The Yale Insulin Protocol Updated 2011 4z2q6y

Title: Adapting to the New Consensus Guidelines for Managing Hyperglycemia During Critical Illness: The Updated Yale Insulin Infusion Protocol. Authors: Shilpa Shetty, MD1; Silvio E. Inzucchi, MD2,4; Philip A. Goldberg, MD2; Dawn Cooper, RN4; Mark D. Siegel, MD3,4; Shyoko Honiden, M.Sc. MD3,4

1

Department of Medicine, Griffin Hospital, Derby, CT; Sections of 2Endocrinology and

3

Pulmonary & Critical Care Medicine, Department of Medicine, Yale University School of

Medicine, New Haven, CT; 4Yale-New Haven Hospital, New Haven, CT

© 2011 AACE. DOI:10.4158/EP11260.OR

1

ABSTRACT Objective: The current AACE/ADA consensus statement stresses a less stringent blood glucose (BG) target (140-180 mg/dl) in the intensive care unit (ICU) than was previously endorsed. Since 2003, we have utilized a standardized IV insulin infusion protocol (IIP), initially targeting 100-140 mg/dl, revised in 2005 to 90-120 mg/dl. Both have been validated and published and are now used in many US hospitals. In response to the new guidelines, in 2009 we revised our IIP to target 120-160 mg/dl. Research Design and Methods: We prospectively tracked clinical responses to the new IIP in our medical ICU. Results: The IIP was used 115 times in 90 patients (mean age 62±14 years, 51% male, 35% ethnic minorities, 64% with history of diabetes). The mean ission APACHE-II score was 24.4±7.5. The median duration of insulin infusion was 59 hours. The mean baseline BG was 306.1±89.8 mg/dl, with the BG target achieved after a median of 7 hours. Once the target was reached, the mean IIP BG was 155.9±22.9 mg/dl (median 150 mg/dl.) The median insulin infusion rate required to reach and maintain the target range was 3.5 units/hour. Hypoglycemia was rare, with 0.3% of BGs recorded <70 mg/dl and only 0.02% <40 mg/dl. In all cases hypoglycemia was rapidly corrected using intravenous dextrose with no evident untoward outcomes. Conclusions: The updated Yale IIP provides effective and safe targeted BG control in the critically ill, in compliance with recent national guidelines. It can be easily implemented by hospitals now using the original Yale IIP.

2

Introduction Over the past decade there has been a great deal of controversy regarding the optimal management of hyperglycemia in the intensive care unit (ICU). In 2001, Van den Berghe et al (1) demonstrated the benefits of intensive glucose control (80-110 mg/dl) in a single center, prospective randomized controlled trial among surgical ICU patients. As a result of this compelling evidence, institutions around the world adopted insulin infusion protocols (IIP) to achieve stringent blood glucose (BG) targets in critically ill patients. Several subsequent studies (2-4) , however, questioned the efficacy and safety of targeting strict euglycemia and raised the concern of unacceptably high rates of hypoglycemia. The results of the most recent and largest trial to date, Normoglycemia in Intensive Care Evaluation - Survival Using Glucose Algorithm Regulation (NICE SUGAR),(4) in particular, demonstrated a slightly higher mortality among patients randomized to intensive glucose control and called for a re-evaluation of how best to manage hyperglycemia in the ICU. These discrepant findings led to uncertainty regarding the ideal management of BG in the critical care setting. In response, the AACE and ADA published a consensus statement suggesting a moderated target BG range of 140-180 mg/dl in an effort to minimize glycemic excursions at both extremes, as both extremely low and high blood glucose concentrations have been associated with adverse outcomes(5) . The statement also suggested that greater benefit may be realized with mean glucose levels toward the lower end of this range. In order to achieve these goals, it is important to implement a validated and safe IIP. At our institution, IV insulin infusions have been standardized since 2003. In response to emerging professional guidelines at that time endorsing tight ICU glycemic control as a standard of care, we developed a protocol that targeted a BG of 100-140 mg/dl. These data were published (6) and adaptations of our protocol have been used at many other centers and in several other studies (7,8) . Two years later, we revised our protocol (9) , targeting 90-120 mg/dl given the growing consensus to attain

3

euglycemia in the critically ill. With both protocols, our hypoglycemia rates were very low and we could identify no single case of a direct permanent adverse clinical outcome related to low blood glucose. Soon after publication of the NICE SUGAR trial in 2009 and combined AACE/ ADA consensus statement (10) from that same year, we re-evaluated our own protocol. A multidisciplinary group involving of the sections of endocrinology, critical care medicine, surgery, pharmacy and nursing modified our existing IIP to target a somewhat higher BG range of 120-160 mg/dl. Our hope was that we would achieve a median BG toward the lower end of the ADA/AACE recommendation (140-180 mg/dl), as the consensus statement implied this to be potentially more advantageous. Our preliminary experience with the revised, more conservative Yale IIP follows.

