot examinations by healthcare providers should be an integral component of diabetes management to identify persons at risk for ulceration and lower-extremity amputation [Grade C, Level 3 (1,2)] and should be performed at least annually and at more frequent intervals in those at high risk [Grade D, Level 4 (3)]. Assessment by healthcare providers should include the assessment of skin changes, structural abnormalities (e.g. range of motion of ankles and toe joints, callus pattern, bony deformities), skin temperature, evaluation for neuropathy and PAD, ulcerations and evidence of infection [Grade D, Level 4 (3)].


  1. F. Crawford M. Inkster J. Kleijnen T. Fahey Predicting foot ulcers in patients with diabetes: a systematic review and meta-analysis QJM 100 2007 65 86
  2. Y. Feng F.J. Schlösser E. Bauer B.E. Sumpio The Semmes Weinstein monofilament examination is a significant predictor of the risk of foot ulceration and amputation in patients with diabetes mellitus J Vasc Surg 53 2011 220 226
  3. A.J. Boulton D.G. Armstrong S.F. Albert American Diabetes Association American Association of Clinical Endocrinologists Comprehensive foot examination and risk assessment: a report of the task force of the foot care interest group of the American Diabetes Association, with endorsement by the American Association of Clinical Endocrinologists Diabetes Care 31 2008 1679 1685

A tool which you may find beneficial includes:


Chantel Dougall
Coordinator, Clinical Practice Guidelines

What are the reasons behind the new ECG requirements for physical activity?

The Guideline for a baseline ECG prior to exercise is not necessarily a new Guideline. It is also not a Guideline specific to physical activity. This Guideline is in accordance with Chapter 23, "Screening for the Presence of Coronary Artery Disease” which includes these recommendations for ECG testing:

  1. A baseline resting ECG should be performed in individuals with any of the following [Grade D, Consensus]:
    • Age >40 years
    • Duration of diabetes >15 years and age >30 years
    • End organ damage (microvascular, macrovascular)
    • Cardiac risk factors
  2. A repeat resting ECG should be performed every 2 years in patients with diabetes [Grade D, Consensus].

In the physical activity chapter of the 2013 Guidelines, we changed the recommendation to:

  1. People with diabetes with possible CVD or microvascular complications of diabetes who wish to undertake exercise that is substantially more vigorous than brisk walking should have medical evaluation for conditions that might increase exercise-associated risk. The evaluation would include history, physical examination (including funduscopic exam, foot exam, and neuropathy screening), resting ECG and, possibly, exercise ECG stress testing [Grade D, Consensus].

Marni Armstrong PhD, CEP

Do the guidelines remain the same (50g or OGTT at 24-28 weeks) for women pregnant with twins?

Twin (or more) pregnancies were not listed as a risk factor for earlier screening of gestational diabetes mellitus (GDM). Therefore, twin pregnancies can be screened at the same time as singleton pregnancies.

Alice Y.Y. Cheng, MD, FRCPC

I recently had a client tell me they “used to have Type 2 Diabetes but since losing a lot of weight a few years back I don’t have it anymore”. My understanding was that once you are diagnosed with diabetes you will always have diabetes but you can keep it under control. Can you please clarify?

This is an excellent question that is not specifically addressed in the guidelines. There are different opinions out there and here is mine. Type 2 diabetes is typically a chronic progressive disease, due to pancreatic beta-cell dysfunction and insulin resistance that is not reversible. In rare cases of type 2 diabetes (and sometimes seen with diabetes of other causes), with a significant amount of weight loss, the individual has been able to reduce the insulin resistance, and thus maintain normal blood glucose. While the individual technically at that time may not meet the diagnostic criteria for diabetes (could confirm by doing a 75-g OGTT), they would still be at risk for "redeveloping" diabetes or dysglycemia with time or changes in weight. A term that may better reflect this could be "well-controlled diabetes".

