only factor promoting diabetes in cancer patients; diabetes is limited because of patients with long-term steroid diabetes, education by a certi …
HOW
WE
DO
IT
How to Manage Steroid Diabetes in the Patient With Cancer
David S Oyer, MD, FACE, Ajul Shah, BS, and Susan Bettenhausen, APRN, CDE
G
lucocorticosteroids steroids have profound effects on glucose metabolism, particularly on postprandial hyperglycemia Patients with cancer often receive steroids as a component of their chemotherapy, as a measure to treat or prevent nausea, or as adjuvant therapy following neurosurgical procedures The oncology caregiver may not notice steroid-induced hyperglycemia, either because it is not considered or because steroids affect post-meal glucose much more so than morning fasting sugars A recent study suggests that even a few days of hyperglycemia have deleterious effects on the immune system1 The current trend for maintaining near-euglycemia in hospitalized patients is the use of intensive insulin therapy in hyperglycemic patients with or without diabetes This article will discuss steroid effects on glucose metabolism, recommend levels at which therapy should be considered, and discuss the options available for treating hyperglycemia caused by steroids At this time, there are no official guidelines for the cancer patient with steroid
diabetes, but the guidelines for diabetes care in general will be reviewed
Pathophysiology
Steroids induce a state of relative insulin resistance Steroid effects on glucose metabolism include down-regulation of glucose transporter 4 GLUT-4 in the muscle so that more insulin is needed for the uptake of glucose into cells Steroids may also promote glucose production in the liver, reduce binding of insulin to the insulin receptor on cells, and decrease insulin secretion from the islet cell In patients known to have diabetes, steroids will worsen the hyperglycemia, whereas non-diabetic patients, depending on the state of their islet cell reserve, may experience hyperglycemia or even overt diabetes In rare instances, hyCorrespondence to: David S Oyer, MD, 211 E Chicago Avenue, Suite 1050, Chicago, IL 60611; telephone: 312 9446677; fax: 312 944-3346; e-mail: d-oyer@northwesternedu
J Support Oncol 2006;4:479483 2006 Elsevier Inc All rights reserved
perglycemic nonketotic hyperosmolar coma may even ensue Increased steroid levels are not the only factor promoting diabetes in cancer patients; infection, inactivity, emotional stress, intravenous glucose, and high carbohydrate diets also
increase the tendency toward hyperglycemia Multiple reviews have emphasized the importance of intensive insulin therapy in hospitalized patients,25 and several studies reinforce the importance of tight glucose control in this patient population Furnary et al1,6,7 showed that aggressive control of postoperative blood glucose levels in diabetic patients who had undergone a coronary artery bypass graft reduced sternal wound infections and also improved morbidity and mortality The Diabetes Mellitus, Insulin-Glucose Infusion in Acute Myocardial Infarction DIGAMI study8 showed that, compared with conventional therapy, 48 hours of tight glucose control in patients with acute myocardial infarction reduced mortality The Van den Berghe study9 may be relevant to steroid-induced hyperglycemic patients, since the study population consisted largely of patients in the intensive care unit ICU not known to be diabetic who developed hyperglycemia in the hospital This response may have been related to the infused glucose and the stress-induced endogenous steroid production Patients who had a glucose reading above 110 mg/dL were randomized to receive intensive insulin therapy or conventional therapy
Patients in the intensive insulin group received insulin infusions set to reduce their glucose level to 80110 mg/dL, whereas those in the conventional therapy group received insulin treatment only if the glucose level went above 215 mg/dL and maintenance of glucose at a level between 180 and 200 mg/dL The final glucose averages of the intensive and conventional groups were 103 mg/dL and 153 mg/dL, respectively The benefits of tight glucose control in the intensive insulin group included a reduction in overall mortality, particularly in patients who remained in the ICU 5 days, and a reduced risk of sepsis, transfusions, renal failure, and ICU