Research Design and Methods Setting The Yale New Haven Hospital Medical ICU (MICU) consists of 28 beds in a 966 bed tertiary care referral center located in New Haven, Connecticut. Internal medicine residents, under the supervision of critical care fellows and board certified pulmonary and critical care physicians, as well as hospitalist physicians, care for these patients. The nurse-to-patient ratio in the MICU is 1:1 or 1:2. IIP Implementation & Overview Our MICU nursing staff was already skilled in the implementation of prior Yale IIPs. Accordingly, minimal in-servicing was required, as simple dosing adjustments made to achieve the new BG target range, preserved the overall protocol construct. The Yale IIP consists of three clearly delineated steps as shown in Figure 1 (IIP protocol). The nurse begins by identifying an ICU patient with two BG values greater than 180 mg/dl. Once an appropriate candidate is identified, the responsible physician is alerted, and if clinically indicated the 4

physician enters an order to begin the IIP. After the initial BG is determined, the bolus and initial infusion rate is obtained using a simple calculation outlined under ‘Getting Started’ (Figure 1). Hourly BG measurement, usually via bedside capillary sampling and point-of-care meters begins. The current BG level corresponds to one of four columns in Step 1. Next, the nurse determines the change in rate by subtracting the current blood glucose value from the prior one. This leads to a specific cell within the column. The nurse then moves to the far right of the identified cell for specific instructions in Step 2. The actual changes to the insulin infusion rate are outlined in the lower table under Step 3. Depending on the stability of BG readings, the protocol allows for less frequent monitoring over time. Data Collection Methods Data from the first 115 insulin infusions initiated were consecutively recorded. A patient in whom the IIP protocol was restarted after 72 hours for recurrent hyperglycemia, or a patient in whom the protocol was resumed after ICU reission was treated as a new infusion. Seventeen patients were placed on the drip more than once. Data were collected prospectively by a single abstracter (SS), using a standardized abstraction form. Baseline (ission) variables collected included age, sex, race, height, weight, ission diagnosis, history of diabetes and pre-ICU insulin or oral anti-hyperglycemic drug use (Table 1). An APACHE II score was calculated for each patient, and the need for ventilator assistance was noted. All BG values, insulin doses, nutritional including intravenous dextrose infusions, caloric values for enteral and parenteral nutrition, use of vasopressors, corticosteroids and hemodialysis or continuous veno-venous hemodialysis (CVVHD) were collected from the hospital record. Patients were followed until 72 hours after MICU discharge. Statistical Analysis Data presented are expressed as means ± SD (or when not normally distributed, as medians with inter-quartile ranges (IQR)), or as a percentage.

5

Results Patients Data collection occurred from September 2009 to January 2010. The new IIP was used 115 times in 90 patients. Baseline characteristics of the patients are shown in Table 1. 64.4% of the patients had known diabetes. 44.4% were on insulin prior to hospital ission, while 24.4% were using oral hypoglycemics. The most frequent itting diagnosis was acute respiratory failure. The mean APACHE II score was 24.4 ± 7.5 and the mean overall length of ICU stay was 19.5 ± 24.8 days (median 10 days, 419.5, interquartile range (IQR)) indicating an extremely ill cohort of patients. Although 20% of patients were never intubated, those who required mechanical ventilation spent an average of 22.9 ± 27.4 days (median 10.5 days, 5-28.25 IQR) on the ventilator. The overall hospital length of stay was 36.4 ± 30.6 days (median 25 days, 14-59.5, IQR). The study cohort spent 10,874 hours receiving insulin infusion, with 8272 BG measurements taken during that time. The estimated average time interval between BG checks was 1.31 hours. Glycemic Control The mean baseline pre-infusion BG level was 306.1 ± 89.8 mg/dl (Table 2). The mean time taken to reach target BG was 8.3 ± 5.7 hours. Once target levels were reached, the IIP effectively maintained stable glycemic control. The mean BG value after the target was achieved was 155.9 ± 22.9 mg/dl (median 150 mg/dl, IQR 127-180). Once the target was reached, the IIP maintained 42% of BGs within the target range and 76% of BGs <180 mg/dl (Figure 2). The mean insulin infusion rate required was 3.9 ± 2.1 units/hour, The median number of hours patients remained on the infusion was 59 hours, IQR 25127 with 41 (36%) still running at 72 hours. Mean BG concentration during the 72 hour period after drip cessation climbed to 183.5 ± 48.0 mg/dl, when excluding those individuals who were made ‘comfort care only’ after transfer from the ICU.