Dr. Catherine Yu MD FRCPC MHSc

We know improving A1C improves outcomes but that implies an A1C derived from tight glucose control. What do we do with patients who are getting info from their HCPs that they are doing a good job because A1C is 6.5% - but if provider actually looked at BG log, they would discover widely fluctuating BGs – i.e. 3 to 18.

A1C represents an average of blood glucose levels over 3 months and is not a reliable marker of glycemic variability. The importance of glycemic variability is being studied and clarified. At this time, there is no easy-to-use test to measure the degree of glycemic variability. In addition, the “ideal” degree of glycemic variability has not been determined. Therefore, glycemic variability is not a recommended target of treatment. However, hypoglycemia is recognized as a negative outcome and treatments should be altered to minimize hypoglycemia.

Can you comment on starting insulin early in the treatment regimen for a short period of time, for the purpose of preserving beta cell function? Not necessarily on a patient with extremely high A1C.

As per the 2013 CDA CPGs, insulin is appropriate for use at any stage of management of a person with type 2 diabetes. The possibility of reducing progression to diabetes (in prediabetes) or even reversing type 2 diabetes (in newly diagnosed individuals) has been studied and ongoing trials will provide further information.

What is a reasonable approach for dose adjustment for gliclazide in renal impairment?

Some people have written off TZDs as a group. In what patient population would you use pioglitazone?

TZDs remain a valid treatment option after metformin. As with any other agent choice, one needs to consider both patient and agent characteristics to determine if it is appropriate. As per the Table provided in the Pharmacotherapy for Type 2 Diabetes tool, one can determine the pros and cons of the class.

Should one be concerned about the effect of statins on blood glucose levels in a person with diabetes?

This is an excerpt from the Dyslipidemia chapter of the 2013 CDA CPGs:

“Although statins are the cornerstone of lipid-altering therapy for CVD risk reduction in people with or without diabetes, recent evidence has suggested that chronic statin use is associated with an increased risk of incident diabetes. The interplay between statin therapy and incident diabetes was highlighted in a prespecified analysis of the West of Scotland Coronary Prevention Study (WOSCOPS), which actually showed a decrease in the incidence of new-onset diabetes with statin therapy (88). In contrast, Justification for the Use of Statins in Prevention: an Intervention Trial Evaluating Rosuvastatin (JUPITER) showed an increase in incident diabetes (89). Several meta-analyses suggest that there is indeed a small overall increase in diabetes with chronic statin use (90,91) and that this risk may be related to the statin dose (92).
Although this finding is of little relevance to patients with established diabetes, it may be of relevance to patients who are at risk for developing diabetes irrespective of statin treatment, such as those who are obese and/or who manifest metabolic syndrome. However, as discussed earlier, even these patients with risk factors for the development of diabetes enjoy a marked benefit in CVD risk reduction through the LDL-C–lowering effects of statins, which appears to far outweigh any small risk of new-onset diabetes (47,48). Accordingly, these recent analyses do not affect the recommendation that statins are the preferred agents for lowering LDL-C in most instances, including in patients with established diabetes or in those with risk factors for developing the disease.”

What is the effect of ethnicity on A1C for diagnosis of type 2 diabetes?

Studies of various ethnicities indicate that African Americans, American Indians, Canadian First Nations, Hispanics and Asians have A1C values that are up to 0.4% higher than those of Caucasian patients at similar levels of glycemia. Until ethnic specific thresholds for A1C as a diagnostic test for diabetes are determined, it is recommended that the other validated tests for diabetes diagnosis be used for these ethnicities.

Please comment on the use of point-of-care capillary glucose or A1C testing by diabetes centres for screening.

Point-of-care capillary glucose or A1C testing should not be used for definitive diabetes diagnosis, and any abnormal values should be confirmed with our standard laboratory tests.

How accurate is an A1C – error in assay? Should we check liver function / enzyme and ferritin to use A1C for diagnosis?

