A 63‐year‐old female with a history of essential thrombocythemia and hypertension presented with a 4‐week history of a worsening ulcer on her right second digit. Initially, the patient attributed the wound to a dog scratch but sought further treatment at an outside clinic when she did not see improvement. She was given a diagnosis of cellulitis and was treated with unknown oral antibiotics and silvadene cream. The ulcer continued to worsen and the patient presented to our hospital. On physical exam, an 8 cm 3 cm ulcer was observed on the right second digit. It had violaceous rolled up borders, granulation tissue, fibrinous exudates, and areas of necrotic tissue (Figure 1). The remainder of the physical examination was unremarkable. Initial laboratory values included hemoglobin 12.5 gm/dL, white blood cell count 31.2 K/UL, and platelets 625 gm/dL. An x‐ray of the hand showed soft tissue swelling with no evidence of osteomyelitis. The ulcer was evaluated and treated as an infected wound. The patient was started on broad spectrum intravenous antibiotics and underwent excisional debridement with biopsy. Blood and wound cultures were negative for aerobic and anaerobic bacteria, fungi, and acid‐fast bacilli. Pathology from the biopsy showed extensive necrosis and acute inflammation. The patient was discharged home with 10 days of oral antibiotics, and instructions for wound care. Upon follow‐up 1 week later, the patient complained of intense pain and worsening of the ulcer prompting readmission. Dermatology was consulted and diagnosed pyoderma gangrenosum (PG). The patient was started on prednisone, 60 mg daily and azathioprine, 50 mg daily. The ulcer slowly improved (Figure 2) and the steroid dosage was tapered. She was finally discharged home with a 6‐week taper of prednisone, azathioprine, and home health consultation for assistance with wound care.0, 0

Figure 1

Figure 2

PG is an ulcerative neutrophilic dermatosis. In up to 50% of cases, PG is associated with either inflammatory bowel disease, collagen vascular disease, or hematologic disorders.1 Although an immune‐modulated pathway may be involved, the etiology and pathophysiology of PG is still unknown.1 Furthermore, PG is a diagnosis of exclusion.1 However, PG does have clinical findings which favor the diagnosis. There are 4 main subtypes of PG; ulcerative or classic, pustular, bullous, and vegetative.1 Although myeloproliferative disorders are more specifically associated with the bullous form, our patient presented with the classic subtype.2, 3 In the classic subtype, patients will often describe an initial pustule which then necroses, forming an ulcer with a reddish/purple or gray undermined border and a red halo surrounding the ulcer. PG can occur anywhere on the body however it is more frequently seen on the legs. A clinically relevant feature of PG, emphasized in this case, is pathergy. Thus, PG can develop or worsen secondary to mild trauma. PG has been reported to form after mild trauma such as an insect bite or dog scratch and has been documented to worsen with debridement, skin grafting, and biopsies.1 Another feature and clinical clue of PG as manifested by our patient is intense pain. The skin biopsy, however, is usually nonspecific and can reveal findings which include edema, neutrophil infiltration, abscess formation, necrosis, and thrombosis of vessels.1 In patients with PG associated with myeloproliferative syndromes, no correlation has been shown between the time of diagnosis and the severity of the underlying myeloproliferative syndrome.2, 3 Treatment for PG depends on extent of involvement and association with underlying disease and can include local, oral, or intravenous corticosteroids, immunosuppressants, appropriate wound care, and treatment of associated disease.4

PG is a diagnosis of exclusion. Underlying infection, vasculitis, malignancy, and Sweet's syndrome should be considered in the differential. However, one must consider PG in the differential diagnosis of an ulcer in a patient with an underlying predisposing illness, when the ulcer has characteristics of pathergy and intense pain, and is not healing appropriately as illustrated in this case.

A 63‐year‐old female with a history of essential thrombocythemia and hypertension presented with a 4‐week history of a worsening ulcer on her right second digit. Initially, the patient attributed the wound to a dog scratch but sought further treatment at an outside clinic when she did not see improvement. She was given a diagnosis of cellulitis and was treated with unknown oral antibiotics and silvadene cream. The ulcer continued to worsen and the patient presented to our hospital. On physical exam, an 8 cm 3 cm ulcer was observed on the right second digit. It had violaceous rolled up borders, granulation tissue, fibrinous exudates, and areas of necrotic tissue (Figure 1). The remainder of the physical examination was unremarkable. Initial laboratory values included hemoglobin 12.5 gm/dL, white blood cell count 31.2 K/UL, and platelets 625 gm/dL. An x‐ray of the hand showed soft tissue swelling with no evidence of osteomyelitis. The ulcer was evaluated and treated as an infected wound. The patient was started on broad spectrum intravenous antibiotics and underwent excisional debridement with biopsy. Blood and wound cultures were negative for aerobic and anaerobic bacteria, fungi, and acid‐fast bacilli. Pathology from the biopsy showed extensive necrosis and acute inflammation. The patient was discharged home with 10 days of oral antibiotics, and instructions for wound care. Upon follow‐up 1 week later, the patient complained of intense pain and worsening of the ulcer prompting readmission. Dermatology was consulted and diagnosed pyoderma gangrenosum (PG). The patient was started on prednisone, 60 mg daily and azathioprine, 50 mg daily. The ulcer slowly improved (Figure 2) and the steroid dosage was tapered. She was finally discharged home with a 6‐week taper of prednisone, azathioprine, and home health consultation for assistance with wound care.0, 0