6

Rates of hypoglycemia were low (Table 3). Severe hypoglycemia (BG<40 mg/dl) occurred in just 1 out of every 5000 BG determinations. Of the 2 episodes of severe hypoglycemia recorded, one appeared to result from inappropriate continuation of an insulin infusion rate that required reduction, based on the previous hour’s BG reading. In all cases, hypoglycemia was rapidly corrected using intravenous dextrose and no deleterious effects were observed.

Comparison with 2003 and 2005 IIPs Our revised IIP (target 120 – 160 mg/dl) is based on the original Yale IIP (2003, target 100 – 140 mg/dl), later modified in 2005 (target 90 – 120 mg/dl). Table 3 shows a comparison of the glycemic outcomes with the three IIPs. The median time spent on the insulin infusion is similar in the patient population we surveyed as in the 2 earlier IIPs introduced in the MICU. The baseline glucose level, when compared with the patients examined in the 2003 and 2005 IIPs, is higher, likely in part explained by the higher severity of illness of the current cohort. Not unexpectedly, hypoglycemia is less common with the higher target, suggesting that it is affording a larger margin of safety, a specified goal of the 2009 AACEADA recommendations.

Discussion The importance of avoiding hyperglycemia in the critically ill has been well established(1, 11-13) . The Leuven study in 2001(1) reported a 42% relative reduction in mortality with intensive insulin therapy aimed at maintaining BG in the 80-110mg/dl range when compared to conventional therapy (180200 mg/dl), among 1548 patients in the surgical ICU. Following this, the same investigators implemented a similar study in their MICU(14) . Although their results revealed no reduction in mortality overall, a subgroup of patients who required 3 or more days of critical care experienced considerable benefit. Subsequently, however, multicenter studies have not been able to reproduce these findings,(2-4) and have 7

had unacceptably high rates of hypoglycemia. The most recent and largest trial to date, NICE-SUGAR, randomly assigned intensive glycemic control (targeting BG 81-108mg/dl) or moderate glycemic control (targeting BG 140-180 mg/dl) in 6104 mixed medical-surgical ICU patients. Surprisingly, the investigators revealed a significantly increased mortality (odds ratio 1.14, confidence interval 1.02-1.28, p=0.02) in the group undergoing intensive insulin therapy. There remains, accordingly, substantial clinical equipoise regarding the expected benefits of tight glycemic control in the critical care setting and emerging concerns about the added risks of insulin-induced hypoglycemia. In response to these findings, the AACE/ADA released a consensus statement(10) that recommended maintaining BG levels in a moderated range of 140 – 180 mg/dl in critically ill patients, while also implying the potential for greater benefits toward the lower end of this range. In response to these national guidelines, we adapted our previously successful IIP to a target of 120 – 160 mg/dl. Although our rates of hypoglycemia have been historically low even with the lower targets, mounting evidence about the potential risks of hypoglycemia in the ICU prompted us to bring our target in line with current national guidelines. The ease of use of our IIP allowed for successful implementation by a busy MICU nursing staff with minimum expert supervision as demonstrated in a previous report(6) . The modified IIP was successful in facilitating and maintaining stable glycemic control within the targeted range of 120 – 160 mg/dl while maintaining very low rates of hypoglycemia. This is particularly reassuring, as many of our critically ill patients had risk factors for hypoglycemia as previously identified, such as need for vasopressor and renal replacement therapy, prior history of diabetes, and sepsis (15). Additionally, our rates of hypoglycemia are significantly lower than other published protocols and similar to those observed in the conventional/standard treatment arms of most randomized clinical trials of intensive insulin therapy in the critical care setting (1-4,6,14,16-21). We acknowledge that our work has a few limitations. First, the study was conceived as a quality improvement project aimed at ensuring the safety and efficacy of the Yale IIP, as we adjusted for new targets in line with the 2009 AACE-ADA consensus statement. We do not know whether our target of 8

