A 50-year-old male patient presented to the emergency department at 2 AM with vomiting and abdominal pain. He had a 2-week history of polyuria and polydipsia, accompanied by a 20-pound weight loss and blurred vision. His medical history was unremarkable, except for being treated for hypertension with amlodipine 10 mg daily, which provided good control. (His blood pressure on admission was 135/80.)
Case Study 1: Patient with Newly Diagnosed Type 1 Diabetes
Author: Zachary T. Bloomgarden, MD
A 50-year-old male patient presentedto the emergency department at 2 AM with vomiting and abdominal pain. He had a2-week history of polyuria and polydipsia, accompanied by a 20-pound weightloss and blurred vision. His medical history was unremarkable, except for beingtreated for hypertension with amlodipine 10 mg daily, which provided goodcontrol. (His blood pressure on admission was 135/80.)
Results of hospital laboratory studies (Table 1-1) revealedthat the patient's initial blood glucose level was 1192 mg/dL and clinicalpresentation and laboratory findings were consistent with a diagnosis ofdiabetic ketoacidosis (DKA). The patient reported no family history ofdiabetes. His father died at age 35 of renal failure.
The patient was treated successfully for DKA and dischargedfrom the hospital 3 days later on an insulin regimen consisting of 30 units ofNPH/regular human insulin 70/30 mixture (70/30 mix) before breakfast, 15 unitsof regular human insulin before dinner, and 20 units of NPH insulin at bedtime.
On discharge, he was instructed to perform blood glucosemeasurements 4 times a day. The patient was seen as an outpatient 4 days afterhe is discharged from hospital.
Table 1-1. Hospital Laboratory Studies
| 2 AM | 4 AM | 5 AM | 9 AM | 11 AM | 1 PM |
| Plasma |
|
|
|
|
|
|
| Glucose mg/dL | 1192 | 958 | 718 | 358 | 288 | 222 |
| Sodium mEq/L | 154 | -- | 158 | 167 | -- | 161 |
| Potassium mEq/L | -- | -- | 3.3 | 4.0 | -- | 3.5 |
| Creatinine (mg/dL) | 1.7 | 1.6 | -- | -- | 0.9 | -- |
| pH | 7.34 | -- | -- | -- | -- | -- |
| Urine |
|
|
|
|
|
|
| Acetone | -- | -- | -- | 2+ | -- | -- |
| Case Notes This patient presented to the emergency department with acute-onset diabetes with classic symptoms of insulin deficiency compatible with a diagnosis of type 1 diabetes. Approximately 25% of patients that present with DKA have new onset of type 1 diabetes. Antibody testing was not performed, presumably because of the typical type 1 presentation. During his hospital stay, the patient received instructions regarding diet, medication schedule, and home glucose monitoring. The patient was discharged on an insulin regimen designed for ease of administration, consisting of a premixed 70/30 mix before breakfast. The evening insulin regimen included regular insulin before dinner and NPH at bedtime. Because the peak action of NPH is expected approximately 8-10 hours following administration, giving the evening dose of NPH at bedtime rather than before dinner avoids the nocturnal (2-3 AM) hypoglycemia that is often associated with dinnertime NPH administration. |
|
|
|
|
|
|
|
|
|
First Office Visit: Evaluating Glycemic Control
At the first office visit afterdischarge, the patient reported feeling well. Physical examination showedhyporeflexia, but was otherwise normal. No evidence of nephropathy orretinopathy was found. Blood and urine samples were collected for later analysis.
The previous 4 days of self-monitoring of blood glucose(SMBG) revealed a pattern of initial hyperglycemia followed by some improvement(Table 1-2).
| Day 1 | Day 2 | Day 3 | Day 4 | Day 5 |
| Breakfast |
| 416 | 159 | 94 | 164 |
| Lunch |
| 246 | 151 | 127 | 73 |
| Dinner |
| 181 | 117 | 124 | Office visit |
| Bedtime | 357 | 133 | 75 | 177 | -- |
Analyses of blood and urine samplescollected at the first office visit showed elevated hemoglobin A1c(HbA1c) (Table 1-3).
