DIABETES: MANAGING INSULIN/HOSPITALIZED PATIENTS
From Mayo Clinic’s Clinical Reviews 2008: A Primary Care and Internal Medicine Update
Educational Objectives
| The goal of this program is to improve the management of diabetes and hyperglycemia in the hospitalized patient. After
hearing and assimilating this program, the participant will be better able to:
|
 | 1. Describe the mechanism of action and pathophysiology of insulin and recognize the importance of controlling blood
glucose (BG) in hospitalized patients.
|
| 2. Evaluate the efficacy of the various forms of insulin and differentiate among those with different pharmacokinetics.
|
| 3. Recognize the barriers to effective control of BG in diabetic patients and prescribe the appropriate form of insulin
therapy.
|
| 4. Analyze reports of the prevalence and consequences of hyperglycemia in hospitalized patients and appropriately diagnose
and treat their condition.
|
| 5. Assess the role of sliding scale insulin regimens for treatment of hospitalized patients with diabetes and
hyperglycemia.
|
Faculty Disclosure
In adherence to ACCME Standards for Commercial Support, Audio-Digest requires all faculty and members of the planning
committee to disclose relevant financial relationships within the past 12 months that might create any personal conflicts
of interest. Any identified conflicts were resolved to ensure that this educational activity promotes quality in health
care and not a proprietary business or commercial interest. For this program, Drs. Hogan and Whitaker and the planning
committee reported nothing to disclose.
Acknowledgements
Lectures given by Drs. Hogan and Whitaker were recorded at the Mayo Clinic’s Clinical Reviews 2008: A Primary
Care and Internal Medicine Update, held March 26-29, 2008, in Scottsdale, AZ. The Audio-Digest Foundation thanks
the speakers and the Mayo Clinic for their cooperation in the production of this program.
| RATIONALE FOR INSULIN THERAPY —Michael J. Hogan, MD, MBA, Assistant Professor, Department of Consultative
Medicine, Mayo Clinic College of Medicine, Scottsdale, AZ
|
| Physiology of insulin: insulin produced as metabolically inactive prohormone in islet cells of pancreas; cleaved in
blood to active A and B chains, C (connecting) peptide removed; when glucose absorbed in jejunum, serum concentration
rises, stimulating beta cells of pancreas to produce more insulin; 60% to 65% of insulin goes through portal vein to
liver, inhibiting glycogenolysis and gluconeogenesis; also inhibits glucose production in renal cells, and decreases glucose
in periphery by increasing uptake and utilization in muscle and adipose tissue
|
| Mechanism: insulin and insulin-like growth hormone very similar and act on similar receptors to affect growth of cells;
effect of insulin on growth both positive and negative (eg, vascular disease); transporters in cell membranes—enable
uptake of glucose; insulin reacts with membrane-bound receptor and initiates energy-producing steps that cause glucose
transporters to work; normally, blood glucose (BG) rises quickly after each meal with rapid corresponding rise in insulin
|
| Importance of glucose control: essential that treatment go beyond controlling ketoacidosis and urine output and
achieve good control of BG; study—meta-analysis showed difference in rate of development of nephropathy among patients
who had standard treatment, compared to those who had intensive treatment; similar differences seen in nerve conduction
(eg, tibial, peroneal, sural nerves); effects on development of retinopathy take longer (5 yr), but risk eventually
much greater in patients who had standard treatment
|
| Insulin physiology: genetic engineering of human insulin (combined with Zn or other moiety to enable injection and
persistence in tissue) influenced treatment; absorption and onset of action—injected insulin self-aggregates, forming
dimers or larger complexes that limit rate of absorption; normally, insulin increases almost instantaneously with meal and
decreases (but doesn’t disappear) after meal; injected insulins include regular or rapid-onset (onset of action begins 30-60
min after administration, concentration peaks after 2-3 hr and lasts ≤10 hr), as well as neutral protamine Hagedorn (NPH)
or intermediate, and extended forms; difficulty maintaining normal levels of insulin by injection leads to hyper- and hypoglycemia
|
| Barriers to control of BG: variable absorption of intermediate- and longer-acting insulins if not administered correctly;