120-160 mg/dl would yield similar morbidity and mortality outcomes as the 140-180mg/dl target utilized in the NICE-SUGAR study, but our median BG value after reaching target (150 mg/dl) is similar to the time weighted mean BG reported in the NICE-SUGAR conventional treatment cohort (144 ± 23mg/dl). Indeed, our study was not aimed at assessing clinical outcomes beyond biochemical results. We also do not have a randomized comparison group, given the nature of our investigation. In summary, this study describes our experience with the implementation of a safe, effective IIP with revised targets. Our IIP protocol is well established. To our knowledge there are no published data regarding safety and efficacy of IIP in a real world, non-research setting in the post-NICE SUGAR era, that is in line with the latest national recommendations for glycemic control in critically ill patients. Our revised protocol has low rates of hypoglycemia, was effective at maintaining good glycemic control and was readily accepted and easily implemented by the MICU nursing staff. The revised IIP could be easily used by other hospitals that are currently using the original Yale IIP and, with modest training, should be readily adopted by most ICUs with qualified critical care nurses and the monitoring capability needed to manage patients similar to our own. Further studies are needed to determine whether these new targets actually improve patient outcomes.

9

Table 1. Baseline variables, ission diagnosis and relevant clinical interventions employed for 89 patients who were placed on the revised Yale IIP. Characteristic

IIP Patients

n

115

Age (years) ±SD

62.4 ±14.4

Male sex

51.3%

Mean BMI (kg/m2) ±SD

31.8 ± 9.3

Race: Caucasian

65.2%

African American

16.5%

Hispanic

13.0%

Asian

5.2%

History of diabetes

64.4%

Use of insulin prior to ission

44.4%

Use of oral anti-hyperglycemic agents prior to ission

24.4%

Most frequent ission diagnoses: Acute respiratory failure

28.7%

Sepsis, pulmonary

25.2%

Bacterial pneumonia

11.3%

Acute Respiratory Distress Syndrome

7.8%

APACHE II score Mean ±SD

24.4 ± 7.5

Median

23 (19 -29, IQR)

Clinical Interventions: Corticosteroid therapy

60%

Vasopressor therapy

36.5%

Ventilator days (Mean)

11 (5 – 28, IQR) (20 % never intubated)

Hemodialysis

16.5%

CVVHD

9.6%

Mean caloric intake while on drip

1001 cals ± 711*

Parenteral nutrition

6.1%

*does not include caloric intake for 12 patients who were exclusively taking in an oral diet

10

Table 2: Results Pre-infusion BG (mg/dl)

Mean

306.1 ± 89.8

Median 309 (251 – 359, IQR) BG during infusion (mg/dl)

Mean

179.3±33.0

Median 166 (158 – 194, IQR) Nadir BG during infusion (mg/dl)

Mean

92.5 ± 26.2

Median 89 (80 – 101, IQR) Time to target (BG < 160 mg/dl) (hours)

Mean

8.3 ± 5.7

Median 7 ( 5 – 12, IQR) BG once target reached (mg/dl)

Mean

155.9 ± 22.9

Median 150 (127 – 180, IQR) Hours on infusion

Mean

94.6 ± 104.2

Median 59 (25 – 127, IQR) Infusion dose during infusion (units/hr)

Mean

3.9 ± 2.1

Median 3.5 (2.5 – 4.5, IQR) ICU length of stay (days)

Mean

19.5 ± 24.8

Median 10 (4 - 19, IQR) Hypoglycemia rate (% of BG measurements) < 70 mg/dl

0.30%

< 40 mg/dl

0.02%

BG during 72 hours after infusion

Mean

183.5 ± 48.0

discontinued (mg/dl)

11

Table 3. Comparison with 2003 and 2005 Yale IIPs

Yale Insulin Infusion Protocols, 2004-2011 Protocol

1

Year published Clinical setting

2004 Medical ICU

BG target (mg/dl) N (infusions) Hours on infusion (median (IQR)) Baseline BG (mg/dl) (±SD) Hours to target (median (IQR)) Mean BG* (mg/dl) (±SD)

100-140 69 61 (27-128)

Median BG* (mg/dl) (IQR)

3

2 2004 Cardiothoracic ICU 100-140 137 16 (12-27)

2011 Medical ICU

90-120 63 63 (28-133)