While there can be slight intraindividual variability in A1C testing, this parameter is less variable than the alternative tests of FPG or 2h-PG in an 75-g OGTT. A1C is misleading in hemoglobinopathies, hemolytic anemia, iron deficiency, and severe renal or liver disease, so it would be important to know if your patient has these conditions before using A1C. Clinical judgment is recommended in looking for these conditions, but since CBC, ALT and creatinine are fairly routine measures, information suggestive of the above conditions will often be available prior to A1C testing.

Is anyone researching First NationsCanadian A1C thresholds for diagnosis?

This is an area that still requires active investigation in various First Nation populations in order to determine the threshold level of A1C for the risk of retinopathy, and hence the cut point A1C for diabetes diagnosis in this population.

When confirming the diagnosis of diabetes, when should the A1C be repeated (if that was the initial screening test)?

The confirmatory test for diabetes diagnosis following a single test in the diabetes range can be done at any time following the initial test, preferably in a timely fashion so that diagnosis is not delayed. There is no need to wait 3 months for a confirmatory A1C.

When confirming the diagnosis of diabetes, when should the A1C be repeated (if that was the initial screening test)?

The confirmatory test for diabetes diagnosis following a single test in the diabetes range can be done at any time following the initial test, preferably in a timely fashion so that diagnosis is not delayed. There is no need to wait 3 months for a confirmatory A1C.

Is there a version of the diabetes flow sheet for EMRs?

We are in the process of creating flowsheets to facilitate guideline implementation. Currently this has been completed in 2 systems: Optimed-Accuro and MedAccess. Please contact us at guidelines@diabetes.ca if you are interested in integrating this into your EMR. When these are more systematically completely, we will update guidelines.diabetes.ca with this information, under “Team care and organization of care”.

Why does clear insulin need to be drawn into the syringe first?

As you may know, mixing of insulins and administration of insulins via syringe and needle is quite rare in comparison to the use of pre-mixed insulinsand the administration of insulin using an insulin pen and pen needles.

There is no therapeutic reason for drawing clear insulin first. Neither insulin should be contaminated with insulin from the other vial. “Best practice" states that if you overdraw on the second, cloudy insulin, you cannot push the mixed solution from the syringe back into the long-acting cloudy insulin vial. Rather, you would have to remove the syringe from the second, cloudy insulin vial, discard the syringe and restart the procedure of drawing up each insulin, beginning with the clear insulin before the cloudy insulin.

In addition, there is no adverse effect on the pharmacokinetics of the insulins. A slight decrease in the absorption rate, but not the total bioavailability, is seen when rapid-acting and protamine-stabilized insulin(NPH) are mixed. In clinical trials, however, the postprandial blood glucose response was similar when rapid-acting insulin was mixed with either NPH or ultralente. When rapid-acting insulin is mixed with either an intermediate- or long-acting insulin, the mixture should be injected within 15 min before a meal. (Reference: http://care.diabetesjournals.org/content/25/suppl_1/s112.full)

Perhaps, the suggestion to draw clear insulin first exists to reduce errors. If you add cloudy insulin to clear insulin, you can see the solution change from clear to cloudy. The reverse is not true. If you begin witha syringe of cloudy insulin, you cannot "see" clear insulin going into cloudy insulin. This visual effect provides an added step of confirmation which may reduce unfortunate human error.

After drawing up both insulins, it is NOT necessary to tip the syringe back and forth to mix the insulins further before administering. The mixing should have occurred before the insulin was drawn into the syringe. The tipping and rolling of a cloudy insulin vial/cartridge/pre-filled pen prior to withdrawing a dose is due to the need to ensure a homogenous suspension of the cloudy insulin. This is important to ensure the most consistent insulin release over the course of the entire contents of the vial/cartridge/pre-filled pen. However, once this tipping and rolling has occurred and the withdrawn solution was properly mixed, then there is no further need to tip the syringe.

