sign in sign up
<<
Home , Bile acid sequestrant, Colestilan, Colestipol

The Effect of on Glycemic Control in Patients With Type 2 Diabetes

Elizabeth A. Sauter, PharmD; Sindhu Abraham, PharmD, BCPS; Tania G. John, PharmD; Seema Kapadia, PharmD, BCACP; and Judith A. Toth, PharmD, CGP, CDE

Using an electronic chart review, the authors investigated the effect of colestipol, a acid sequestrant, on glycemic control in patients with type 2 diabetes.

iabetes mellitus is a complex cations. In general, every percentage group and the placebo group of disease, characterized by hy- point drop in glycosylated hemoglo- − 0.5% (P = .007). Additionally, in perglycemia, and can be as- bin (A1C) can reduce the risk of mi- participants with a baseline A1C ≥ sociated with abnormalities crovascular complications (ie, eye, 8%, the difference in mean change in D 2 in fat, carbohydrates, and protein kidney, and nerve diseases) by 40%. A1C was − 1.0% (P = .002). Treat- that result from defects Improved control of LDL-C can re- ment with was also in insulin secretion, insulin action, duce cardiovascular complications by associated with a reduction in post- or both. Type 2 diabetes is the most 20% to 50%.2 Current American Dia- prandial glucose (− 31.5 mg/dL, common form of diabetes in the betes Association (ADA) guidelines P = .026) and LDL-C (− 11.7%, United States, accounting for as many recommend an A1C of < 7% and an P = .007). The study authors con- as 90% to 95% of all cases.1 As of LDL-C < 100 mg/dL in patients with cluded that colesevelam may improve 2011, 25.8 million patients were di- diabetes.4 However, to achieve these both glycemic and lipid control in agnosed with diabetes, or 8.3% of the goals, patients with diabetes often patients with type 2 diabetes.5 Fol- U.S. population.2 Type 2 diabetes is need to use multiple .1,2 lowing the publication of this trial, more common in African Americans, As a result, recent attention has colesevelam received an FDA indica- Mexican Americans, Native Ameri- shifted to the use of single agents tion as an adjunctive treatment in pa- cans, Asian Americans, Native Ha- with both glucose and lipid-lowering tients with type 2 diabetes, in addition waiians, and other Pacific Islanders, effects. One such class of agents is the to its prior indication for hypercholes- as well as the elderly population.1,3 sequestrants. To date, both terolemia. Type 2 diabetes is characterized by colesevelam hydrochloride and cho- Another similar trial evaluated the insulin resistance and a relative lack lestyramine have demonstrated A1C use of cholestyramine for dyslipid- of insulin secretion. Insulin is neces- and glucose-lowering effects in clini- emia in patients with type 2 diabetes. sary for the breakdown of sugars and cal trials. The study researchers also examined carbohydrates in the body and plays The Glucose Lowering Effect the effect of cholestyramine on glu- a vital role in the uptake of glucose of Welchol Study (GLOWS) evalu- cose levels as a secondary endpoint. into cells for energy.1 ated the A1C lowering effect of co- This study was a randomized, dou- Uncontrolled diabetes is associ- lesevelam hydrochloride in subjects ble-blind, crossover study of 21 par- ated with numerous complications, with type 2 diabetes who were inad- ticipants receiving cholestyramine including blindness, kidney damage, equately controlled by oral antihyper- 8 g bid vs a placebo for 6 weeks heart disease, and lower-limb am- glycemic agents alone. Participants each. Study participants had type 2 putations. In patients with diabetes, were on a stable dose of metformin diabetes that was well controlled for optimal control of glucose and low- or a sulfonylurea for ≥ 90 days before at least 1 month prior to the initia- density lipoprotein (LDL- the initiation of colesevelam. Sixty- tion of cholestyramine, using either C) has been shown to delay or even five study participants who had an insulin or glyburide therapy, and an prevent the development of compli- A1C of 7% to 10% were random- LDL-C of > 130 mg/dL. Addition- ized to receive either colesevelam ally, patients were on a stable dose of 3.75 g/d or placebo for 12 weeks. insulin or glyburide throughout the Dr. Sauter, Dr. Abraham, Dr. John, Dr. Kapadia, After 12 weeks, results demon- duration of the study. After 12 weeks, and Dr. Toth are all clinical pharmacy specialists in the Department of Pharmacy with the Jesse strated a difference in mean change results demonstrated that treatment Brown VA Medical Center in Chicago, Illinois. in A1C between the colesevelam with cholestyramine reduced total