2005 Cardiothoracic ICU 90-120 54 15 (11-18)

2005 Medical ICU

120-160 115 59 (25-127)

299 ±96

218 ±53

238 ±76

189 ±44

306 ±90

9 (7-13)

5 (3-8)

6 (4-9)

7 (5-9)

7 (5-12)

119.9 ±28.6

112.0 ±21.8

155.9 ±22.9

118 (101-134)

110 (100-122)

150 (127-180)

129.5 ± 34.3

124 (107-146)

124.6 ± 31.0

121 (104-138)

% BGs <60

0.3%

0.2%

0.4%

0.3%

0.1%

% BGs <40

0.05%

0%

0.02%

0%

0.02%

*after target achieved

12

Figure Legends: Figure 1- Yale Insulin Infusion Protocol (IIP). Figure 2 - Performance of the updated IIP (data points represent the first 72 hours of insulin infusion). All BG levels are shown as medians (IQR).

13

References

1.

van den Berghe G, Wouters P, Weekers F, et al. Intensive insulin therapy in the critically ill patients. N Engl J Med. 2001;345:1359-1367.

2.

Preiser JC, Devos P, Ruiz-Santana S, et al. A prospective randomised multi-centre controlled trial on tight glucose control by intensive insulin therapy in adult intensive care units: the Glucontrol study. Intensive Care Med. 2009;35:1738-1748.

3.

Brunkhorst FM, Engel C, Bloos F, et al. Intensive insulin therapy and pentastarch resuscitation in severe sepsis. N Engl J Med. 2008;358:125-139.

4.

Finfer S, Chittock DR, Su SY, et al. Intensive versus conventional glucose control in critically ill patients. N Engl J Med. 2009;360:1283-1297.

5.

Inzucchi SE, Siegel MD. Glucose control in the ICU--how tight is too tight? N Engl J Med. 2009;360:1346-1349.

6.

Goldberg PA, Siegel MD, Sherwin RS, et al. Implementation of a safe and effective insulin infusion protocol in a medical intensive care unit. Diabetes Care. 2004;27:461-467.

7.

Faraon-Pogaceanu C, Banasiak KJ, Hirshberg EL, Faustino EV. Comparison of the effectiveness and safety of two insulin infusion protocols in the management of hyperglycemia in critically ill children. Pediatr Crit Care Med. 2010;11:741-749.

8.

Tamaki M, Shimizu T, Kanazawa A, et al. Efficacy and safety of modified Yale insulin infusion protocol in Japanese diabetic patients after open-heart surgery. Diabetes Res Clin Pract. 2008;81:296-302.

9.

PA G, MG R, Inzucchi SE: Clinical results of an updated insulin infusion protocol in critically ill patients. Diabetes Spectrum. 2005;18:188-191. 14

10.

Rodbard HW, Jellinger PS, Davidson JA, et al. Statement by an American Association of Clinical Endocrinologists/American College of Endocrinology consensus on type 2 diabetes mellitus: an algorithm for glycemic control. Endocr Pract. 2009;15:540-559.

11.

Malmberg K, Ryden L, Efendic S, Herlitz J, Nicol P, Waldenstrom A, Wedel H, Welin L. Randomized trial of insulin-glucose infusion followed by subcutaneous insulin treatment in diabetic patients with acute myocardial infarction (DIGAMI study): effects on mortality at 1 year. J Am Coll Cardiol. 1995;26:57-65.

12.

Malmberg K: Prospective randomised study of intensive insulin treatment on long term survival after acute myocardial infarction in patients with diabetes mellitus. DIGAMI (Diabetes Mellitus, Insulin Glucose Infusion in Acute Myocardial Infarction) Study Group. BMJ. 1997;314:15121515.

13.

Furnary AP, Zerr KJ, Grunkemeier GL, Starr A. Continuous intravenous insulin infusion reduces the incidence of deep sternal wound infection in diabetic patients after cardiac surgical procedures. Ann Thorac Surg. 1999;67:352-360; discussion 360-352.

14.

Van den Berghe G, Wilmer A, Hermans G, Meersseman W, Wouters PJ, Milants I, Van Wijngaerden E, Bobbaers H, Bouillon R: Intensive insulin therapy in the medical ICU. N Engl J Med. 2006;354:449-461.

15.

Vriesendorp TM, van Santen S, DeVries JH, et al. Predisposing factors for hypoglycemia in the intensive care unit. Crit Care Med. 2006;34:96-101.