formation of aggregates (beneficial for longer-acting forms but impedes absorption in short-acting); difficulty
maintaining regular mealtimes; nocturnal hypoglycemia
|
| Shorter-acting insulins: designed to aggregate less, for more rapid absorption and turnover, higher peak concentration,
and shorter duration of action; properties of insulin lispro protamine recombinant—minimal aggregation, unaltered receptor
affinity, more rapid absorption, higher peak concentration, and shorter duration of action (more like normal insulin
behavior); insulin aspart recombinant compared to human insulin—kinetics of binding and activating receptor similar;
can be safely used intravenously (IV), eg, with pumps, because safety profile similar; rate of absorption and maximum concentration
2-fold higher; shorter duration of action seen in both rapid-onset insulins
|
| Longer-acting insulins: provide steady levels of insulin to avoid postprandial and nighttime hypoglycemia; insulin
glargine recombinant—introduced first; stable in acidic injection medium; absorption rate 50% of NPH insulin; absorption
curve resembles straight line (unlike NPH insulin, which has arc-shaped curve) to reduce risk for hypoglycemia
after meals; slightly lower affinity for receptor than human insulin; rate of dissociation from receptor also similar
(important in limiting growth-stimulating mitogenic effects in vasculature); metabolic effects (eg, lipogenesis) slightly
lower; lower affinity for insulin-like growth factor (IGF); little difference between glargine and NPH insulins in fasting
BG, but significant improvement in nocturnal hypoglycemia; twice daily use of NPH insulin may cause hypoglycemia
in evening if patient skips snack; detemir (Levemir) insulin—newer long-acting form with long-chain myristic acid
residue added; increased aggregation prolongs persistence in tissues; binds reversibly to albumin in serum to maintain
constant release; study using clamp technology—patients given constant infusion of insulin, and glucose added to
maintain euglycemia; found detemir mimics effects of human insulin on liver more than other insulins, possibly because
of binding to albumin; receptor binding and release similar to human insulin, so less mitogenic effect; activity
profile intermediate between glargine and NPH, mimicking basal insulin levels and response to meals seen in normal
individuals
|
| Inhaled insulin: much faster onset than injected form, but withdrawn from market; long-term efficacy less than subcutaneous
insulin, but patient acceptance greater
|
| Cost: important element of compliance; injection pens and newer forms of insulin more expensive
|
| Goals for treatments in development: increased selectivity (to reduce mitogenic effects)—injectable proinsulin
not much more effective than intermediate-acting insulin, but shows less aggregation; thyroxyl-insulin complexes modify
aggregation and absorption; other delivery systems and improvements—better stability and ease of use; less variability;
ultrarapid onset for use after or with meals; ultra long-acting forms (24 hr); better BG control without mitogenic effects
|
| INTENSIVE INPATIENT DIABETES MEDICINE —Michael D. Whitaker, MD, Assistant Professor, Department of Endocrinology,
Mayo Clinic College of Medicine, Scottsdale
|
| Background: high prevalence of hyperglycemia in severely ill hospitalized patients not previously diagnosed with diabetes;
50% of septic nondiabetic patients and many trauma patients in intensive care unit (ICU) have elevated BG; 100% of
patients undergoing liver transplantation have hyperglycemia (some severe) within 12 hr
|
| Evidence of hyperglycemia in hospitalized patients: study—found 33% of 1034 hospitalized patients without
known diabetes had BG >200 mg/dL; second study—found almost 40% of patients in ICU needed insulin to control BG,
but only 15% had previous diagnosis of diabetes
|
| Associated effects: study—found almost 40% of ≈2000 patients admitted to hospital had hyperglycemia (BG >180),
26% had known diabetes, and 12% had newly diagnosed diabetes; mortality 1.7% overall for patients without hyperglycemia,
and much higher among patients with hyperglycemia; in ICU, 10% mortality for patients with normoglycemia,
and 4-fold higher mortality in patients with hyperglycemia or diabetes; mortality among patients in ICU with known diabetes
similar to normoglycemic patients (10%), but much higher among patients with new hyperglycemia; retrospective
study—followed patients with diabetes (mostly type 2) undergoing coronary artery bypass graft (CABG); 98% had BG