MAY 2012 • FEDERAL PRACTITIONER • 23 GLYCEMIC CONTROL

Table 1. Data collected using the electronic chart

Demographic Age information Gender Race Baseline information A1C (within the 6 months Lipids (LDL-C, TGs, HDL-C) prior to initiation of LFTs (within normal limits [WNL] vs elevated) colestipol) After initiation of A1C (at least 3 months after starting colestipol) colestipol Lipids (LDL-C, TGs, HDL-C) (at least 4 weeks after starting colestipol) LFTs (WNL vs elevated) Any subsequent values for the above parameters following dose titration of colestipol for up to 6 months Adverse effects intolerance as documented in the progress notes Medication Colestipol information • Initial dose • Maximum tolerated dose • Date of initiation • Date of discontinuation (if applicable) Concomitant oral antihyperglycemic medications (including sulfonylureas, biguanides, insulin, thiazolidinediones, α-glucosidase inhibitors, meglitinides, dipeptidyl pepti- dase-4 inhibitors, glucagon-like peptide-1 agonists, and dopamine agonists) • Dose when colestipol was started • Dosage changes since colestipol was started • Date of initiation • Date of discontinuation (if applicable) Concomitant lipid-lowering medications (including HMG-CoA reductase inhibitors, nicotinic acid, intestinal absorption inhibitors, fish oil, and fibric acid derivatives) • Dose when colestipol was started • Dosage changes since colestipol was started • Date of initiation • Date of discontinuation (if applicable) Medication Determined by the refill history in the electronic medical record and documentation compliance of nonadherence in the progress notes Patient education Counseling regarding the proper administration of colestipol in relation to other medications to minimize any possible interference with absorption (take other medications at least 1 hour before or 4 hours after colestipol) as documented in the progress notes or stated on the medication label cholesterol by 18% and LDL-C by that cholestyramine therapy effec- monotherapy, which does not com- 28%, when compared with placebo. tively reduced LDL-C and may also ply with the current clinical approach In addition, cholestyramine therapy improve glycemic control in patients in the United States of using bile acid improved glycemic control. Mean with diabetes.6 sequestrants as adjunctive therapy in plasma glucose levels decreased by It is important to note that there patients with diabetes. In other coun- 13%; a mean reduction in urinary glu- is 1 additional agent in this class of tries, has been proven effec- cose of 0.22 g/d was noted, medications: Colestilan. However, this tive in reducing both A1C and fasting as well as a decrease in A1C values of agent is not approved in the United plasma glucose in patients with type 2 0.5%. The study authors concluded States and has been evaluated only as diabetes.7

Continued on page 27

24 • FEDERAL PRACTITIONER • MAY 2012 GLYCEMIC CONTROL

Continued from page 24

Patient charts reviewed (239)

Excluded Included (189) (50)

Non- < 3 months No type 2 Antihyperglycemic No A1C adherence therapy diabetes changes (42) (8) (13) (30) (96)

Figure 1. Study design.

In general, bile acid sequestrants tory action increases LDL-C clearance demonstrated beneficial effects on act by binding bile acids, forming a from the , resulting in lowered both LDL-C and A1C in patients with complex that is then excreted in the LDL-C levels.8,9 type 2 diabetes, the specific effect of feces, thus preventing reabsorption The mechanism by which bile acid colestipol on glycemic control has yet in the intestines. This action also re- sequestrants exert their glycemic ef- to be evaluated. Currently, colestipol sults in partial removal of bile acids fects is not fully understood. One is FDA approved for the treatment of from the enterohepatic circulation, proposed mechanism is that bile acid primary , and which prevents its reabsorption. This sequestrants cause a reduction in glu- it is the that is leads to the depletion of serum bile cose absorption or a change in the currently on the formulary at Jesse acids, which activates cholesterol time course of glucose absorption in Brown VAMC. 7 alpha-hydroxylase to convert cho- the . Another the- lesterol into bile acids. Consequently, ory is that by interrupting the entero- METHODS a demand for cholesterol in the hepatic pathway of bile metabolism, The purpose of this study was to is increased, resulting in increased bile acid sequestrants deactivate the evaluate the effect of colestipol on 3-hydroxy-3-methyl-glutaryl-CoA farnesoid X receptor. This receptor is glycemic control in patients with (HMG-CoA) reductase transcription a bile acid-activated nuclear receptor type 2 diabetes in a veteran popula- and activity, and an increased num- that plays a significant role in the me- tion. This study was an Institutional ber of LDL-C receptors. HMG-CoA tabolism of bile acids, cholesterol, and Review Board and VA Research and reductase is the rate-controlling en- glucose.8,9 Development Committee approved, zyme in the metabolic pathway that While previous trials with both co- retrospective, electronic chart re- produces cholesterol. This compensa- lesevelam and cholestyramine have view of patients with an ICD-9 di-

