Search tips
Search criteria 


Logo of nihpaAbout Author manuscriptsSubmit a manuscriptHHS Public Access; Author Manuscript; Accepted for publication in peer reviewed journal;
Jt Comm J Qual Patient Saf. Author manuscript; available in PMC 2013 September 1.
Published in final edited form as:
PMCID: PMC3534988

Discontinuation of Antihyperglycemic Therapy After AMI: Medical Necessity or Medical Error?

Kay O Lovig, MD, Clinical Instructor of Medicine, Leora Horwitz, MD, Assistant Professor of Medicine, Kasia Lipska, MD, Endocrinology Fellow, Mikhail Kosiborod, MD., Associate Professor, Harlan M Krumholz, MD, Professor of Medicine, and Silvio E Inzucchi, MD, Professor of Medicine


From a national Medicare database, we recently reported that 1 in 8 older patients with diabetes (DM) was discharged (DC) off all anti-hyperglycemic medication (AHM) following acute myocardial infarction (AMI). This practice was associated with increased 1-yr mortality, but the reasons for stopping AHM were not known. We sought to determine whether such practice might be due to medical necessity (i.e., new contraindication) or oversight – in which case a quality improvement opportunity might exist.

We identified 327 DM patients hospitalized with AMI over a 1-yr period at an academic medical center and an affiliated community hospital. We conducted detailed chart reviews on the 217 patients with AMI as a principal diagnosis who were admitted on AHM (insulin, 81). 25 (11.5%) were DC off AHM, 24 (96%) of whom received some AHM in the hospital, mostly insulin sliding scale. One patient’s (4%) AHM was stopped due to a change in care goals, a second developed recurrent hypoglycemia, and a third had entirely normal in-hospital blood glucose (BG) after AHM discontinuation. We categorized the remaining 88% as being DC off AHM without justification. The demographic/clinical characteristics of those DC on vs. off AHM were similar, except for better left ventricular ejection fraction (LVEF) in the latter.

In this study, the proportion of DM patients DC off AHM following AMI was nearly identical to that in a national database (~1/8). We found no clear reason for this practice in nearly 90% of cases, suggesting that it may often constitute a medical error. Based on our prior finding of associated increased mortality, this may represent an opportunity to improve the care of the growing population of DM patients with ischemic heart disease.

Accurate medication reconciliation has become a growing area of concern as multiple studies have demonstrated the frequency of medication discrepancies between admission and discharge,14 which may contribute to adverse clinical outcomes.5,6 Therefore, the Joint Commission designated medication reconciliation as a “National Patient Safety Goal” in 2005.79 The most common type of medication discrepancy identified in multiple studies is the omission of medications at discharge.1,2 In particular, one study found that medications associated with endocrine conditions are amongst the most common to be omitted at discharge.1

From a large, national database, we recently reported that 1 out of 8 older diabetic patients following acute myocardial infarction (AMI) was discharged off all anti-hyperglycemic therapy (AHT), a practice that was associated with increased 1-year mortality.10 However, due to data limitations, details regarding the reason(s) for discontinuation of AHT were not available. Therefore, whether AHT discontinuation is medically justified, or whether it is simply due to clinicians’ oversight, remains unclear.

Most studies evaluating long term intensive glycemic control in diabetic patients have shown no significant overall reduction in macrovascular events. 1113 Trials examining the benefits of acute intensive blood glucose control during hospitalization for AMI have shown disparate results.14,15 Few data exist, however, on the importance of glucose management soon after discharge for acute myocardial infarction – a possibility that was raised by our findings.

In this context, we sought to determine whether the omission of anti-hyperglycemic therapy upon discharge following AMI was related to new contraindications that had emerged during the course of the hospitalization or whether this simply reflected an inadvertent omission of an important treatment program. In the case of the latter scenario, an important quality improvement opportunity might exist.

Research design and methods

Data Source

IRB approval was obtained for a retrospective chart review, while the need for informed consent was waived, given the nature of the investigation. We conducted a detailed chart review of patients with diabetes admitted with AMI over a 1-year period, January 2008 to December 2008, at an academic medical center and an affiliated community hospital. Inclusion criteria were: established diagnosis of diabetes as documented in the record, the use of at least one anti-hyperglycemic drug on admission, a primary discharge diagnosis of AMI which was the cause of the admission, and discharge to home or a chronic care facility.

