In our last two posts on diabetes in pregnancy, we discussed the importance of diet and exercise and the value of self-monitoring as the foundations of blood sugar control. If these measures fail to achieve normalization of blood sugars (fasting values < 95 mg/dL and 2 hour after meal values of < 120 mg/dL), then medical therapy should be started. Although there are various opinions regarding approaches to medical therapy and endpoints for adjusting and monitoring efficacy of therapy, most will agree the primary goal should be normalization of blood sugars. Secondary goals include, among others, minimizing fetal overgrowth (macrosomia), avoiding intrauterine fetal death, reducing the risks for fetal damage during the delivery (e.g., shoulder dystocia, nerve palsies, bone fractures) resulting from fetal macrosomia, and preventing neonatal complications related to hypoglycemia and hyperbilirubinemia, as well as admissions to the neonatal intensive care unit.
During my training (too many years ago), insulin was really our only option for therapy for the woman with gestational diabetes mellitus (GDM) and it is still the ‘gold standard’ against which other approaches to therapy must be measured. Despite the fact that insulin is comparatively expensive, requires multiple injections during the day combining different types of insulin to optimize control, requires frequent adjustments in dosing regimen (requirements change and usually increase as the placenta grows, hormone production increases and pregnancy progresses), is prone to greater risks for complications related to self-administration dosing errors, variability of response, and hypoglycemia, and often seems to terrify patients, in my experience its use can be learned by most patients (with or without the help of other family members) and, when well-understood by the patient in terms of her own responses, offers remarkable flexibility in management, accommodating variability in dietary intake and activity levels.
When estimating a starting dose for insulin, we use a combination of maternal weight (in kg) and pregnancy trimester. First, a 24-hour total insulin dose is calculated using 0.7 units (U)/kg in first trimester, 0.8 U/kg in second trimester, and 0.9 U/kg in third trimester. (To quell the usual concerns of our patients that this is “too much,” I usually tell them that others recommend starting doses as high as 0.8 U, 1.0 U, and 1.2 U, respectively, in each trimester). Two-thirds of the total dose is given in the morning before breakfast and one-third at night (half of that before supper and half prior to bedtime). Two-thirds of the morning insulin dose is then given as an intermediate-acting insulin (such as NPH) and one-third as a short-acting insulin (such as Regular insulin or Humolog) with both being given in a single injection. The goal of the combination therapy is to distribute the insulin action throughout the day, minimizing the number of injections that are required, and accommodating for the meals and snacks that will be consumed during that period of time. The insulin used just prior to supper is again the short-acting type alone, and prior to bedtime, the intermediate acting type alone. To give an example of how this works, suppose we have a 100 kg woman who is diagnosed with gestational diabetes at 30 weeks gestation:
_The TOTAL 24-hour insulin dose is 100 kg X 0.9 U/kg = 90 U insulin;
_The prebreakfast dose then totals 60 U (two-thirds of 90 U), with 40 U (two- thirds of 60 U) given as intermediate-acting insulin and 20 U as short-acting insulin;
_The evening dose then totals 30 U, divided as a presupper dose of 15 U of short-acting insulin; and,
_A bedtime (qhs) dose of 15 U of intermediate-acting insulin.
Some women can be adequately controlled by combining both of the evening doses of insulin prior to supper (just as before breakfast), but we have found over the years that, generally, distribution of dose and insulin type as detailed above achieves a more stable, physiological control of blood sugars and minimizes risks for large swings in glucose levels over the 24-hour period.
The combination of short- and intermediate-acting insulin allows for altering doses in a way that readily accommodates the patient’s eating and activity habits. Within a week (if not sooner) after starting insulin, adjustments in the insulin regimen can be made that are based on the fasting and 2-hour after meal blood sugar determinations (detailed in our last post on this subject). For example, if the blood sugars two hours after breakfast are still elevated, but the 2-hour after lunch levels are normal, then the morning dose of short-acting insulin can be increased independently of the intermediate-acting insulin to improve control earlier in the day with very little, if any, effect on control later on. Similarly, if the after breakfast blood sugars are good to start with, but the afternoon ones are elevated, increasing the morning intermediate-acting insulin alone might adequately improve control. I think you get the picture without belaboring all the alternatives! Other options for administration of insulin exist, for example insulin pumps, more frequent injections of short-acting insulin, and ‘sliding scales’ based on individual calculated responses to various foods, or to cover unexpectedly (and intermittently) elevated blood glucose levels, but most women are well-handled on the regimen detailed above.
One of the big fears (other than that of the shots themselves) that women starting insulin have is that they will get low blood sugar (hypoglycemia). As we discussed earlier in this series, gestational diabetes is, usually, more akin to type 2 diabetes than to type 1 diabetes. Women with GDM, therefore, tend to be relatively ‘insulin-resistant’, especially if they are overweight to start. Indeed, many women with GDM require much higher doses of insulin than is calculated as a ‘starting’ dose based on weight and gestational age. Hypoglycemia, therefore, is not a frequent problem in most cases of GDM and when it does occur, it is usually accompanied by symptoms (such as shakiness, palpitations, sweating, anxiety, weakness, difficulty talking) and it is rarely as profound, or as unpredictable, as it is in women with long-standing type 1 diabetes. Since the ‘symptoms’ of hypoglycemia are so common in pregnant women anyway, we encourage our patients on insulin not to overreact when they think their blood sugars are getting low, but to recheck their blood glucose levels before ‘self-medicating’ with additional caloric intake. Most women with GDM do not need prescriptions for glucagon to manage periodic lows in their blood sugars. However, as a word of caution, I always do worry if a woman’s insulin requirements begin to drop dramatically over what has been needed to maintain good glycemic control. This may be a sign of ‘placental insufficiency’, signaling a baby or a pregnancy at risk for complications but it is, again, more of a concern in type 1 diabetics than in women with GDM.
Although the primary reason for starting medical therapy in women with GDM is control of blood sugar levels, others have suggested that maternal blood sugar control alone, even if it appears adequate, may not prevent complications of pregnancy. Buchanan and colleagues (Diabetes Care 1994;17:275-83) identified women between 29 and 33 weeks with GDM that appeared to be well-controlled on diet alone, whose babies had abdominal circumference measurements that were greater than the 75th percentile, suggesting risk for fetal macrosomia , and hence risk for many of the other complications detailed in the first paragraph above. These women were randomized either to continue on diet alone or to have insulin added to their diet treatment. Interestingly, despite no significant differences in blood sugar measurements between the two groups, the insulin-treated women had only a 13% rate of fetal macrosomia compared to the 45% rate seen in the women treated with diet alone.
I have begun the discussion of medical therapy for GDM with a basic review of insulin therapy simply because of its historical significance, because it remains the ‘gold standard’ for care, and because it is the final resort if other approaches to therapy fail. However, in recent years the safety and efficacy of oral hypoglycemic agents have also been well-documented and, indeed, in many institutions they have become the treatment of first choice in women with GDM. A discussion of these drugs will be the focus of our next post on the topic of diabetes in pregnancy…