Table 2. Demographic information Gender n (%) Age (y) Race n (%) Male Female Mean ± SD African American White Pacific Islander Unknown 50 (100) 0 (0) 70.9 ± 8.1 26 (52) 15 (30) 2 (4) 7 (14)

MAY 2012 • FEDERAL PRACTITIONER • 27 GLYCEMIC CONTROL

70 paired t test and Wilcoxon signed (n = 30) rank test, depending on the data dis- 60 tribution. Data analyzed included change in A1C (Wilcoxon), lipid pa- 50 rameters (t test), LFTs (none), the oc- currence of counseling and adverse 40 effects (none), and dose-dependent changes in A1C (Wilcoxon). 30 (n = 10) RESULTS 20 Based on the aforementioned crite- (n = 5) ria, a total of 239 patient charts were

Patient population (%) population Patient 10 (n = 3) (n = 1) (n = 1) reviewed, and 50 patients were ulti- 0 mately included in the study (Figure 5.5 6.5 7.5 8.5 9.5 10.5 1). Overall, 100% of the patients were male, with a mean age of 70.9 years. A1C (%) Additionally, the majority of the pa- tients were African American (52%), Figure 2. Baseline A1C. followed by white (30%) (Table 2). The majority of patients had a baseline A1C between 6% and 7% agnosis of type 2 diabetes and with served as his or her own control. (Figure 2 ), which explains the lack of an active prescription for colestipol The primary efficacy endpoint was changes in concomitant antihypergly- anytime between January 1, 2005 a change in A1C from baseline to fol- cemic medications during the study and June 15, 2010. Patients aged low-up after the initiation of colesti- period. On average, patients receiv- ≥ 18 years with a diagnosis of type pol. Secondary endpoints included ing colestipol experienced a statis- 2 diabetes and with a prescription the percent change in lipid parame- tically significant reduction in A1C for colestipol were included in the ters and percentage of patients expe- of 0.24% (P < .0001). When divid- study. Study participants who were riencing an increase in liver function ing patients into different categories not receiving treatment with colesti- tests (LFTs) from baseline to follow- based on their A1C level at baseline pol for a minimum of 12 weeks were up after initiation of colestipol. Ad- (< 7%, 7%-8%, or > 8%), there was excluded from the study. Addition- ditional secondary endpoints included a trend toward a greater reduction in ally, patients with any changes in the documentation of appropriate A1C in those patients with a higher their antihyperglycemic medications counseling regarding the proper ad- initial A1C (0.16%, P = .001; 0.34%, during the 3-month period before or ministration of the medication, as well P = .01; and 0.56%, P = .31, respec- after the initiation of colestipol and as the occurrence of adverse events tively). However, due to the small pa- patients that lacked an A1C within related to the study medication. tient population in the > 8% baseline the 6 months prior to or following the The data collected, using the elec- category (n = 7), there may not have initiation of colestipol were excluded. tronic chart, are shown in Table 1. been enough power to determine Patients were followed throughout Statistical analysis of the collected statistical significance. Additionally, the study period, and each subject data was performed using both a there was not a statistically signifi- Table 3. A1C reduction Parameter Baseline (%) Final (%) Change (%) P value Average overall A1C (%) 6.9 6.7 − 0.24 < .0001 < 7 6.5 6.3 − 0.16 = .001 7-8 7.4 7.1 − 0.34 = .01 > 8 9.0 8.5 − 0.56 = .31