Study Design

We examined the frequency of discontinuation of all anti-hyperglycemic agents upon discharge among those patients who had been admitted on a pharmacological diabetic regimen. Among those discharged off all AHT, one investigator (K.L.) proceeded to conduct a detailed chart review to identify the reasons for its discontinuation. Selected findings were confirmed by a second investigator (S.I.). More specifically, we reviewed whether a reason for discontinuing AHT was documented, or whether there was evidence of a clinical reason to stop the medications. Plausible explanations included a hypoglycemic event either on admission or during the course of the hospital stay, acceptable blood glucose control without any AHT, a new contraindication for resuming AHT (e.g., renal failure, heart failure), and/or a change in the overall goals of care. We also evaluated the rate of continuation of other standard medication after AMI including aspirin, beta blocker and HMG-COA reductase inhibitors.

Statistical Methods

We examined whether patients discharged of AHT differed from those discharged on AHT in terms of age, sex, race, mean blood glucose in hospital, any blood glucose <70 or ≥ 200, left ventricular ejection fraction, chronic renal insufficiency, mean hemoglobin (Hb) A1c, ST elevation MI, serum creatinine ≥ 1.5 at discharge, discharged without beta blocker, aspirin and HMG-COA reductase inhibitor, and discharge disposition. We used the chi square test, Fisher’s exact test or Student’s t-test, as appropriate. All statistical tests were two-tailed and we considered a p-value ≤ 0.05 to be statistically significant. Analyses were conducted using SAS 9.2 (SAS Institute, Cary, NC).


A total of 327 patient with diabetes were admitted with AMI over a 1-year period; 297 at the academic medical center and 30 at the affiliated community hospital (n=30). Of these we excluded 110 patient for the following reasons: (1) no documentation of AHT prior to admission (n=37), (2) no documented history of DM prior to admission (18), (3) chart not available at time of review (18), (4) transfer to another facility (16), (5) AMI was not the primary diagnosis (13), and (6) death during hospitalization (8). Of the 217 meeting all inclusion criteria, 97 (45%) were being treated with metformin, 86 (40%) with insulin secretagogues (mostly sulfonylureas), 81 (37%) with insulin, 30 (14%) with thiazolidinediones, and 14 (6%) with other anti-hyperglycemic medications (drug categories not mutually exclusive due to the use of combination regimens). Of the 217 included patients, 25 (11.5%) were discharged off AHT, 24 (96%) of whom received some AHT in the hospital. Of these, 16 (64%) were treated as inpatients with regular insulin sliding scale (RISS), 7 (28%) with a basal insulin plus RISS, 1 (4%) with basal insulin only, 5 (20%) with an oral anti-hyperglycemic agent. Prior to admission, these 25 patients were treated with metformin in (n=10), insulin (n=10), sulfonylurea in (n=6), thiazolidinediones (n=4), and other medicines (n=3). Of these 25 patients 19 (76%) were treated with one anti-hyperglycemic agent, 4 (16%) with 2 agents, and 2 (8%) with 3 agents.

Only one patient (4%) had a documented explanation (change in goals of care) and 2 other (8%) were deemed to have a justifiable explanation for discontinuation of AHT at discharge after the detailed chart review. One of these individuals developed recurrent hypoglycemia (capillary blood glucose levels by point-of-care meter of 43, 50, 56, and 59 mg/dl) while hospitalized both on and off of an oral hypoglycemic agent (sulfonylurea). A second individual had entirely normal in-hospital blood glucose levels after AHT was discontinued, with an average morning serum glucose of 103 mg/dl and a range of 85–128 mg/dl. We categorized the remaining 88% as being discharged off AHT without justification.

The demographic and clinical characteristics of those discharged on versus off AHM were similar, as shown in the Table. The sole exception was better left ventricular ejection fraction (LVEF) in the latter group. More specifically, there was no statistical difference in the rate of hypoglycemic or hyperglycemic episodes, nor in the serum creatinine concentration at discharge. Additionally, there were no statistical differences in glycemic measures, such as mean laboratory plasma glucose level during hospitalization, as well as Hb-A1c obtained over the past six months, when it was available. For example, among the 218 charts reviewed, 90 (41%) had Hb-A1c obtained in the past six months available for review. Of these, 35 patients had a Hb-A1c <7%, 7 (20%) of whom were discharged off of AHM, whereas 55 patients had a Hb-A1c > 7%, of whom 5 (9%) were discharged off of AHM, (P=0.20).