Dr Oyer is Assistant Professor of Clinical Medicine, Division of Endocrinology, Feinberg School of Medicine, Northwestern University, Chicago, Illinois Mr Shah is a medical student, the University of Texas Southwestern Medical School, Dallas Ms Bettenhausen is a Clinical Nurse Specialist and Certified Diabetes Educator, Associates in Internal Medicine, Ltd, Chicago, Illinois
VOLUME 4, NUMBER 9
OCTOBER 2006
wwwSupportiveOncologynet
479
Steroid Diabetes in the Patient With Cancer
Table 1
Preparations of Glucocorticosteroids
PREPARATION
GLUCOCORTICOID POTENCY MINERALOCORTICOID EFFECT HALF-LIFE
Cortisone Prednisone Methylprednisolone Dexamethasone
10 40 48 200
/ 0
46 h 612 h 612 h 12 d
neuropathy In a more recent study, Van den Berghe et al10 achieved reduced morbidity, but not reduced mortality, using an intensive insulin therapy protocol in a medical ICU MICU Mortality was reduced only in those in the MICU for longer than 3 days Hyperglycemic patients not known to be diabetic suffered greater morbidity from hyperglycemia than known diabetics with the same degree of hyperglycemia A 2005 study11 showed less morbidity postoperatively if the glucose was controlled during surgery to a level 140 mg/dL by insulin infusions compared with a control group that averaged 180 mg/dL during surgery In all of these studies, insulin was used for glycemic control, so it is unknown if the benefits result from the lower glucose or possibly from direct effects of the insulin itself12,13 In vitro studies have shown that insulin can reduce many of the factors elicited in stress and infection, such as nuclear-factor kappa B,14 and these factors may be as important in the cancer patient as they are in the ICU patient, which is
one reason to prefer insulin therapy over oral therapy The aforementioned studies were all performed on inpatients, so we can only assume that outpatients might benefit as well There is no proof that the benefits of intensive glucose control will extrapolate to hyperglycemic cancer patients on steroids, and prospective studies targeting this group are needed
glucose levels; the most sensitive time to test for hyperglycemia is 2 hours after lunch The normal range for fasting glucose is below 100 mg/dL, and the normal level for the 2-hour postprandial reading is below 140 mg/dL Diabetic levels are 127 mg/dL or more for fasting glucose and 200 mg/ dL or more for the 2-hour postprandial level Glucose values between 100 and 126 mg/dL represent impaired fasting glucose, and postprandial glucose levels of 141199 mg/dL represent impaired glucose tolerance There is no way to predict whether glucose values will return to normal after cessation of steroids, since steroids may have unmasked a pre-existing tendency toward diabetes Current glucose monitors for home use are accurate to within 10, and there are few drugs that interfere with the current monitoring strips Severe anemia may cause
false meter readings A1C testing is not indicated to monitor short-term hyperglycemia, since it represents a 90-day summation of all the glucose variations but should be done at least once whenever hyperglycemia is noted A high A1C would indicate undetected previous diabetes It should be noted that the A1C test is only valid if the life of the red cell is normal; in transfused or anemic patients, false low A1C readings may be seen
When to Treat
For outpatients with only a few days of mild steroid hyperglycemia, therapy may not be needed Treatment would require patients to learn to test their glucose, take pills, or give insulin, all with no proven benefit In the hospital, however, this type of management is possible and should be considered for even brief episodes of hyperglycemia If the patient has only a few days of steroid therapy, and hyperglycemia is not causing symptoms, no therapy is indicated However, patients in the hospital for 3 days or more with a fasting glucose level over 110 mg/dL or a postprandial glucose level over 140 mg/dL would be candidates for therapy
Preparations of Glucocorticosteroids
Steroid diabetes is related to the dose of steroids used but not the