Two references are below for more information:

Susie Jin (Pharmacist)

What is the recommended blood pressure target for people with prediabetes?

There are no blood pressuretargets specific for prediabetes. Therefore, one would target as they would anyone else with hypertension which is <140/90 mmHg.

We work with long term care and advanced care facilities, and there is a question as to whether they should be recapping pens. There was a theoretical contamination issue brought forth by one of the nurses. The pens are only being used for one patient so it is not a question of cross-contamination risk. However, the question is specifically with regards to what is the risk of recapping the pen cap after use, and is there one?

The 2013 CDA Guidelines do not specifically address insulin injection techniques however; CDA has reviewed a document that you might find helpful. FIT for SAFETY is an initiative of many diabetes educators in the country to establish safe and effective injection technique. It has long been established that re-capping needles with the inner smaller cap increases the risk of accidental needle stick injury and, therefore is not to be performed. The outer needle cap is used to unscrew / disconnect the needle from the pen device. I hope this will help you with your practices with needle re-capping. You may access more information at the fit4diabetes website.

Lori Berard RN CDE

I am looking for some sort of definition or calculation tool to be able to identify "limited life expectancy". I am a pharmacist working in our diabetes care center and have been unable to get any sort of answer.

This is a good question and the reason you have not been able to find a definition anywhere is because there isn't a good one. The concept is that it takes time for complications of poor glycemic control to manifest. Therefore, if someone doesn't have time ... then no need to worry about preventing long term complications. It probably takes about 5 years for complications to develop from scratch. Based on opinion (and not from the CPGs), if someone is not expected to live another 5 years, then that may be considered as "limited". As you well know, we do not have a way to accurately predict life expectancy and comorbidities will only provide us with clues. Therefore, this is a "clinical judgement" kind of statement.

Could you please recommend the appropriate interval of face-to-face follow up for people with diabetes? We discussed that it would depend on the individual's stability and risk factors and we practice the recommendation to monitor HbA1C every 3 months. But as a person hopefully does well with self-management, and their medical treatment plan has stabilized, how often do you recommend there is a face-to-face appointment with the care provider team?

You are correct that it will depend on the individual's level of control, stability, risk factors etc. If everything is at target, then once every 6 months may be reasonable. If not, then once every 3 months (for the A1C) and perhaps more frequently (for other adjustments) would be appropriate. There is no single right answer here!

The guidelines recommend that all people with diabetes whose age is 40 years or older should be offered statin therapy with an LDL-cholesterol target of ≤2.0 mmol/L. What should we do with patients whose baseline LDL-cholesterol is already less than 2.0 mmol/L?

If the decision is made to treat a patient with statin therapy (ie. age ≥40 years), then titrate the statin until the LDL is ≤2.0 mmol/L. If the LDL is already ≤2.0 mmol/L at baseline, then just start low dose statin therapy and there is no need to repeat the LDL since it will only go lower.

Does Appendix 7 (Sick day medications) apply to all patients with type 2 diabetes or just patients with nephropathy?

The Sick Day Medication List is intended for anyone with diabetes (type 1 or type 2) but particularly those with any renal impairment since they are at higher risk of worsening renal function in the context of dehydration. Clinical judgement is required when deciding to use the tool. For example, I'll give you 2 cases:.

Case A: Patient who already has a reduced eGFR (45 mL/min) ... you would definitely be concerned because if he/she gets a little dehydrated, that eGFR may fall to <30 mL/min quickly.

Case B: Patient with a baseline eGFR of 80 mL/min. That person can also have significant drop in eGFR with dehydration but it would have to be more severe dehydration to do that.

I cannot give you clearer definitions of dehydration or a specific level of GFR but the concept applies to everyone with diabetes (and frankly, even those without diabetes!).

Is doing an A1C in the first trimester an acceptable screening method for pregestational diabetes? And if so, what would be the threshold value?