28 • FEDERAL PRACTITIONER • MAY 2012 GLYCEMIC CONTROL

cant difference in the change in A1C Table 4. Maximum dose between the 3 baseline categories Average change in (P = .38) (Table 3). Colestipol dose A1C (%) Patients (n) P value The average maximum tolerated 1 g bid − 0.2 7 = .15 dose of colestipol was 4.5 g/d. When comparing the average A1C reduc- 2 g bid − 0.2 32 = .01 tion with the maximum tolerated 3 g bid − 0.3 5 = .25 dose of colestipol (1 g bid, 2 g bid, 4 g bid − 0.3 6 = .06 3 g bid, and 4 g bid), there was a trend toward a greater reduction in A1C Table 5. Lipid parameters with a higher colestipol dose (0.2%, P = .15; 0.2%, P = .01; 0.3%, P = .25; Parameter Baseline Final Change (%) P value and 0.3%, P = .06, respectively). How- LDL-C (mg/dL) 130.8 112.3 − 13.4 < .0001 ever, similar to the previous discus- sion, there may not have been enough HDL-C (mg/dL) 40.5 40.4 0.6 = .89 patients in the various dosage groups TGs (mg/dL) 153.8 170.4 18.9 = .26 to determine statistical significance. Additionally, there was not a statisti- Table 6. LFT (AST/ALT) changes cally significant difference in the change in A1C levels between the 4 Classification Baseline (n) Final (n) Change (%) dosage categories (P = .60) (Table 4). WNL 42 42 0 All the study patients failed to Elevated 8 8 0 achieve therapeutic lipid goals with AST/ALT: aspartate aminotransferase/alanine aminotransferase. optimal doses of or had docu- mented intolerance(s) to other avail- use or hepatitis (Table 6). Education agents used in the previously men- able agents, and were, therefore, regarding the proper administra- tioned studies were higher in com- initiated on colestipol for additional tion of colestipol, in relation to other parison with the average colestipol LDL-C lowering. Patients receiv- medications, was documented in 13 dose observed in this study. A trend ing colestipol experienced a statisti- of 50 patients (26%). Finally, gas- toward a greater reduction in A1C cally significant reduction in LDL-C trointestinal-related adverse events, with higher initial A1C levels at base- of 13.4% (P < .0001). Following which are commonly associated with line was observed; however, due to initiation of colestipol, patients ex- the administration of colestipol, were the small patient population, these re- perienced an increase in reported by 2 of 50 patients (4%) fol- sults did not achieve statistical signifi- (TGs) of 18.9% (P = .26) and overall, lowing dose titration of colestipol. cance. Additionally, a trend toward a high-density lipoprotein cholesterol In both instances, the patients ex- greater reduction in A1C with higher (HDL-C) remained relatively un- perienced , the dose of colestipol doses was noted. Similarly, changed, increasing slightly by 0.6% colestipol was reduced to the previ- the sample size may have limited the (P = .89) (Table 5). Of note, 4 of the ously tolerated dose, and the patients ability to determine statistical signifi- 50 patients (8%) had concomitant were able to continue therapy. cance. Of note, the maximum recom- changes to their other lipid-lowering mended dose of colestipol is 16 g/d in medications during colestipol therapy DISCUSSION divided doses; the average maximum (medication initiation [n = 2], medi- On average, patients receiving ther- tolerated dose among patients in this cation conversion [n = 1], and medi- apy with colestipol for additional study was 4.5 g/d. cation discontinuation [n = 1]). LDL-C lowering experienced an Patients receiving therapy with In terms of the additional second- overall reduction in A1C of 0.24%. colestipol experienced a reduction in ary endpoints, LFTs remained stable This value is slightly lower than the LDL-C of 13.4%, an increase in TGs of following initiation of colestipol and reductions observed with other agents 18.9%, and a slight increase in HDL-C throughout the study period. Eight of this class: 0.3% to 0.7% with co- of 0.6%. These results are fairly con- of 50 patients (16%) had persistently lesevelam and 0.5% with cholestyr- sistent with the documented thera- elevated LFTs due to either alcohol amine. However, the doses of those peutic effects of colestipol showing