type Insulin resistance causes primarily postprandial hyperglycemia This causal relationship is particularly evident with morning doses of prednisone and could be partially related to the steroid effect wearing off overnight, but the improvement in glucose overnight is also seen with the longer-acting dexamethasone A typical patient will have elevated glucose values after breakfast, lunch, dinner, and at bedtime but will have a significant drop toward normal glucose overnight Therefore, hyperglycemia is greatest 12 hours after a meal, with persistent elevation until the following meal, followed by a return to normal between 4:00 am and 7:00 am Table 1 shows different preparations of glucocorticosteroids; it is not clear that the hyperglycemic effects will differ despite the different half-lives
Options for Therapy
The first steps in the management of steroid diabetes are diet and exercise The appropriate diet is low in carbohydrates to minimize postprandial hyperglycemia, but if the cancer patient has anorexia or cachexia, limiting the choices of food is not desirable In these cases, nutritional consultation is recommended Avoidance of other medications that may promote
hyperglycemia, such as hydrochlorothiazide 125 mg/d, niacin, and some selective serotonin reuptake inhibitors, might be helpful
ORAL AGENTS
Monitoring Glucose
Ideally, all cancer patients who receive steroids should be monitored with both pre-meal glucose and 2-hour post-meal
The role of oral agents in the oncology patient with steroid diabetes is limited because of potential side effects, slow onset of action, and lack of flexibility Oral secretagogues Table 2 may be useful in mild cases, although 24-hour secretagogues sulfonylureas do not selectively target postprandial hyperglycemia and may increase the risk of hypoglycemia in the morning or if the patient misses a meal There are two short-acting secretagogues that have unique features The shortest-acting
480
wwwSupportiveOncologynet
THE JOURNAL OF SUPPORTIVE ONCOLOGY
Oyer, Shah, and Bettenhausen
Table 2
Table 3
Uses of Oral Secretagogues
In cases of mild hyperglycemia During short-term steroid use For outpatients For patients unable or unwilling to give injections To bridge therapy until diabetic education is complete
Common Insulins
HUMAN INSULIN AND INSULIN ANALOGUES ONSET PEAK DURATION
agent is nateglinide
Starlix, which can be given immediately before meals to control postprandial glucose and has a half-life of about 46 hours The effect of nateglinide typically wears off overnight, an advantage in a patient prone to low glucose levels in the morning If the patient does not eat, he or she just skips the pill If the patient takes the pill and then is unable to eat, the risk of hypoglycemia is low Repaglinide Prandin is another pre-meal insulin secretagogue that can be given with meals, but it has a longer half-life than nateglinide and may lead to hypoglycemia in the morning if given with dinner Both of these secretagogues should be taken before eating, since their effect is blunted if taken during or after a meal The other oral antidiabetic agents are even less useful Metformin has multiple problems in the patient with cancer, with side effects including nausea, diarrhea, or vomiting It needs to be started at a low dose and increased gradually, and it does not target post-meal glucose Metformin is contraindicated in renal failure and must be stopped for any iodinated contrast dye studies; it may be resumed 48 hours later, once the creatinine level is documented as normal Metformin is
not indicated in patients with liver disease The greatest fear with metformin is lactic acidosis, and thus it is not recommended in patients with any medical condition predisposing them to sepsis, dehydration, or hypoxemia The thiazolidinediones have appeal because they directly treat the insulin resistance caused by steroids and seldom cause hypoglycemia, but their use is not often practical They have a long onset of action 12 weeks and prolonged effect after discontinuation also 12 weeks, which does not allow for short-term titration or adjustment These agents also promote weight gain, fluid retention, and edema, although overt heart failure rarely occurs The resulting fluid retention may cause a drop in the hemoglobin level of up to 12 g/dL If the patient does not have issues with fluid retention, and the duration of steroid use is prolonged and constant, the thiazolidinediones might play a useful supporting role The dose would be titrated primarily against the pre-meal morning glucose result The incretin mimetics are a new class of drugs for treating type-2 diabetes Only the glucagon-like peptide 1 mimetic exenatide Byetta is currently on the market, but other compounds are
