Rapid, patient-led titration of basal insulin can be used to achieve tight glycaemic control in gestational diabetes mellitus with insulin resistance

Abstract

A woman with gestational diabetes mellitus (GDM) and significant insulin resistance in her third pregnancy was diagnosed with a fasting blood glucose reading of 5.7 mmol/L (103 mg/dL) at 28+1 weeks gestation and referred to our diabetes team. Using a rapid, patient-led approach to basal insulin titration this patient achieved therapeutic doses and glucose targets in the limited time available during pregnancy, without causing significant hypoglycaemia. This method of insulin titration empowers women with GDM to take control of their own management and could reduce complications in GDM pregnancies at negligible additional cost. The only additional cost being that of the higher insulin doses used.

Background

Many women with gestational diabetes mellitus (GDM) have insulin resistance and frequently require exogenous insulin. For women with GDM, our centre uses a novel patient-led approach to insulin titration (figure 1) which facilitates rapid glucose control even in the numerous women who eventually require very high insulin doses. This case illustrates the utility of patient-led insulin titration in the management of women with GDM and significant insulin resistance.

Figure 1.

The gestational diabetes mellitus treatment process adopted in our centre. Note: International recommendations for the management of post prandial glucose vary; the American Diabetes Association recommends avoiding the use of metformin in pregnancy if possible, whereas the National Institute for Health and Care Excellence (NICE) supports its use as a first line therapy. In our centre we provide both treatment options. The patient-led daily insulin intensification algorithm is described by the section below the dotted line. Daily increments are cautiously increased to 6 then 10 units in women who have glucose persistently above the fasting glucose target and where insulin resistance is suspected. *National criteria used for diagnosis in our clinic are those recommended by NICE (NICE guideline NG3).¹ Figure produced by the authors.

Case presentation

This woman in her early 40s, of white ethnicity, was diagnosed with GDM in her third pregnancy. The patient’s first pregnancy, 6 years prior, was uneventful except for an emergency lower segment Caesarean-section (LSCS), performed at 41+5 weeks for fetal distress and small size for gestational age (ninth centile) on antenatal scanning. A healthy female infant was born (birth weight: 3317 g (z-score −0.6; 27% centile)). During her second pregnancy, 3 years prior, the patient was referred to the diabetes team at 28+5 weeks after an abnormal oral glucose tolerance test (OGTT) performed due to her family history of her father and maternal grandmother having type 2 diabetes (T2DM), her prepregnancy weight was 86 kg (body mass index (BMI) 29 kg/m²). A diagnosis of GDM was made having excluded Glucokinase Maturity-Onset Diabetes of the Young (MODY-2) with genetic testing. Glycaemic control was eventually achieved between 35+6 weeks and delivery at 38+4 weeks using the nurse/clinician-led insulin titration approach in use in our clinic at the time. The patient’s maximum daily insulin dose was 62 units insuman basal (0.6 units/kg). A healthy female infant was born by elective LSCS (birth weight: 3160 g (z-score 0.0; 50% centile)) and insulin was discontinued postdelivery.

In her third pregnancy, OGTT was performed at 16+1 weeks due to previous GDM and her fasting (5.4 mmol/L (97 mg/dL)) and 120 min glucose (5.8 mmol/L (104 mg/dL)) were within the normal range.¹ At 28+1 weeks, she was seen in our multidisciplinary diabetes antenatal clinic following a fasting plasma glucose of 5.7 mmol/L (103 mg/dL) and a diagnosis of GDM was made. OGTT was unavailable due to COVID-19 limitations. She was then referred and seen at 28+1 weeks in our multidisciplinary diabetes antenatal clinic.

Her notable GDM risk factors included a prepregnancy weight, which had increased from her previous pregnancy to 104 kg (BMI 34.0 kg/m²), coupled with previous GDM and her family history. Her medical history included well controlled Crohn’s disease; antiphospholipid syndrome, Sweet syndrome and a pulmonary embolus 6 years prior. Her regular medications were enoxaparin, sertraline and aspirin.

Treatment

Throughout her third pregnancy, tight glucose control was maintained using our rapid patient-led insulin titration approach and metformin to achieve our fasting glucose target <5.3 mmol/L (<95 mg/dL) (figure 2). After 1 week of glucose testing at 29+4 weeks gestation insuman basal was commenced at an initial dose of 14 units nocte. The patient was advised to up-titrate her insulin dose by 4 units every evening following a fasting blood glucose reading of ≥5.0 mmol/L (90 mg/dL) as per our algorithm (figure 1). She followed this advice and at 30+2 weeks her dose was 30 units per day (0.29 units/kg) and her daily insulin dose increment was increased to 6 units. Despite this her fasting glucose remained persistently high and her increment was increased to 10 units per day at 30+5 weeks meaning her daily dose increased from 130 units to 260 units (1.25 to 2.50 units/kg) daily at 32+4 and 36 weeks. The patient reported no symptomatic episodes of hypoglycaemia at any point. Her dose of insulin was reduced at 36+1 weeks from 260 units to 250 units following an asymptomatic fasting glucose reading of 3.8 mmol/L (68 mg/dL). This dose then remained constant until delivery without any further low glucose readings.