We also examined the percentage of patients discharged on other standard cardiovascular therapies following AMI such as aspirin, beta blocker, and hmg coa reductase inhibitor. The reasons for omission of these agents were not addressed by our investigation.

Finally, 40% of the 25 patients discharged off AHT, had follow up office or clinic visit dates documented on their discharge summaries, which was not statistically significant from those discharged on AHT (30%)(p=0.32).


In this two-site study, the proportion of diabetic patients discharged off all AHT following AMI was nearly identical to our previous report using a large, national database, at 11.5%.10 Surprisingly, we could find no documentation of the reason for medication discontinuation in 96% of medical records. Further exploration failed to reveal an apparent reason in 84%. These data suggest that such practice is likely due to clinicians’ oversight, rather than valid medical reasoning. Based on our previous finding of increased mortality in this group,10 an opportunity to improve the care of the growing population of patients with diabetes and ischemic heart disease might exist.

Frequently, anti-hyperglycemic agents (particularly oral agents) are appropriately held during hospitalizations for a variety of reasons.16 The American Diabetes Association (ADA) recommends scheduled subcutaneous insulin with basal, nutritional, and correction components as the preferred method for achieving and maintaining glucose control in noncritically ill patients. 17 For critically ill patients admitted to the hospital, the ADA recommends that hyperglycemia greater than 180 mg/dl be treated with intravenous insulin with a blood glucose goal of 140–180 mg/dl. The ADA has also stated that there is no clear evidence for specific blood glucose goals in non-critically ill patients, however subcutaneous insulin is the preferred method for achieving and maintaining glucose control in this patient population, with a pre-meal blood glucose target of < 140 mg/dl. 17 Therefore, the majority of patient with hyperglycemia during the in-hospital stay are treated with insulin therapy. In our cohort the failure to resume AHT occurred just as frequently in patients using insulin before admission (40%) compared to patients using oral AHT (60%), suggesting that the conversion to insulin therapy during hospitalization does not fully explain this practice. In addition there was no difference in the number of patients on vs. off AHT who received insulin during hospitalization, 180 (94%) vs. 24(96%) respectively (p=1.0). There were small but similar percentages of patient in both groups who were discharged without standard cardiovascular therapies (aspirin, beta-blockers, HMG-COA reductase inhibitor) following AMI, or without documented follow-up visits, suggesting that the failure to resume AHT was not merely reflective of overall substandard discharge planning.

Of course, it is possible that health care providers assume that patients will continue their diabetic regimens after discharge, or that AHT will be addressed during outpatient follow-up. It may also be that providers simply feel uncomfortable addressing therapeutic agents for a chronic condition, such as diabetes, at discharge. Nevertheless it is clear that failure to resume any AHT regimen at discharge in these individuals does not appear to be indicated in the vast majority of cases and the failure to record any intended follow-up of AHT in the medical records represents, at minimum, less than optimal documentation. One study assessing medication reconciliation at admission and discharge of general medicine, cardiology and general surgery patients found that patients were explicitly alerted on their discharge instructions to changes made in their medications at discharge only 44% of the time.18

Although in randomized clinical trials in patients with diabetes at high cardiovascular risk, intensive blood glucose control appeared to have no major impact on the incidence of cardiovascular events,11,12 we know less about the acute effects of hyperglycemia soon after MI. Several studies have investigated the role of management of hyperglycemia in diabetic patient during acute coronary syndrome with variable results. 1921 One study examining this point, DIGAMI, found major reductions in mortality in diabetic patients with AMI whose blood glucose was managed more intensively during the hospital with intravenous insulin and during the 3 post-discharge months with multiple subcutaneous insulin injections.19 Another trial, HI-5, found no difference in all cause mortality during hospitalization, and at 3 and 6 months, between diabetic patients treated aggressively with insulin infusion versus those in whom hyperglycemia was managed permissively.21 A third study, which was observational, suggested that normalization of glucose values in patients with acute MI after admission is associated with better survival. 22 However the literature demonstrating that hyperglycemia during or soon after admission for acute MI is associated with mortality is far from extensive. 15,23, 24