Table 1-3. Laboratory Results
| Blood | Patient | Normal Values |
| HbA1c | 10.4% | < 6%* |
| Plasma |
|
|
| Glucose (mg/dL) | 80 | < 110* |
| Creatinine (mg/dL) | 1.0 | ≤ 1.2 mg/dL |
| Total cholesterol (mg/dL) | 143 | < 200 mg/dL |
| Low-density lipoprotein cholesterol (LDL [mg/dL]) | 83 | < 130 mg/dL |
| High-density lipoprotein cholesterol (HDL [mg/dL]) | 45 | ≥ 35mg/dL |
| Triglycerides (mg/dL) | 74 | < 200 mg/dL |
| Urine |
|
|
| Albumin (mcg/mg creatinine) | 6 | < 30 |
For patients with diabetes, the American DiabetesAssociation recommends glucose values in the range of 90-130 mg/dL for plasma(80-120 mg/dL for whole blood) and HbA1c levels < 7%.[1]
| Case Notes This patient was placed on a multiple daily insulin injection (MDI) regimen and showed good adherence to home glucose monitoring and willingness to participate in close follow-up with weekly (or more frequent) telephone calls to the office and monthly office visits. The high HbA1c supports the existence of sustained hyperglycemia prior to hospitalization. There may be some peripheral neuropathy, but no evidence of nephropathy or retinopathy is found. Over the next few weeks, the goal was to stabilize his insulin regimen to his diet and lifestyle. |
Adjusting Insulin Regimen to Optimize GlycemicControl
Subsequent to starting on 30 unitsof 70/30 mix before breakfast, 15 units of regular insulin before dinner, and20 units of NPH bedtime, the patient's blood glucose levels were somewhat low.The insulin regimen was modified to 28 units of 70/30 mix in the morning, 12units of regular insulin at dinner, and 12 units of NPH at bedtime. Amlodipinewas discontinued. Home glucose monitoring showed recurring lunchtimehypoglycemia and bedtime hyperglycemia (Table 1-4).
| Day 1 | Day 2 | Day 3 |
| Breakfast | 100 | 176 | 142 |
| Lunch | 72 | 54 | 50 |
| Dinner | 149 | 117 | -- |
| Bedtime | 207 | 209 | -- |
In response to the lunchtimehypoglycemia, the 70/30 mix was discontinued in favor of having the patient mixNPH and regular insulin. The patient was given instruction on how to combinethe insulins, and was started on 20 units of NPH and 4 units of regular insulinin the morning, 14 units of regular insulin in the afternoon, and 10 NPH atbedtime.
| Case Notes With low noon blood glucose levels (50-72 mg/dL) and high bedtime levels (207-209 mg/dL), it was decided that this patient would benefit from the added flexibility of altering his insulin regimen beyond what the premixed formulation allowed. However, it is important to consider that the extra time and effort of mixing insulins may impose an added burden on a patient already overwhelmed by coping with the recent diagnosis of type 1 diabetes. In addition, there is potential for errors in mixing insulin, which may ultimately result in a deterioration of glycemic control. |
A Change in Insulin Routine: Introducing aRapid-Acting Insulin Analogue
Continued low lunchtime bloodglucose levels prompted a discontinuation of regular insulin and replacementwith the rapid-acting analogue insulin lispro. To start, the patient wasinstructed to take 16 units of NPH and 2 units of insulin lispro in themorning, 10 units of insulin lispro before dinner, and 10 units of NPH atbedtime.
However, because of persistently low lunch time glucoselevels, the morning dose of insulin lispro was discontinued, and the insulinregimen altered to 8 units of NPH before breakfast, 8 units of insulin lisprobefore dinner, and 10 units of NPH at bedtime.