Figure 2.

Daily fasting blood glucose (mmol/L) and basal insulin (units/kg) from insulin initiation at 29+4 weeks gestation to 36+4 weeks. Glucose data for the patients’ final week of pregnancy (delivered at 37+6 weeks) was not uploaded to the online platform and is therefore not available. The patient reports have no further low glucose readings (<4.0 mmol/L (<72 mg/dL)) during this period.

Outcome and follow-up

At 36+5 weeks her antenatal growth scan predicted fetal weight as 4058 g (>95% centile). An elective LSCS was performed at 37+6 weeks and a healthy male infant was born (birth weight: 3740 g (z-score 1.32; 91% centile)). There was no neonatal hypoglycaemia, and the patient discontinued basal insulin immediately after delivery. Her HbA1c 3 months postdelivery was normal at 38 mmol/mol. When asked, after delivery, about her experience of using the patient-led titration approach compared with the nurse and doctor-led approach the patient reported a very favourable experience (see below).

Discussion

Questions raised:

1. Given the short timeframe in GDM to achieve glucose control; what is the best strategy for safely achieving glucose target?

2. What is the existing evidence-base for insulin titration approaches in GDM?

Commentary:

Gestational diabetes mellitus is common and may lead to complications for both mother and baby if blood glucose is not tightly controlled. Moreover, there is very limited time from GDM diagnosis to delivery, in which to achieve this tight glycaemic control. Data from published observational studies suggest this is a widespread difficulty.^2–4 The patient characteristics and challenges described here are not uncommon and exemplify the need for swift insulin up-titration to maintain glycaemic control.

Current guidelines do not describe an insulin titration strategy in GDM, and a review of the literature did not identify any studies comparing alternative approaches to insulin titration in GDM. Physician-led and patient-led approaches to basal insulin titration in T2DM have been compared. One recent meta-analysis, including six studies (n=12 409, T2DM individuals) found that patient-led titration was correlated with higher basal insulin dose (5.9 units/day; 95% CI) 0.2 to 11.8), lower HbA1c −0.12% (95% CI −0.16 to −0.07) and lower fasting glucose (−0.29 mmol/L (−5.2 mg/dL); 95% CI −0.52 to −0.07 mmol/L (−9.3 to −1.2 mg/dL)) than physician-led titration. Patient-led insulin titration was associated with a negligible increase in the risk of hypoglycaemia (relative risk (RR)=1.12; 95% CI 1.02 to 1.23) however the risk of severe hypoglycaemia (requiring intervention) was not significant (RR=1.20; 95% CI 0.73 to 1.98).⁵ These results were consistent with the GDM-based findings of Mayne et al.⁶ Fasting glucose targets were also compared in a randomised control trial of patient-led basal insulin titration in people with T2DM which found that hypoglycaemic events were not significantly increased in patients with the lower glycaemic threshold (<5.3 mmol/L (<95 mg/dL)).⁷

A rapid approach to insulin titration is required in GDM and existing patient-led, T2DM titration strategies are inappropriately slow for adequate glycaemic control in these women. Most studies trialled 2–3 unit changes every 3–4 days, with the aim to achieve glycaemic control over a period of months.^8–12 In our patients’ case, the application of even the most rapid of these algorithms from 28+6 weeks (date of first blood glucose measurement) would have achieved a maximum insulin dose of just 64 units on the day of delivery (less than a quarter of the dose that was eventually required for our patient of 260 units).

In summary, the patient presented demonstrates the doses of insulin which may be needed in GDM and shows that a rapid, patient-led approach to insulin dose titration is a solution to achieve this without causing problematic hypoglycaemia. While our 4-unit daily insulin titration works for most women with GDM,^{6 13} this patient illustrates that even more rapid dose escalation may be required in women with insulin resistance. With this approach our patient had a healthy baby with normal birth weight and felt in control of her own management during pregnancy. It should, however, be noted that our algorithm has not been tested with long-acting insulin analogues as we use isophane insulin for GDM treatment in our centre.

There are numerous women with GDM, similar to our patient, who require large insulin doses to achieve glycaemic targets. Without a similar rapid dose escalation approach these women will not achieve therapeutic doses and are therefore exposed to increased risk of pregnancy complications. Our dose titration algorithm can achieve this at negligible additional cost attributed to increased insulin dose requirements, while empowering our patients to play a leading role in their diabetes care.

Patient’s perspective.

(The dose advice) was pretty simple for me. I liked being responsible for my treatment and felt more in control as a result. (I could) respond immediately to what my body was doing rather than waiting for a weekly review and worrying about the impact on the baby in the meantime.

Learning points.

• A rapid patient-led approach to basal insulin titration enables even significantly insulin resistant patients to achieve therapeutic doses and glucose targets in the limited time available during pregnancy and without causing significant hypoglycaemia.

• International adoption of this approach could reduce complications in gestational diabetes mellitus (GDM) pregnancies at negligible additional cost. The only additional cost being that of the higher insulin doses used.

• This method of insulin titration enables women with GDM to take control of their own management and makes them feel empowered.

Ethics statements

Patient consent for publication

Consent obtained directly from patient(s).