Our study has several limitations. Being a retrospective chart review, the data collected is based solely on the accuracy of documentation. In addition relying on chart review to determine which omissions may have been deliberate may not be accurate. That is, it is certainly possible that additional discharge instructions were provide to the patient regarding their AHT that were not recorded in the medical record. Also, we had no access to outpatient notes and therefore could not determine how quickly (if at all) AHT was resumed upon discharge. In addition, conducting this retrospective chart review, we did not contact the patient’s discharging physician in attempt to attain insight as to why discontinuation may have occurred. Finally, for similar reasons, we were not able to track the patients’ glycemic control in the weeks to months after discharge. Accordingly, no causative link between adverse outcomes in patients discharged off AHT and uncontrolled hyperglycemia can be made based on our data.

In conclusion, we confirmed our findings from a prior investigation that approximately 1 out of 8 diabetic patients hospitalized for AMI is discharged off all AHT. Of these, almost all had no documented explanation for this practice, with the vast majority having no obvious justification upon further examination of the medical record. At the very least, our findings call for more accurate medication reconciliation in patients with diabetes following hospitalizations for AMI. Despite a paucity of supporting data from randomized clinical trials10,1921, given our previous findings of increased mortality in such individuals and the reasonable biological basis of such an association, our data suggests a potential opportunity for improvement in the quality of care provided to hospitalized patients with diabetes and cardiovascular disease.

Table 1
Characteristics of diabetic patients with AMI discharged on vs. off AHT.



Dr. Horwitz is supported by the National Institute on Aging (K08 AG038336) and by the American Federation for Aging Research t2hrough the Paul B. Beeson Career Development Award Program.

Dr. Horwitz is also a Pepper Scholar with support from the Claude D. Pepper Older Americans Independence Center at Yale University School of Medicine (#P30AG021342 NIH/NIA).

Dr. Kosiborod is supported by grant by the American Heart Association and Medtronic Diabetes Dr. Kosiborod is a consultant for Medtronic Diabetes, Sanofi-Aventis, Boehringer-Ingelheim, Gilead, Genentech, and Kowa Pharmaceuticals.

Dr. Krumholz is supported by grant U01-HL105270-02 (Center for Cardiovascular Outcomes Research at Yale University) from the National Heart, Lung, and Blood Institute.

Dr. Krumholz discloses that he chairs a cardiac scientific advisory board for UnitedHealth and is the recipient of a research grant from Medtronic, Inc. through Yale University.

Dr. Inzucchi is consultant/advisor for Merck, Takeda, Boeringher-Ingelheim, and has research funding from Eli Lilly Co.

Contributor Information

Kay O Lovig, Division of General Internal Medicine, University of Pittsburgh Medical Center.

Leora Horwitz, Division of General Internal Medicine, Yale University School of Medicine, Yale-New Haven Hospital.

Kasia Lipska, Yale University School of Medicine, Yale-New Haven Hospital.

Mikhail Kosiborod, Saint-Luke’s Mid America Heart Institute, University of Missouri-Kansas City.

Harlan M Krumholz, Section of Cardiovascular Medicine and Robert Wood Johnson Clinical Scholars Program, Department of Internal Medicine, Yale University School of Medicine, Center for Outcomes Research and Evaluation, Section of Health Policy and Administration, Yale School of Public Health.

Silvio E Inzucchi, Yale University School of Medicine, Director, Yale Diabetes Center, Yale-New Haven Hospital.