16.

Goldberg PA, Sakharova OV, Barrett PW, et al. Improving glycemic control in the cardiothoracic intensive care unit: clinical experience in two hospital settings. J Cardiothorac Vasc Anesth. 2004;18:690-697.

17.

Hemmila MR, Taddonio MA, Arbabi S, Maggio PM, Wahl WL. Intensive insulin therapy is associated with reduced infectious complications in burn patients. Surgery. 2008;144:629-635; discussion 635-627.

15

18.

Cao SG, Ren JA, Shen B, Chen D, Zhou YB, Li JS. Intensive versus conventional insulin therapy in type 2 diabetes patients undergoing D2 gastrectomy for gastric cancer: a randomized controlled trial. World J Surg. 2011;35:85-92.

19.

Eachempati SR, Hydo LJ, Shou J, Barie PS. Implementation of tight glucose control for critically ill surgical patients: a process improvement analysis. Surg Infect (Larchmt). 2009;10:523-531.

20.

Arabi YM, Dabbagh OC, Tamim HM, et al. Intensive versus conventional insulin therapy: a randomized controlled trial in medical and surgical critically ill patients. Crit Care Med. 2008;36:3190-3197.

21.

De La Rosa Gdel C, Donado JH, Restrepo AH, et al. Strict glycaemic control in patients hospitalised in a mixed medical and surgical intensive care unit: a randomised clinical trial. Crit Care. 2008;12:R120.

16

FINAL 3-3-11

Yale-New Haven Hospital

ICU Insulin Infusion Protocol (IIP) for Adults The following IIP is intended for use in hyperglycemic adult patients in the ICU, adapted from our earlier protocols, in keeping with the latest glucose guidelines from national organizations. It should NOT be used in diabetic ketoacidosis (DKA) or hyperosmolar hyperglycemic state (HHS), as these patients may require higher initial insulin doses, IV dextrose at some point, and important adjunctive therapies for their fluid/acid-base/electrolyte/divalent status. (See 'DKA Guidelines' in YNHH Clinical Practice Manual (M) for further instructions.) In any patient with BG >500 mg/dL, the initial orders should also be carefully reviewed with the MD, since a higher initial insulin dose and additional monitoring/therapy may be required. If the patient’s response to the insulin infusion is at any time unusual or unexpected, or if any situation arises that is not adequately addressed by this protocol, the MD must be ed for assessment and further orders.

Getting Started 1.) PATIENT SELECTION: Begin IIP in any ICU patient with more than 2 BGs >180 mg/dl who is not expected to rapidly normalize their glycemic status. Patients who are eating (see #9 below); transferring out of ICU imminently (<24 hrs); or pre-terminal or being considered for CMO status are generally not appropriate candidates for this IIP. 2.) TARGET BLOOD GLUCOSE (BG) RANGE: .120-160 mg/dL. 3.) ORDERS: MD order required for use in the ICU. 4.) INSULIN INFUSION SOLUTION: Obtain from pharmacy (1 unit Regular Human Insulin / 1 cc 0.9 % NaCl). 5.) PRIMING: Before connecting, flush 20 cc infusion through all tubing. 6.) ISTRATION: Via infusion pump in 0.5 units/hr increments. 7.) BOLUS & INITIAL INFUSION RATE: Divide initial BG level by 100, then round to nearest 0.5 units for bolus AND initial infusion rate. Examples: 1.) Initial BG = 325 mg/dL: 325 ÷ 100 = 3.25, round ↑ to 3.5: IV bolus 3.5 units + start infusion @ 3.5 units/hr. 2.) Initial BG = 274 mg/dL: 274 ÷ 100 = 2.74, round ↓ to 2.5: IV bolus 2.5 units + start infusion @ 2.5 units/hr. 8.) CAUTION: If enteral/parenteral (TPN, PPN, Tube feeds) nutrition abruptly stopped, reduce infusion rate by 50%. 9.) Patients requiring IV insulin are usually NPO. In the rare patient who is eating, consider giving SQ Aspart PC to ‘cover’ the meal (ister 1 unit /15 grams carbohydrates consumed (usual dose 3-6 units.) In this circumstance don’t increase infusion rate during the first 3 hrs PC. 10.) Patients with T1DM, insulin-requiring T2DM, and those requiring >1 unit/hr should be transitioned to SQ insulin prior to discharge from ICU. BG Monitoring While on infusion, use glucose meter to check BG hourly. Once stable (3 consecutive values in target range), may reduce checks to q 2 hr. If stable for 12-24 hrs, may space checks to q 4 hr. Resume hourly checks until stable again if: any BG out of range; any change in insulin infusion rate; any significant change in clinical condition; initiation/discontinuation of steroids, pressors, TPN/PPN/tube feeds, dialysis, CVVH, or CAVH. In patients who are vasoconstricted/hypotensive, capillary BG (i.e., fingersticks) may be inaccurate; venous or arterial blood is preferred in this setting. Adjusting Infusion Rate If BG < 50 mg/dL: .D/C INSULIN INFUSION. & ister 1 amp (25 g) D50 IV; recheck BG q 15 minutes until ≥90 mg/dl. Ã Then, recheck BG q 1 hr; when ≥140 mg/dL, wait 30 min, restart insulin infusion at 50% of most recent rate If BG 50-74 mg/dL: .D/C INSULIN INFUSION. & ister 1/2 Amp (12.5 g) D50 IV; recheck BG q 15 minutes until ≥90 mg/dl. Ã Then, recheck BG q 1 hr; when ≥140 mg/dL, wait 30 min, then restart infusion at 50% of most recent rate. If BG 75-99 mg/dL: .D/C INSULIN INFUSION.. Recheck BG q 15 minutes until BG reaches or remains ≥90 mg/dl. Ã Then, recheck BG q 1 hr; when ≥140 mg/dL, wait 30 min, then restart infusion at 75% of most recent rate.