As seen in Table 1-5, the patient continued to show lowglucose levels at lunch and dinner. Therefore, the morning NPH dose wasdecreased to 6 units. He was also given nutritional counseling and prescribed a1500-calorie/day diet, with 150 g carbohydrates distributed over 5 feedings.
| Day 1 | Day 2 | Day 3 | Day 4 | Day 5 |
| Breakfast | 113 | 103 | 98 | 104 | 104 |
| Lunch | 71 | 75 | 51 | 57 | 105 |
| Dinner | 80 | 85 | 91 | 113 | 105 |
| Bedtime | 105 | 84 | 137 | 114 | 140 |
| Case Notes Insulin lispro has an onset of action 15 minutes after injection, a maximal effect 60-90 minutes after dosing, and a duration of action of 3-4 hours. Regular insulin has an onset of action 1-2 hours after injection, a maximal effect 3-4 hours after dosing, and a duration of action of 6 or more hours. Thus, the pharmacokinetics of insulin lispro are more closely matched to the normal rise in postprandial glucose levels, whereas that of regular insulin is often found to lead to hypoglycemia 3-4 hours after the meal, when nutrient absorption has ceased and insulin action remains high.[2,3] Although often considered a treatment particularly effective for type 1 diabetes, insulin lispro also benefits patients with type 2 diabetes.[4] |
|
|
|
|
|
|
|
|
Reevaluating Metabolic Status
During the following weeks, thepatient's blood glucose became more stable. However, he continued to experiencesome fluctuation in glucose levels, especially lows at lunchtime. The morninginsulin dose was discontinued. However, during this time he expressed feelingsof depression and was referred for mental health counseling. Ophthalmologicreferral was made as well. The HbA1c was 7.8%.
| Case Notes Although the patient's HbA1c improved, it remained higher than recommended by American Diabetes Association guidelines. However, because HbA1c reflects glycemic status for approximately a 12-week period, SMBG values (if considered adequately reliable) remain the most important tool for monitoring glycemia. Because the emotional state of the patient will directly affect his ability to manage his diabetes, it is important to consider those needs by recommending counseling. |
Six months after initial diagnosis,SMBG showed consistent readings between 63 and 127 mg/dL, and normal HbA1c(5.4%). Because of hypoglycemia, insulin was completely discontinued. Treatmentwith repaglinide, a short-acting insulinotropic agent, is initiated at a doseof 1 mg twice daily. To transition to oral agents, the patient was instructedto taper the current insulin regimen of 6 units of NPH before breakfast, 8units of insulin lispro before dinner, and 10 units of NPH at bedtime asfollows: decrease to 4 units of each insulin during the first week, to 2 unitsof each insulin during the second week, and discontinue the insulin thereafter.The patient was monitored by weekly telephone contact during this process. Hewas started on ramipril 5 mg daily, which was subsequently increased to 10 mgdaily and then 20 mg daily to achieve blood pressure control and preserve renalfunction. The HbA1c levels remained normal (5.5%).
| Case Notes The lack of glycemic variability on a stable insulin regimen suggests type 2 diabetes. Because of the patient's tendency to become hypoglycemic, he was started on the short- acting insulin secretagogue repaglinide due to its association with a low incidence of hypoglycemia. It is important to consider that the patient may have type 1 diabetes and is in an extended "honeymoon period," and thus DKA may recur. This is a concern, but the clinical picture is more suggestive of "type 1 1/2 diabetes," which can initially present with DKA, but subsequently attain adequate glycemic control with oral agents.[5] Indeed, the argument can be made that by undergoing an initial period of intensive insulin treatment, the patient was sensitized to allow optimal response to oral agents.[6] The cornerstone of management is ongoing home glucose monitoring so that if changes in metabolic status occur, insulin can promptly be restarted. |
Rethinking the Initial Diagnosis of Type 1 Diabetes
Fourteen months after his episode ofDKA, the patient was demonstrating good metabolic control (Table 1-6). Hecontinues to take repaglinide without needing insulin. He is also taking 20mg/day of ramipril, 81mg/day of aspirin, a multivitamin, and a magnesiumsupplement. The HbA1c was 5.3% and lipids and blood chemistries werewithin the normal range.