1. Herrero-Herrero JI, Garcia-Aparicolo J. Medication discrepancies at discharge from aninternal medicine service. Eur J Intern Med. 2011;22(1):43–48. [PubMed]
2. Grimes TC, et al. Medication details documented on hospital discharge: cross-sectional observational study of factors associated with medication non-reconciliation. Br J Clin Pharmacol. 2011;71(3):449–457. [PMC free article] [PubMed]
3. Pippins JR, et al. Classifying and predicting errors of inpatient medication reconciliation. J Gen Intern Med. 2008;23(9):1414–1422. [PMC free article] [PubMed]
4. Tam VC, et al. Frequency, type and clinical importance of medication history errors at admission to hospital: a systematic review. CMAJ. 2005;173(5):510–515. [PMC free article] [PubMed]
5. Coleman EA, et al. Post-hospital medication discrepancies: prevalence and contributing factors. Arch Intern Med. 2005;165(16):1842–1847. [PubMed]
6. Boockvar KS, et al. Medication reconciliation for reducing drug-discrepancy adverse events. Am J Geriatr Pharmacother. 2006;4(3):236–243. [PubMed]
7. Joint Commission on Accreditation of Healthcare Organizations. [Accessed April 18, 2008.];Critical access hospital and hospital national patient safety goals. 2006
8. Cornish PL, et al. Unintended medication discrepancies at the time of hospital admission. Arch Intern Med. 2005;165(4):424–429. [PubMed]
9. Schnipper JL, et al. Role of pharmacist counseling in preventing adverse drug events after hospitalization. Arch Intern Med. 2006;166(5):565–571. [PubMed]
10. Lipska KJ, et al. Discontinuation of antihyperglycemic therapy and clinical outcomes after acute myocardial infarction in older patients with diabetes. Circulation: Cardiovascular Quality and Outcomes. 2010;3(3):236–242. [PubMed]
11. Gerstein HC, Miller, et al. Effects of intensive glucose lowering in type 2 diabetes. N Engl J Med. 2008;358:2545–2559. [PMC free article] [PubMed]
12. Patel A, et al. Intensive blood glucose control and vascular outcomes in patients with type 2 diabetes. N Engl J Med. 2008;358:2560 –2572. [PubMed]
13. Duckworth W, et al. Glucose control and vascular complications in veterans with type 2 diabetes. N Engl J Med. 2009;360:129–139. [PubMed]
14. Stranders I, et al. Admission blood glucose level as risk indicator of death after myocardial infarction in patients with and without diabetes mellitus. Arch Intern Med. 2004;164:982–988. [PubMed]
15. Kosiborod M, et al. Admission glucose and mortality in elderly patients hospitalized with acute myocardial infarction: implications for patients with and without recognized diabetes. Circulation. 2005;111:3078–3086. [PubMed]
16. Moghissi ES, et al. American Association of Clinical Endocrinologists and American Diabetes Association consensus statement on inpatient glycemic control. Diabetes Care. 2009;32:1119–1131. [PMC free article] [PubMed]
17. Standard of medical care in diabetes - 2011. America Diabetes Association. Diabetes Care. 2011;34(1):S11–S61. [PMC free article] [PubMed]
18. Unroe KT, et al. Inpatient medication reconciliation at admission and discharge: A retrospective cohort study of age and other risk factors or medication discrepancies. Am J Geriatric Pharmacothery. 2010;8(2):115–126. [PMC free article] [PubMed]
19. Mellbin LG, et al. Prognostic implications of glucose-lowering treatment in patients with acute myocardial infarction and diabetes: experiences from an extended follow-up of the Diabetes Mellitus Insulin-Glucose Infusion in Acute Myocardial Infarction (DIGAMI) 2 Study. Diabetologia. 2011;54(6):1308–1317. [PubMed]
20. Malmberg K, et al. Intense metabolic control by means of insulin in patients with diabetes mellitus and acute myocardial infarction (DIGAMI 2): effects on mortality and morbidity. Eur Heart J. 2005;26:650–661. [PubMed]
21. Cheung NW, Wong VW, McLean M. The Hyperglycemia: Intensive Insulin Infusion in Infarction (HI–5) study: a randomized controlled trial of insulin infusion therapy for myocardial infarction. Diabetes Care. 2006;29:765–770. [PubMed]
22. Kosiborod M, et al. Glucose normalization and outcomes in patients with acute myocardial infarction. Arch Intern Med. 2009;169:438–446. [PubMed]
23. Bellodi G, et al. Hyperglycemia and prognosis of acute myocardial infarction in patients without diabetes mellitus. Am J Cardiol. 1989;64:885–888. [PubMed]
24. Capes SI, et al. Stress hyperglycemia and increased risk of death after myocardial infarction in patients with and without diabetes: a systematic overview. Lancet. 2000;355:773–778. [PubMed]