>110 mg/dL first day after surgery; risk for complications increased by 17% for every 18-mg/dL increase in BG >110
mg/dL; study of patients with stroke—among patients without known diabetes, BG >110 mg/dL predicted higher risk
for mortality and poor recovery after stroke
|
| Study of patients in cardiac ICU: among 1500 adult patients, 13% had known diabetes; 75% of patients without
known diabetes had BG >110 mg/dL (6.1 mmol); only 12% of cohort had BG <110 mg/dL; patients received either
conventional (ie, BG goal of 180-200 mg/dL) or intensified (ie, BG goal 80-110 mg/dL) insulin drip; on average, patients
needed ≈70 U insulin/day
|
| Mortality: intensive therapy group had much lower mortality rate in ICU; similar results for overall rate of death in hospital;
overall mortality reduced by one-third in patients receiving intensive treatment
|
| Morbidity: patients receiving conventional treatment had greater morbidity; also determined that intensive control of hyperglycemia
improved kidney function and lowered rates of sepsis, electromyographic changes, and neuropathy
|
| Rationale for cutoff of 110 mg/dL: found 1.3-fold (30%) increase in risk for death for every 20-mg/dL increase in BG,
and BG >200 mg/dL gave 2.5-fold higher risk for death
|
| Study limitations: not blinded; oriented toward cardiac surgery patients; following European practice, patients received
high levels of IV glucose within first 24 hr after ICU admission and total parenteral nutrition (TPN) 24 hr later (use of
excess glucose possibly caused hyperglycemia and increased mortality); high mortality rate (5%) for cardiac patients;
unexplained discrepancy between large reduction in mortality (34%) and small decrease in BG (50 mg/dL)
|
| Study of patients in medical ICU: 1200 patients expected to be in medical ICU 3 days received either intensive BG
control or conventional therapy; 1 in 6 patients had known diabetes; mean BG elevated; mortality—overall, no difference
in total number of ICU and in-house deaths between 2 regimens, but patients who remained in ICU >3 days benefitted
from intensive treatment; morbidity—patients receiving conventional treatment did worse than those receiving
intensive treatment, regardless of length of stay in ICU; hypoglycemia more common in intensively treated group and
also independent predictor of death; study limitations—similar to those of previous study, ie, patients also received
excess glucose and TPN
|
| Recommendations: do not ignore hyperglycemia in ICU; glycemic threshold and ideal level of control of BG still unclear;
consider tight control for patients expected to remain >3 days; need reassessment of role of early treatment with
TPN
|
| Future treatment in ICU: GLUControl Trial with 3500 medical and surgical patients and Normoglycemia in Intensive
Care Evaluation and Survival Using Glucose Algorithm Regulation (NICE-SUGAR) trial for medical and surgical
ICU patients evaluating tight and conventional control of BG
|
| Sliding scale insulin (SSI) for patients on ward: efficacy—in 1997, first study of stand-alone SSI adjustment
evaluated rates of hypo- and hyperglycemic episode in ≈170 patients with diabetes; 130 patients on SSI regimen (66%
on conservative SSI, 33% on aggressive SSI), with or without NPH or oral insulin; among patients without oral insulin,
25% had hypoglycemic episodes, and 40% had significant hyperglycemia (BG 180-200 mg/dL); found 3-fold higher
incidence of hyperglycemia in patients on SSI; SSI beneficial for patients on oral or NPH insulin
|
| Problems with SSI: retrospective study results—hospitalized patients treated with stand-alone SSI had 33% rate of uncertainties
in dosing information; about 10% rate of reaching BG goal; BG remained elevated in 84%; regimen never
adjusted in 80%; only 6% had good control of BG over 5 days; prospective study results—among 107 patients with
diabetes or hyperglycemia, 33% had hyperglycemia; <50% of patients on SSI received basal insulin, and only 4% received
bolus insulin after meals; treatment unmodified for 66% of patients with high BG; SSI associated with increased
BG of 20 mg/dL per patient per day
|
| Recommendations: use standing regimen of rapid or long-acting insulin or oral agent, check BG ante cibum (ac), and adjust
therapy; avoid using intermediate-acting NPH insulin; use SSI only as adjunct; supporting evidence—study of
130 insulin-naive patients received either basal/bolus insulin glargine recombinant (Lantus) and insulin glulisine recombinant
(Apidra) or SSI (4 times daily for BG >140 mg/dL); 70% of patients on basal/bolus had BG <140 mg/dL vs
