Review of Continuous Subcutaneous Insulin Infusion (Insulin pump) Therapy

1. Review of Continuous Subcutaneous Insulin Infusion (Insulin pump) Therapy Endocrine Morning Rounds October 8, 2008 Arthur Chung PGY-3

2. Introduction • Insulin pumps: how it works • Insulin pumps in Diabetes Mellitus, type 1 • advantages • disadvantages • Insulin pumps in Diabetes Mellitus, type 2 • Insulin pumps in pregnancy • Insulin pumps in perioperative care

3. What is insulin pump therapy? • The device includes: • The pump itself (including controls, processing module, and batteries) • A disposable reservoir for insulin (inside the pump) • A disposable infusion set, including a cannula for subcutaneous insertion (under the skin) and a tubing system to interface the insulin reservoir to the cannula. • With pump therapy, basal insulin is supplied in the form of a continuous infusion (comprising between 40 and 60 percent of the total daily dose) with pre-meal bolus doses given to minimize postprandial glucose excursions.

5. Good candidates: • Patients must monitor their blood glucose at least 4 times daily • learn to carbohydrate count • have a “rescue algorithm” (eg, a correction factor) • be sufficiently motivated • have an appropriate support system • possess problem-solving skills. Choice of insulin:  • Only short-acting insulin (regular), or rapid-acting insulins are used. • A double-blind crossover trial, insulin lispro, compared to regular insulin, resulted in greater reduction in postprandial blood glucose concentrations (by 25 mg/dL [1.4 mmol/L]), lower A1C (by 0.3 percent), fewer episodes of hypoglycemia, and less weight gain (Diabetes 1997 Mar;46(3):440-3). Similar results have been seen with insulin aspart (Diabetes Care 2002 Mar;25(3):439-44).

6. Getting started:  • Convert a patient from an MDI regimen to insulin pump therapy based on pre-pump level of chronic glycemia and total daily dose of insulin. • Eg. If previously well controlled with A1C <7.0 percent, the initial total daily dose of insulin administered by pump may be 10 to 20 percent less than the total daily dose of the previous regimen. • Approximately ½ of total daily dose = basal rate. • Eg. basal rates are in the range of 0.01 to 0.015 units per kg per hour (ie, for a 60 kg woman approximately 0.6 to 0.9 units per hour). The basal rates are adjusted empirically based on glucose monitoring results. • Infusion rates depending on individual factors including life-style and the "dawn phenomenon" which often occurs between 2 AM and 8 AM. • Most pumps allow for pre-programmed changes in basal rate to accommodate these requirements.

7. Insulin pump therapy in DM 1 What are the advantages of insulin pump therapy? 1) Improved glycemic control and less hypoglycemic events • Numerous trials have been done comparing glycemic control and hypoglycemia with CSII vs. MDI. • Using an insulin pump eliminates unpredictable effects of intermediate- or long-acting insulin. • Insulin absorption is less variable from day to day. Both the small subcutaneous depot, and the constancy of the injection site and depth for the two to three days with each catheter, contribute to the relative consistency of absorption.

8. Jeitler et al. 2008. Comparing CSII vs. MDI in DMI and DMII. • Methods: A meta-analysis of RCTs up to March 2007. • 22 trials: 17 on type 1 diabetes mellitus (908 patients), two on type 2 diabetes mellitus (234 patients), three on children (74 patients). • Adult DM I: 12 studies were used for the meta-analysis. • In 6 studies using HbA1c  −0.4% WMD in favour of CSII therapy over MDI. • In 6 studies using HbA1  - 0.6% WMD (not statistically significant) • No meta-analysis was performed on hypoglycemic events. • Mild or overall hypoglycemic events were comparable between the two groups: • CSII group: 1.9 [0.9–3.1] events per patient per week • MDI group: 1.7 [1.1–3.3] events per patient per week • No overall conclusions were possible for severe hypoglycaemia and adverse events for any of the different patient groups due to rareness of such events, different definitions and insufficient reporting.