likely to follow Incretin mimetics target postprandial hyperglycemia but have not been studied in steroid diabetes Worrisome side effects are their tendency to cause nausea and vomiting in about 50 of patients and to decrease appetite with resultant weight loss
Basal insulins NPH 13 h Glargine 24 h Detemir 24 h Human short-acting insulin Regular 3060 m Rapid-acting analog insulins Lispro 1015 m Aspart 1015 m Glulisine 1015 m
Abbreviation: NPH neutral protamine Hagedorn
410 h 412 h 412 h 24 h 12 h 12 h 12 h
1018 h 1824 h 1224 h 48 h 36 h 36 h 36 h
Insulin Therapy
For most patients, insulin will be a more appropriate therapy than oral agents Available insulins are listed in Table 3 For patients in the ICU or MICU who are not eating, the current standard of care is an infusion of rapid-acting insulin targeting a glucose level of 80115 mg/dL Various insulin protocols exist, and examples have been published on the web sites for the American Association of Clinical Endocrinologists15 and the American Diabetes Association16 and in the reviews by Clement3 and Furnary1 Although insulin infusions increase the work of nurses, the benefits justify this extra effort Other long-term trials
are in progress to evaluate the advantages of intensive insulin therapy17 Subcutaneous insulin replacement therapy is used for patients who are eating There are three components to insulin therapy: basal insulin, prandial insulin, and supplemental insulin Basal insulin is most often a long-acting insulin used to suppress glucose production by the liver; it controls the glucose during fasting and overnight Prandial insulin is needed to prevent a glucose rise after ingestion of food Supplemental insulin is used to lower the glucose level when it is high, either with a meal or at other times
BASAL INSULIN
Three basal insulins are available: neutral protamine Hagedorn NPH, glargine Lantus, and detemir Levemir In the steroid-induced diabetic, basal insulins should be administered in the morning rather than bedtime Insulin requirements will be higher during the day and the lowest in the early morning, so the slight peak that the basal insulins have at 410 hours post administration should occur during the day and not at night NPH has a greater tendency to peak than the two analog basal insulins and a shorter duration of action, both of which might be an advantage in steroid diabetes The
newer analog insulins, however, have a flatter and more consistent time course, which is more likely to last 24 hours A common starting dose of basal insulin is
VOLUME 4, NUMBER 9
OCTOBER 2006
wwwSupportiveOncologynet
481
Steroid Diabetes in the Patient With Cancer
Table 4
Ten Key Facts About Steroid Diabetes
1 2 3 Primary effect is on postprandial glucose level Glucose values tend to normalize overnight Glucose levels should be tested before as well as 2 hours after a meal 4 Oral agents are usually inappropriate, ineffective, or too inflexible 5 Insulin is generally the best therapy 6 Prandial insulin is the primary need 7 Prandial insulin should be titrated to the glucose 2 hours post prandially or the next meal 8 Basal insulin should be given in the morning and titrated to the glucose level from the following morning 9 Target glucose levels are 115 mg/dL pre-meal and 140180 mg/dL 2 hours postprandially 10 Steroid diabetes is difficult to control: consults with endocrinologists, certified diabetes educators, and nutritionists are appropriate
added to the prandial dose The insulin sensitivity factor, or correction factor, is often about 40, indicating that 1 U of
insulin lowers the glucose level by 40 mg/dL We have a supplemental insulin scale based on low, medium, and high sensitivity, which correlates with correction factors of 30, 40, and 50 mg/dL per U of insulin A target glucose level is usually set at 100120 mg/dL To calculate the supplemental insulin dose, we use this formula: current glucose level target glucose level / correction factor units of supplemental insulin required
TIMING OF THERAPY