33% on SSI alone; mean daily BG 27 mg/dL higher in SSI group; 14 patients on SSI had BG >240 mg/dL; no difference
in hypoglycemia or length of stay; American Association of Clinical Endocrinologists (AACE)—guidelines set
110 mg/dL goal for patients in ICU; 110 mg/dL preprandial and 180 mg/dL maximum for those not in ICU
|
| Questions and answers: insulin pumps—speaker does not favor pumps because studies show no difference between
pumps and multiple daily injections in controlling BG; lower acceptance by younger patients; higher cost and small risk
for mechanical failure; use of insulin and exenatide (Byetta) or sitagliptin (Januvia): Byetta genetically engineered injectable
analogue of glucagon-like pancreatic peptide hormone that affects gastric emptying and other functions; Januvia
inhibits degradation of peptides; both expensive, and role in insulinopenic patients unclear; oral administration of Januvia
may benefit patients who are not insulinopenic; differentiation of diabetes types 1 and 2 in young patient—need for insulin
determined by clinical picture; obese patient probably insulin-resistant and will respond to weight reduction, exercise,
and maybe oral agent; thin, very hyperglycemic patient needs insulin, and insulin injections have better efficacy; use
of insulin drip—essential for acutely ill patients; for stable patients in ICU, long-acting insulin (glargine) useful when
BG monitored every 4 to 6 hr and SSI adjusted; dosing of glargine—speaker favors morning dose to avoid nocturnal hypoglycemia
(check BG before breakfast); for very low (5-10 U) or high (200-300 U) doses, use twice daily; vitamin D
dosing—400 U/day (current recommended daily allowance [RDA]) not sufficient for bone health; minimum dose 800 to
1000 U/day; use of short-acting insulin for patients with type 2 diabetes who require high doses of insulin—
combination of oral agents with insulin no better than adequate insulin alone; speaker favors treating only with insulin;
avoid oral agents that cause hypoglycemia (eg, sulfonylureas); insulin sensitizers (eg, metformin) less likely to cause hypoglycemia
|
Suggested Reading
Baker ST et al: Outcomes for general medical inpatients with diabetes mellitus and new hyperglycaemia. Med J Aust
188:340, 2008; Becker RH, Frick AD: Clinical pharmacokinetics and pharmacodynamics of insulin glulisine. Clin
Pharmacokinet 47:7, 2008; Braithwaite SS: Inpatient insulin therapy. Curr Opin Endocrinol Diabetes Obes 15:159,
2008; Donner TW, Flammer KM: Diabetes management in the hospital. Med Clin North Am 92:407, 2008; Furie
K, Inzucchi SE: Diabetes mellitus, insulin resistance, hyperglycemia, and stroke. Curr Neurol Neurosci Rep 8:12,
2008; Gandhi GY et al: Effect of perioperative insulin infusion on surgical morbidity and mortality: systematic review
and meta-analysis of randomized trails.7. Mayo Clin Proc 83:418, 2008; Green DE: New therapies for diabetes. Clin
Cornerstone 8:58, 2007; Kitabchi AE et al: Evidence for strict inpatient blood glucose control: time to revise glycemic
goals in hospitalized patients. Metabolism 57:116, 2008; Leichter S: Is the use of insulin analogues cost-effective? Adv
Ther 25:285, 2008; Matheny ME et al: Treatment intensification and blood glucose control among hospitalized diabetic
patients. J Gen Intern Med 23:184, 2008; Mills RD et al: Evaluation of diabetes management in a rural community
hospital. Endocr Pract 14:50, 2008; Morales J: Defining the role of insulin detemir in Basal insulin therapy. Drugs
67:2557, 2007; Pickup JC, Renard E: Long-acting insulin analogs versus insulin pump therapy for the treatment of
type 1 and type 2 diabetes. Diabetes Care 31:S140, 2008; Retnakaran R, Zinman B: Type 1 diabetes, hyperglycaemia,
and the heart. Lancet 24:1790, 2008; Rosetti P et al: Superiority of insulin analogues versus human insulin in the
treatment of diabetes mellitus. Arch Physiol Biochem 114:3, 2008; Smith SA, Murad MH: Review: long-acting insulin
analogues do not improve glycaemic control but reduce nocturnal hypoglycaemia. Evid Based Med 13:79, 2008;
Trujillo JM et al: Improving glycemic control in medical inpatients: a pilot study. J Hosp Med 3:55, 2008; Umpierrez
GE et al: Sliding scale insulin use: myth or insanity? Am J Med 120:563, 2007; Webster KA: Stress hyperglycemia
and enhanced sensitivity to myocardial infarction. Curr Hypertens Rep 10:78, 2008; Wexler DF, Cagliero E:
Inpatient diabetes management in non-ICU settings: evidence and strategies. Curr Diabetes Rev 3:239, 2007.
|