9. Jeitler et al. Diabetologia (2008) 51:941–951.

10. Pickup and Sutton 2008 meta-analysis of CSII vs. MDI in DMI • Reviewed the frequency of severe hypoglycemia in CSII vs. MDI • Studied more recent trials because of poor quality of earlier trials • Analyzed the idea that CSII may be more beneficial in DMI patients who fail to achieve glycemic control on MDI. Method: • Databases and literature (1996–2006) were searched for randomized controlled trials (RCTs) and before/after studies of ≥ 6 months duration CSII and with severe hypoglycaemia frequency > 10 episodes/100 patient years on MDI. Results: 22 studies (21 reports). 1414 DMI patients for mean CSII duration of 6 – 28 months. • Severe hypoglycaemia during MDI was related to diabetes duration (P = 0.038) and was greater in adults than children (100 vs. 36 events/100 patient years, P = 0.036).

11. Severe hypoglycaemia was reduced during CSII compared with MDI, with a rate ratio of 2.89 (95% CI 1.45 to 5.76) for RCTs and 4.34 (2.87 to 6.56) for before/after studies [rate ratio 4.19 (2.86 to 6.13) for all studies].

12. The mean difference in glycated haemoglobin (HbA1c) between treatments was less for RCTs [0.21% (0.13–0.30%)] than in before/after studies [0.72% (0.55–0.90%)] but strongly related to the initial HbA1c on MDI (P < 0.001).

13. The worse controlled subject on MDI injections enjoyed the most improvement on the insulin pump therapy.

14. Pickup et al. 2002 meta-analysis of CSII and MDI in DMI Method: Meta­analysis of 12 randomised controlled trials (1975 -2000). 301 people with DMI allocated to insulin infusion and 299 allocated to insulin injections for between 2.5 and 24 months. Results: Glycated haemoglobin was also lower in people receiving insulin infusion (0.44, 0.20 to 0.69), equivalent to a difference of 0.51%. BMJ 2002 Mar 23;324(7339):705

15. Blood glucose variability is less in CSII compared to MDI • Bruttomeso et al. 2007. • Method: Randomized cross-over study. Compared CSII and MDI in patients with DM I. • 39 DM I patients with previously well controlled diabetes on CSII for at least 6 months were randomly assigned to continue CSII with lispro or switch to MDI with lispro and glargine. • After 4 months they were switched to the alternative treatment. • Primary end-points: • Blood glucose variability was analysed using various measurements: glucose standard deviation, mean amplitude of glycaemic excursions (MAGE), lability index and average daily risk range (ADRR). • Secondary end-points: blood glucose profile, HbA1c, number of episodes of hypo- and hyperglycaemia, lipid profile, free fatty acids (FFA), growth hormone and treatment satisfaction.

16. Results: • Overall, CSII glucose variability was 5–12% lower than during MDI with glargine. • The difference was significant only before breakfast considering glucose standard deviation (P = 0.011), significant overall using MAGE (P = 0.016) and lability index (P = 0.005) and not significant using ADRR. • Although HbA1c was similar during both treatments, during CSII blood glucose levels were significantly lower, hyperglycaemic episodes were fewer, daily insulin dose was less, • FFA were lower and treatment satisfaction was greater than during MDI with glargine. The frequency of hypoglycaemic episodes was similar during both treatments.

17. Diabet. Med. 25, 326–332 (2008).

18. Improved Quality of Life with CSII vs. MDI • Flexibility in the timing of meals: • Ability to use bolus insulin (rapid-acting insulin) whenever needed. • Ideal for patients who eat often or erratically or for patients who want the flexibility to eat what and when they want without compromising glycemic control. • Using an insulin pump means eliminating individual insulin injections • Insulin pumps allow you to exercise without having to eat large amounts of carbohydrate

19. Scheidegger et al. 2007. CSII: effects on quality of life. Methods: • Diabetes-specific quality of life was measured with the DSQOLS-Questionnaire in 81 adult subjects with DM I on MDI and 78 subjects on CSII (cross-sectional study). • 19 subjects were followed prospectively, measuring their quality of life before and after switching from MDI to CSII (longitudinal study). Results: • Patients on CSII were more satisfied with their treatment in respect to: • their metabolic goals, psychosocial factors, physical performance, protection from long-term complications and hypoglycaemia. • Furthermore, the subjects on CSII experience greater flexibility in their daily routines, leisure time and diet than the subjects on MDI. • Overall quality of life (+29 points, 95%CI 3 to 54) were significantly better in CSII compared to MDI only in the longitudinal study.