2

If BG ≥ 100 mg/dL:

STEP 1:

Determine the CURRENT BG LEVEL - identifies a COLUMN in the table:

BG 100-119 mg/dL

STEP 2:

BG 120-159 mg/dL

BG 160-199 mg/dL

BG ≥ 200 mg/dL

Determine the RATE OF CHANGE from the prior BG level - identifies a CELL in the table - Then move right for INSTRUCTIONS: [Note: If the last BG was measured 2 or more hrs before the current BG, calculate the hourly rate of change. Example: If the BG at 2PM was 150 mg/dL and the BG at 4PM is 120 mg/dL, the total change over 2 hours is -30 mg/dL; however, the hourly change is –30 mg/dL ÷ 2 hours = -15 mg/dL/hr.]

BG 100-119 mg/dL

BG 120-159 mg/dL

BG ↑ by > 40 mg/dL/hr BG ↑ BG UNCHANGED OR

BG ↓ by 1-20 mg/dL/hr BG ↓ by > 20 mg/dL/hr † see below

BG 160-199 mg/dL

BG ≥ 200 mg/dL

INSTRUCTIONS*

BG ↑ by > 60 mg/dL/hr

BG ↑

↑ INFUSION by “2Δ”

BG ↑ by 1-60 mg/dL/hr

BG UNCHANGED

OR

OR

↑ INFUSION by “Δ”

BG UNCHANGED

BG ↑ by 1-40 mg/dL/hr, BG UNCHANGED, OR BG ↓ by 1-20 mg/dL/hr

BG ↓ by 1-20 mg/dL/hr

BG ↓ by 1-40 mg/dL/hr

BG ↓ by 21-60 mg/dL/hr

NO INFUSION CHANGE

BG ↓ by 21-40 mg/dL/hr

BG ↓ by 41-60 mg/dL/hr

BG ↓ by 61-80 mg/dL/hr

↓ INFUSION by “Δ”

BG ↓ by > 40 mg/dL/hr

BG ↓ by > 60 mg/dL/hr

BG ↓ by > 80 mg/dL/hr

HOLD x 30 min, then ↓ INFUSION by “2Δ”

†

D/C INSULIN INFUSION; √BG in 15 min to be sure ≥90 mg/dl. Then recheck BG q 1 hr; when ≥140 mg/dl, restart infusion @75% of most recent rate.

STEP 3: CHANGES IN INFUSION RATE* (“Δ”) are determined by the current rate: Current Rate (Units/hr) < 3.0 3.0 – 6.0 6.5 – 9.5 10.0 – 14.5 15 – 19.5 ≥ 20*

Δ = Rate Change (Units/hr) 0.5 1 1.5 2 3* 4*

2Δ = 2X Rate Change (Units/hr) 1 2 3 4 6* 8*

© Yale Diabetes Center & Yale-New Haven Hospital (July 2009, revised 8/30/10, 11/18/10, 1/3/11)