10 U or 02 U/kg, whichever is higher The dose is then adjusted based on the morning glucose result
PRANDIAL INSULIN
Prandial insulin can be regular or analog insulin The advantage of the new analog insulins is a more rapid onset of action, a quicker peak, and shorter duration of action This profile allows them to be given right before a meal, whereas regular insulin ideally should be given 3045 minutes before a meal Analog insulin may also be given immediately after a meal, an advantage in nauseated patients unsure of how much they will consume One disadvantage of the analog insulin for prandial coverage is the shorter duration of action compared with regular insulin, which may allow the glucose level to rise before the next meal or stay
high until bedtime High bedtime readings need to be treated cautiously, if at all, since that is the time of day when glucose starts to decrease naturally When possible, prandial insulin doses should be evaluated 2 hours after a meal Doses of short-acting insulin repeated in less than 4 hours need to be reduced to account for the residual effect of the previous dose about 25 an hour to avoid hypoglycemia from the overlapping duration of action To establish a starting prandial dose, we utilize low-dose, medium-dose, and high-dose protocols based on previous results or intuition Common starting doses are 5, 10, or 15 U three times daily, with further adjustments as needed There may be some advantage to teaching patients carbohydrate counting, allowing them to adjust the prandial insulin based on the amount of carbohydrates in a meal A common starting ratio is 1 U for every 10 g of carbohydrate Consultation with a dietician or certified diabetes educator will be necessary if carbohydrate counting is required
SUPPLEMENTAL INSULIN
Prandial insulin should be started first whenever postprandial hyperglycemia is detected A standard dose such as 5 U can be given and adjusted based on the
results, aiming for a postprandial glucose level below 140180 mg/dL Some patients may benefit from the more difficult regimen of adjusting the dose based on the carbohydrate content of meals and prevailing glucose levels A basal insulin given in the morning can help keep glucose levels down during the day but is seldom enough alone to control postprandial hyperglycemia, even with the shorter-acting NPH The longer-acting basal insulins are used primarily to keep the glucose level down overnight or when the patient is not eating Inhaled insulin might prove to be useful in patients who need prandial control but cannot be used in anyone with pulmonary disease For those receiving inhaled insulin, baseline forced expiratory volume in the first second has to be measured and monitored every 3 months Garg et al18 provide a good review of insulin therapy Insulin therapy is complicated and the results are often unsatisfactory; one should never hesitate to ask for help from an endocrinologist
MONITORING RESULTS
The third component of insulin therapy is supplemental insulin, a dose of short-acting insulin used to correct a high glucose level This dose can be given either alone or
An insulin
therapy worksheet and flow sheet are helpful when treating diabetics because the response to previous therapy guides future changes see an example online at http://wwwsupportiveoncologynet/journal/0409html; worksheet adapted courtesy of Marcia Draheim, RN, CDE, of Draheim Dimensional Presentations; flowsheet courtesy of David S Oyer, MD If the steroids are stopped, the carbohydrate intolerance may disappear quickly, and when steroids are tapered, the insulin requirements usually recede proportionally to the reduction Any unexplained hypoglycemia should lead to an automatic reduction in the insulin dose about 20 Insulin doses can be lowered in anticipation of reduced steroid doses When planning repeated cycles of the same steroid doses, historic results can guide therapy In patients who receive dexamethasone as a one-time dose for prevention of nausea, hyperglycemia may last 23 days, and insulin may have to be adjusted with every meal to meet the changing requirements as the dexamethasone wears off The patient should be aware of the symptoms of hypoglycemia and should have snacks at the bedside in case of a low glucose level For patients with long-term steroid diabetes, education by
a certi-