Are there any populations in which testing A1C would be acceptable in GDM? Does the committee see a role for A1C in screening for either pregestational diabetes or gestational diabetes? If so, what would the thresholdsbe?

Does the committee have any comments on recommendations around weight gain during pregnancy in women who have diabetes?

In the 2013 CDA CPGs, the A1C is notrecommended as a screening tool to diagnose a pregnant woman with either gestational diabetes or diabetes that predated the pregnancy. The reason is because the A1C has not been fully validated in a pregnant population for diagnosis. Having said that, you can turn to the IADPSG (International Association of the Diabetes in Pregnancy Study Group) consensus statement (Diabetes Care 2010;33:675-682) to look for others' opinions. They do suggest the use of A1C as an option in early pregnancy BUT they recognize the lack of validation and a variety of caveats. Their suggestion is to measure FPG or random PG or A1C at the initial prenatal visit and they provided thresholds for diagnosis of overt diabetes which correspond with the thresholds outside of pregnancy. There is no A1C threshold for the diagnosis of Gestational diabetes.

The 2013 CDA CPGs do specifically address weight gain in pregnancy. The Institute of Medicine in 2009 put out recommendations for weight gain in pregnancy based on initial BMI. It is shown as Slide #61 in the Pregnancy slide deck available if you go to the Pregnancy chapter under "Browse the 2013 CPGs" or if you go directly to the "Slides and Videos" link on the main navigation menu.

What falls into the category of “hemoglobinopathies”?

Hemoglobinopathies that affect the A1C are known documented ones such as thalassemia, sickle cell, hemoglobin C, E, S or others. Also, any cause of anemia, such as iron deficiency or hemolysis, would also affect the A1C.

In addition, please check out the Canadian Journal of Diabetes, July 2011 issue. There is a position statement on the use of A1C and there is a great table for when A1C can be affected.

What is the proper definition of obese in adolescents? If it is the 97th percentile in Canada, then shouldn’t the 2013 CPGs reflect that in the risk factors for type 2 diabetes for adolescents?

Although the Canadian Paediatric Society and Dietitians of Canada have adopted the WHO charts (defining obesity as 97th percentile), most pediatric centers in Canada have not adopted these charts as they are controversial in their current format, especially for teenagers. In addition, the studies that identified obesity as a risk factor for type 2 in children were performed using the Centre for Diabetes Control (CDC) growth charts and defined obesity at the 95th percentile. That is the why the decision was made to maintain the reference from the CDC growth charts definition. The CDC charts continue to be the more commonly used charts in Canadian pediatric centres.

The definition of obesity is dependent on which growth charts are used: CDC charts = 95th percentile; WHO growth charts = 97th percentile.

Although there will be a small percentage of children who would be characterized as obese with the CDCcharts and not with the WHOgrowth charts, please note the risk of diabetes based on obesity is a continuum. Therefore, in the absence of good evidence for either cut-off on the clinical outcome, there is no evidence to favour one definition over the other in the guidelines. The intent of this recommendation is to screen obese children at large so that those who develop diabetes can be identified earlier.

The CDA paediatrics group endorses the use of either the CDC or the WHO charts definition of obesity to identify children and adolescents at risk. Whichever one you typically use will be the one you will continue to use.

Why did the 2013 CDA Clinical Practice Guidelines not adopt the notion that screening should be based entirely on a validated risk calculator (FINDRISK or CANRISK) rather than from age 40?

The CDA expert committee decided against going exclusively with the risk assessment tool because of the following reasons:

  • CANRISK has only been validated for those age 40-74 years
  • Widespread adoption and use of yet another risk stratification tool will be challenging (The Framingham risk assessment which has been recommended for over 2 decades is still not done universally in the primary care office)
  • There remains a need for a simple threshold for screening for better adoption and use

However, there is recognition that some may choose to use the CANRISK tool, especially if embedded into their electronic systems, so the option to use the risk assessment tool remains.