| Aug 28 | Aug 29 | Aug 30 | Aug 31 | Sep 1 | Sep 2 | Sep 3 | Sep 4 | Sep 5 | Sep 6 |
| Breakfast | 121 | 134 | 69 | 128 | 118 | 122 | 119 | 121 | 132 | 124 |
| Lunch | 116 | 106 | 100 | -- | 99 | 116 | 78 | 129 | 92 | 98 |
| Dinner | 169 | 95 | 83 | 82 | 129 | 117 | 113 | 94 | 89 | 118 |
| Bedtime | 119 | 126 | 114 | 100 | 110 | 128 | 43 | 138 | 103 | 151 |
Case Summary and Perspectives
After an initial aggressive insulintherapy,[7] the patient recovered insulinsecretory capacity and has an excellent short-term prognosis for maintainingglycemic control. Ideally, in view of the severity of the diabetes at the initialpresentation of DKA, the patient most likely would have been maintained oninsulin. This is appropriate for diabetes resulting from autoimmune islet celldestruction, and it has also been argued that in patients with type 2 diabetesthe effect of amylin, which has been implicated in the pathogenesis ofbeta-cell failure, can be minimized with insulin treatment.[8]However, in this case, physiologic as well as psychological factors prompted consideringoral agents, which is a common occurrence in clinical practice.[9] Infact, the patient exhibited excellent glycemic control on oral agents. Therather low dose of repaglinide and the lack of obesity and of dyslipidemiasuggested that the degree of insulin sensitivity in this case was fairly good.[10] Ofcourse, one cannot be certain whether this patient will eventually require theaddition of other oral agents and perhaps insulin, as indeed is the case forthe majority of patients with type 2 diabetes.[11]
1. American Diabetes Association.Standards of medical care for patients with diabetes mellitus. Diabetes Care.2001;24(suppl 1):S33-S43.
2. Wilde MI, McTavish D. Insulinlispro: a review of its pharmacological properties and therapeutic use in themanagement of diabetes mellitus. Drugs. 1997;54:597-614.
3. Vignati L, Anderson JH Jr, IversenPW. Efficacy of insulin lispro in combination with NPH human insulin twice perday in patients with insulin-dependent or non-insulin-dependent diabetesmellitus. Multicenter Insulin Lispro Study Group. Clin Ther. 1997;19:1408-1421.
4. Anderson JH Jr, Brunelle RL, KeohaneP, et al. Mealtime treatment with insulin analog improves postprandialhyperglycemia and hypoglycemia in patients with non-insulin-dependent diabetesmellitus. Multicenter Insulin Lispro Study Group. Arch Intern Med.1997;157:1249-1255.
5. Banerji MA, Chaiken RL, Lebovitz HE.Prolongation of near-normoglycemic remission in black NIDDM subjects withchronic low-dose sulfonylurea treatment. Diabetes. 1995;44:466-470.
6. Ilkova H, Glaser B, Tunckale A,Bagriacik N, Cerasi E. Induction of long-term glycemic control in newlydiagnosed type 2 diabetic patients by transient intensive insulin treatment.Diabetes Care. 1997;20:1353-1356.
7. Herman WH, Dasbach EJ, Songer TJ,Eastman RC. The cost-effectiveness of intensive therapy for diabetes mellitus.Endocrinol Metab Clin North Am. 1997;26:679-695.
8. Lindstrom T, Leckstrom A, WestermarkP, Arnqvist HJ. Effect of insulin treatment on circulating islet amyloidpolypeptide in patients with NIDDM. Diabet Med. 1997;14:472-476.
9. Hunt LM, Valenzuela MA, Pugh JA.NIDDM patients' fears and hopes about insulin therapy. The basis of patientreluctance. Diabetes Care. 1997;20:292-298.
10. Ferrannini E. Insulin resistanceversus insulin deficiency in non-insulin-dependent diabetes mellitus: problemsand prospects. Endocr Rev. 1998;19:477-490.
11. Matthews DR, Cull CA, Stratton IM,Holman RR, Turner RC. UKPDS 26: Sulphonylurea failure in non-insulin-dependentdiabetic patients over six years. UK Prospective Diabetes Study (UKPDS) Group.Diabet Med. 1998;15:297-303.