20. Swiss Med Wkly 2007;137:476–482.

21. Hoogma et al. Comparison of CSII and MDI on quality of life Methods: • The 5-Nations trial was a randomized, controlled, crossover trial conducted in 11 European centres. • 272 patients were treated with CSII  MDI or MDI  CSII x 6 months. • 3 questionnaires: DQofL, SF-12 health survey questionnaire, Lifestyle and manageability of disease Results: • The overall DQoL score was significantly higher for CSII at the end of treatment compared with MDI (75 vs. 71, P < 0.001),indicating a positive impact on QoL. • Improvements in treatment satisfaction (P < 0.001), treatment impact (P < 0.001) and a significant reduction in diabetes related worry (P < 0.01) when using CSII compared with MDI. • SF-12 questionnaire: improvements in perception of mental health when using CSII compared with MDI (P < 0.05). • Lifestyle and manageability questionnaire: more flexibility with regard to eating habits (P < 0.001) and lifestyle flexibility and sleep patterns (P < 0.001) when using CSII compared with MDI.

22. Insulin Pump Therapy in DM2 1) Glycemic control in DM 2 with insulin pump therapy Herman et al. CSII vs MDI in older adults (> 60 years) with DMII and marked obesity: Method: 2-center, 12-month, prospective, randomized, controlled clinical trial compared the efficacy and safety of CSII with that of MDI In this population, patients treated with both CSII and MDI achieved excellent glycemic control with good safety and patient satisfaction.

23. CSII group: HbA1c 6.6% (Mean A1c fell by 1.7 ± 1.0%) MDI group: HbA1c 6.4% (Mean A1c fell by 1.6 ± 1.2%) Diabetes Care. 2005;28:1568-1573.

24. There was no statistically significant difference in hypoglycemic events or severe hypoglycemic events between treatment groups; however, there was a trend for less hypoglycemic events with CSII. • CSII group: minor (self-treated) hypoglycemia 81%. • MDI group: minor (self-treated) hypoglycemia 90%. (P = 0.17) • CSII group: severe hypoglycemia 3 / 48 • MDI group: severe hypoglycemia 6 / 50. (P = 0.49) It is not clear if this study of elderly patients can be extrapolated to most patients with type 2 diabetes. Moreover, it suggests that MDI should be attempted in elderly patients with marked obesity before considering insulin pump therapy.

25. Jeiter et al. meta-analysis of 2 randomized control trials: • In patients with type 2 diabetes mellitus, CSII and MDI treatment showed no statistically significant difference for HbA1c. Cross-over trial of CSII and MDI: • Method: 40 obese, insulin-treated patients with type 2 diabetes were randomized to treatment with CSII or MDI therapy. • At the end of the first 18-week treat treatment period, patients underwent a 12-week washout period during which they were treated with MDI plus metformin. • They were then crossed over to the alternate treatment for an 18-week follow-up period. • Results of the intention-to-treat analysis showed that CSII was superior to MDI in reducing A1C values. There was no significant change in weight or insulin dosage. Diabet Med. 2005;22:1037-1046.

26. Quality of Life improved with CSII compared to MDI in DM2 Method: 132 CSII naive patients with type 2 diabetes were randomly assigned (1:1) to CSII (insulin aspart) or MDI therapy (bolus insulin aspart and basal NPH insulin). • Multicenter, open-label, randomized, parallel-group, 24-week study. • A total of 93% of CSII-treated patients preferred the pump to their previous injectable insulin regimen for reasons of convenience, flexibility, ease of use, and overall preference. Diabetes Care. 2003;26:2598-2603.

27. What are the disadvantages of insulin pumps? 1) Expensive cost of insulin pump therapy: • The costs of the pump and supplies are higher than those of ordinary syringes and needles. • Cost of pump: $3000 - 5000. Insulin infusion lines, syringes, tapes, batteries can add extra $40 – 50 / month. 2) Complications: • Moreover, approximately 40 percent of patients in one large study had one or more acute complications (such as infection at the site of needle insertion) during 3000 patient-months of treatment. (JAMA 1984 Dec 21;252(23):3265-9.) • A second study found that 86 percent of patients had at least one infusion-system failure in a 12-month period. (Diabetes Care 1986 Jul-Aug;9(4):351-5.)