482
wwwSupportiveOncologynet
THE JOURNAL OF SUPPORTIVE ONCOLOGY
Oyer, Shah, and Bettenhausen
fied diabetes educator should begin as soon as possible, and difficult cases should be referred to an endocrinologist
Conclusion
Table 4 lists some of the primary concerns when dealing with steroid diabetes that complicates the therapy of a cancer patient A goal of pre-meal readings of 110 mg/dL and 2-hour postprandial readings of 140180 mg/dL is ideal but difficult to achieve in steroid diabetes Since proof of the benefits of tight control in steroid diabetes does not exist,
safety and avoidance of hypoglycemia remain equally important goals The system of basal bolus insulin therapy is the most flexible option but is complicated to use For many, a basal insulin in the morning and then a standard dose of short-acting insulin before meals will offer the best result Further studies are needed to develop future protocols for the treatment of steroid diabetes In todays clinical setting, it is prudent to devote as much attention to glycemic control in the oncology patient as would be appropriate for patients in the ICU
References
1 Furnary AP, Wu Y, Bookin SO Effect of
hyperglycemia and continuous intravenous insulin infusions on outcomes of cardiac surgical procedures: the Portland Diabetic Project Endocr Pract 2004;10suppl 2:2133 2 The ACE/ADA Task Force on Inpatient Diabetes American College of Endocrinology and American Diabetes Association consensus statement on inpatient diabetes and glycemic control: a call to action Diabetes Care 2006;29:19551962 3 Clement S, Braithwaite SS, Magee MF, et al Management of diabetes and hyperglycemia in hospitals Diabetes Care 2004;27:553591 4 Umpierrez GE, Isaacs SD, Bazargan N, et al Hyperglycemia: an independent marker of in-hospital mortality in patients with undiagnosed diabetes J Clin Endocrinol Metab 2002;87:978982 5 Krinsely J Effect of an intensive glucose management protocol on the mortality of critically ill adult patients Mayo Clin Proc 2004:79:992-1000 6 Furnary AP, Zerr K, Grunkemeier G, et al Continuous intravenous insulin infusion reduces the incidence of deep sternal wound infection in diabetic patients after cardiac surgical procedures Ann Thorac Surg 1999;67:352362 7 Furnary AP, Gao G, Grunkemeier GL, et al Continuous insulin infusion reduces mortality in patients with diabetes undergoing
coronary artery bypass grafting J Thorac Cardiovasc Surg 2003;125:10071021 8 Malmberg K,Ryden L,Efendic S,et alRandomized trial of insulin-glucose infusion followed by subcutaneous insulin treatment in diabetic patients with acute myocardial infarction DIGAMI: effects on mortality at 1 year J Am Coll Cardiol 1995;26:5765 9 Van den Berghe G, Wouters P, Weekers F, et al Intensive insulin therapy in critically ill patients N Engl J Med 2001;345:13591367 10 Van den Berghe G, Wilmer A, Hermans G, et al Intensive insulin therapy in the medical ICU N Engl J Med 2006;354:449461 11 Ouattara A, Lecomte P, Le Manach Y, et al Poor intraoperative blood glucose control is associated with a worsened hospital outcome after cardiac surgery in diabetic patients Anesthesiology 2005;103:687694 12 Vanhorebeek I, Langouche L, Van den Berghe G Intensive insulin therapy in the Intensive Care Unit: update on clinical impact and mechanisms of action Endocr Pract 2006;12suppl 3:1422 13 Ellger B, Debaveye Y, Vanhorebeek I, et al Survival benefits of insulin therapy in critical illness: impact of maintaining normoglycemia versus glycemia-independent actions of insulin Diabetes 2006;55:10961105 14 Dandona P,
Mohanty P, Chaudhuri A, et al Insulin infusion in acute illness J Clin Invest 2005;115:20692072 15 American Association of Clinical Endocrinologists AACE Publications Available at: http://wwwaacecom/pub/ Accessed September 19, 2006 16 American Diabetes Association Clinical practice recommendations Available at: http://wwwdiabetesorg/for-health-professionals-and-scientists/cpr jsp Accessed September 19, 2006 17 Bellomo R, Egi M Glycemic control in the intensive care unit: why we should wait for NICE-SUGAR Mayo Clin Proc 2005;80:15461548 18 Garg S, Dailey G, Hirsch I Practical strategies for introducing insulin therapy in 2006J Fam Pract 2006;55suppl:S1S12
VOLUME 4, NUMBER 9
OCTOBER 2006
wwwSupportiveOncologynet
483
Source:supportiveoncology.net