MAY 2012 • FEDERAL PRACTITIONER • 29 GLYCEMIC CONTROL

a decrease in LDL-C of 15% to 30%, There are several limitations to this tial conflicts of interest with regard to an increase in HDL-C of 3% to 5%, study, including its retrospective de- this article. and either no change or an overall in- sign. Based on the study’s inclusion crease in TGs. Overall, study patients and exclusion criteria, most notably, Disclaimer tolerated therapy with colestipol fairly the occurrence of any change in an- The opinions expressed herein are those well. Two patients reported constipa- tihyperglycemic medications during of the authors and do not necessarily tion following dose titration of colesti- the 3-month period before or after reflect those of Federal Practitioner, pol; however, this was resolved with a the initiation of colestipol, a small Quadrant HealthCom Inc., the U.S. reduction to the previously tolerated number of patients were ultimately Government, or any of its agencies. dose. Additionally, no patients expe- included in the study. This small This article may discuss unlabeled or rienced an increase in LFTs following study population may have limited investigational use of certain . initiation of colestipol. Eight patients the ability to decisively determine sta- Please review complete prescribing in- had persistently elevated LFTs, due tistical significance for a number of formation for specific drugs or to alcohol use or hepatitis, which re- the study endpoints. External valid- combinations—including indications, mained stable throughout therapy ity of the study results is also limited, contraindications, warnings, and ad- with colestipol. Education regarding given the entirely male and elderly verse effects—before administering the proper administration of colesti- study population. Additionally, a lack pharmacologic therapy to patients. pol, in relation to other medications, of documentation regarding patient REFERENCES was documented in only 26% of pa- compliance, outside medications, and 1. American Diabetes Association. Diagnosis and Clas- tients. This is an important counseling concurrent diet and exercise programs sification of Diabetes Mellitus. Diabetes Care. 2010; 33(suppl 1):S62-S69. point in order to minimize any poten- may influence the study outcomes. 2. Centers for Disease Control and Prevention. Na- tial interference with the absorption of To minimize these limitations, larger tional diabetes fact sheet: national estimates and gen- eral information on diabetes and prediabetes in the other essential medications. prospective studies, with a longer ob- United States, 2011. U.S. Department of Health and Studies have demonstrated that servation period, are needed to fully Human Services, Centers for Disease Control and Prevention, 2011. pharmacist involvement in patient evaluate the effect of colestipol on gly- 3. Cowie CC, Rust KF, Byrd-Holt DD, et al. Prevalence care can lead to beneficial outcomes. cemic control. of diabetes and impaired fasting glucose in adults in the U.S. population: National Health and Nutri- One such study evaluated 40 men tion Examination Survey 1999-2002. Diabetes Care. with hypercholesterolemia who were CONCLUSION 2006;29(6):1263-1268. 4. American Diabetes Association. Standards of Medi- randomized to receive either pharma- Overall, therapy with colestipol for ad- cal Care in Diabetes. Diabetes Care. 2012;35(suppl ceutical care (pharmacist-physician ditional LDL-C lowering in patients 1):S11-S63. 5. Zieve FJ, Kalin MF, Schwartz SL, Jones MR, Bai- comanagement) or standard care by a with type 2 diabetes resulted in an av- ley WL. Results of the glucose-lowering effect of physician alone. Patients in the phar- erage overall A1C reduction of 0.24% Welchol study (GLOWS): A randomized, double- < blind, placebo-controlled pilot study evaluating the maceutical care group received edu- (P .0001). The results of this study effect of colesevelam hydrochloride on glycemic cation from a pharmacist regarding contribute to the growing evidence control in subjects with type 2 diabetes. Clin Ther. 2007;29(1):74-83. hypercholesterolemia, colestipol-dose supporting the use of bile acid seques- 6. Garg A, Grundy SM. Cholestyramine therapy for titration, and adverse-effects man- trants as adjunctive therapy in patients in non-insulin dependent diabetes mel- litus: A short term, double-blind, crossover trial. Ann agement. After 52 weeks of colesti- with type 2 diabetes. Although agents Intern Med.1994;121(6):416-422. pol therapy, the pharmaceutical care with more evidence of improved gly- 7. Kondo K, Kadowaki T. Colestilan monotherapy significantly improves glycaemic control and LDL group achieved greater LDL-C reduc- cemic control should be used first, as cholesterol levels in patients with type 2 diabetes: A tions than the standard group (16% recommended by the ADA, provid- randomized double-blind placebo-controlled study. Diabetes Obes Metab. 2010;12(3):246-251. vs 9.4%), and more patients in the ers may consider the use of colestipol 8. Brinton EA. Novel pathways for glycaemic control pharmaceutical care group achieved in patients with type 2 diabetes who in type 2 diabetes: Focus on bile acid modulation. Diabetes Obes Metab. 2008;10(11):1004-1011. their LDL-C goal (29.4% vs 5.0%, need additional LDL-C lowering de- 9. Staels B. A review of bile acid sequestrants: Potential P < .5).10 Therefore, this may represent spite optimal doses of statins or intol- mechanism(s) for glucose-lowering effects in type 2 l diabetes mellitus. Postgrad Med. 2009;121(3)(suppl an opportunity for pharmacists to fill a erability of other agents. 1):25-30. vital role to ensure that the appropri- 10. Konzem SL, Gray DR, Kashyap ML. Effect of phar- maceutical care on optimum colestipol treatment in ate counseling is provided and docu- Author disclosures elderly hypercholesterolemic veterans. Pharmaco- mented. The authors report no actual or poten- therapy. 1997;17(3):576-583.

30 • FEDERAL PRACTITIONER • MAY 2012