28. Blockage or leakage in the syringe or the infusion set or connectors, causing an interruption of infusion flow. These issues now occur less commonly with more modern pumps and supplies. • Any interruption in continuous flow can lead very quickly to hypoinsulinemia, hyperglycemia, and possibly diabetic ketoacidosis. 3) Bothersome to wear an insulin pump: • Patients may feel "tethered" to the pump at all times. • Complaints that the treatment is awkward, uncomfortable, embarrassing, or unpleasant, particularly when bathing or having sexual intercourse.

29. Can insulin pump therapy be used in pregnancy? • A 2008 Cochrane review to compare CSII with MDI of insulin for pregnant women with diabetes. • Search strategy: Articles were searched from the Cochrane Pregnancy and Childbirth Group’s Trials Register (November 2006). • Selection criteria:Randomised controlled trials comparing CSII with MDI for pregnant women with diabetes. • Primary outcome: 1) perinatal mortality 2) fetal anomaly 3) hypoglycemia/hyperglycemic episodes requiring intervention 4) admission and length of stay • Secondary outcomes: multiple additional outcomes for the mother and baby

30. Results: • Two studies (60 women with 61 pregnancies) were included. • There was a significant increase in mean birthweight associated with CSII as opposed to MDI (WMD 220.56, 95% confidence interval (CI) -2.09 to 443.20; two trials, 61 participants). • BUT lack of significant difference in rate of macrosomia (birthweight greater than 4000 g) (relative risk (RR) 3.20, 95% CI 0.14 to 72.62; two trials, 61 participants). • No significant differences were found in any other outcomes measured, which may reflect the small number of trials suitable for metaanalysis and the small number of participants in the included studies. • No significant differences were found in: • perinatal mortality (RR 2.00, 95% CI 0.20 to 19.91) • fetal anomaly (RR 1.07, 95% CI 0.07 to 15.54) • maternal hypoglycaemia (RR 3.00, 95% CI 0.35 to 25.87) • maternal hyperglycaemia (RR 7.00, 95% CI 0.39 to 125.44).

35. Cochrane Conclusion • There is a dearth of robust evidence to support the use of one particular form of insulin administration over another for pregnant women with diabetes. • The data are limited because of the small number of trials appropriate for meta-analysis, small study sample size and questionable generalisability of the trial population. • Conclusions cannot be made from the data available and therefore a robust randomised trial is needed. The trial should be adequately powered to assess the efficacy of continuous subcutaneous insulin infusion versus multiple daily injections in terms of appropriate outcomes for women with diabetes.

36. Can insulin be used perioperatively? • Little evidence regarding CSII and perioperative management • 2 case studies. No randomized controlled trials available. • First case study: DM I who wore his continuous subcutaneous insulin infusion (CSII) pump during general anesthesia for surgical repair of a herniated lumbar disk. Blood glucose levels were stable throughout the perioperative period. Little or no extra work was required of the CRNA. • Second case study: A 61-year-old male, while recovering from a Whipple's procedure for pancreatic carcinoma, was treated for 13 days with an insulin infusion pump for diabetes exacerbated by enteral hyperalimentation. • Treatment with continuous subcutaneous insulin infusion resulted in improved blood glucose control. Associated with this improvement was a reduction in plasma cholesterol, triglyceride and free fatty acid levels. Plasma epinephrine, norepinephrine, glucagon and cortisol concentrations were also lowered although growth hormone levels remained unchanged.

37. Take Home Message • Insulin pump therapy is an alternative to MDI in diabetic patients • In DM 1, insulin pump therapy compared to MDI may lead to improvements in glycemic control with less frequent episodes of hypoglycemia, less glucose variability, and improved quality of life. • There is less evidence available in DM 2, but current trials suggest no additional benefit in glycemic control with CSII compared to MDI. However, CSII may be associated with improved quality of life. • There is currently very little evidence for the benefit of insulin pump therapy in pregnancy or perioperative management of diabetes.

38. Thank you