Figure 1

Figure 2

PG is an ulcerative neutrophilic dermatosis. In up to 50% of cases, PG is associated with either inflammatory bowel disease, collagen vascular disease, or hematologic disorders.1 Although an immune‐modulated pathway may be involved, the etiology and pathophysiology of PG is still unknown.1 Furthermore, PG is a diagnosis of exclusion.1 However, PG does have clinical findings which favor the diagnosis. There are 4 main subtypes of PG; ulcerative or classic, pustular, bullous, and vegetative.1 Although myeloproliferative disorders are more specifically associated with the bullous form, our patient presented with the classic subtype.2, 3 In the classic subtype, patients will often describe an initial pustule which then necroses, forming an ulcer with a reddish/purple or gray undermined border and a red halo surrounding the ulcer. PG can occur anywhere on the body however it is more frequently seen on the legs. A clinically relevant feature of PG, emphasized in this case, is pathergy. Thus, PG can develop or worsen secondary to mild trauma. PG has been reported to form after mild trauma such as an insect bite or dog scratch and has been documented to worsen with debridement, skin grafting, and biopsies.1 Another feature and clinical clue of PG as manifested by our patient is intense pain. The skin biopsy, however, is usually nonspecific and can reveal findings which include edema, neutrophil infiltration, abscess formation, necrosis, and thrombosis of vessels.1 In patients with PG associated with myeloproliferative syndromes, no correlation has been shown between the time of diagnosis and the severity of the underlying myeloproliferative syndrome.2, 3 Treatment for PG depends on extent of involvement and association with underlying disease and can include local, oral, or intravenous corticosteroids, immunosuppressants, appropriate wound care, and treatment of associated disease.4

PG is a diagnosis of exclusion. Underlying infection, vasculitis, malignancy, and Sweet's syndrome should be considered in the differential. However, one must consider PG in the differential diagnosis of an ulcer in a patient with an underlying predisposing illness, when the ulcer has characteristics of pathergy and intense pain, and is not healing appropriately as illustrated in this case.

A 63‐year‐old female with a history of essential thrombocythemia and hypertension presented with a 4‐week history of a worsening ulcer on her right second digit. Initially, the patient attributed the wound to a dog scratch but sought further treatment at an outside clinic when she did not see improvement. She was given a diagnosis of cellulitis and was treated with unknown oral antibiotics and silvadene cream. The ulcer continued to worsen and the patient presented to our hospital. On physical exam, an 8 cm 3 cm ulcer was observed on the right second digit. It had violaceous rolled up borders, granulation tissue, fibrinous exudates, and areas of necrotic tissue (Figure 1). The remainder of the physical examination was unremarkable. Initial laboratory values included hemoglobin 12.5 gm/dL, white blood cell count 31.2 K/UL, and platelets 625 gm/dL. An x‐ray of the hand showed soft tissue swelling with no evidence of osteomyelitis. The ulcer was evaluated and treated as an infected wound. The patient was started on broad spectrum intravenous antibiotics and underwent excisional debridement with biopsy. Blood and wound cultures were negative for aerobic and anaerobic bacteria, fungi, and acid‐fast bacilli. Pathology from the biopsy showed extensive necrosis and acute inflammation. The patient was discharged home with 10 days of oral antibiotics, and instructions for wound care. Upon follow‐up 1 week later, the patient complained of intense pain and worsening of the ulcer prompting readmission. Dermatology was consulted and diagnosed pyoderma gangrenosum (PG). The patient was started on prednisone, 60 mg daily and azathioprine, 50 mg daily. The ulcer slowly improved (Figure 2) and the steroid dosage was tapered. She was finally discharged home with a 6‐week taper of prednisone, azathioprine, and home health consultation for assistance with wound care.0, 0

Figure 1

Figure 2

PG is an ulcerative neutrophilic dermatosis. In up to 50% of cases, PG is associated with either inflammatory bowel disease, collagen vascular disease, or hematologic disorders.1 Although an immune‐modulated pathway may be involved, the etiology and pathophysiology of PG is still unknown.1 Furthermore, PG is a diagnosis of exclusion.1 However, PG does have clinical findings which favor the diagnosis. There are 4 main subtypes of PG; ulcerative or classic, pustular, bullous, and vegetative.1 Although myeloproliferative disorders are more specifically associated with the bullous form, our patient presented with the classic subtype.2, 3 In the classic subtype, patients will often describe an initial pustule which then necroses, forming an ulcer with a reddish/purple or gray undermined border and a red halo surrounding the ulcer. PG can occur anywhere on the body however it is more frequently seen on the legs. A clinically relevant feature of PG, emphasized in this case, is pathergy. Thus, PG can develop or worsen secondary to mild trauma. PG has been reported to form after mild trauma such as an insect bite or dog scratch and has been documented to worsen with debridement, skin grafting, and biopsies.1 Another feature and clinical clue of PG as manifested by our patient is intense pain. The skin biopsy, however, is usually nonspecific and can reveal findings which include edema, neutrophil infiltration, abscess formation, necrosis, and thrombosis of vessels.1 In patients with PG associated with myeloproliferative syndromes, no correlation has been shown between the time of diagnosis and the severity of the underlying myeloproliferative syndrome.2, 3 Treatment for PG depends on extent of involvement and association with underlying disease and can include local, oral, or intravenous corticosteroids, immunosuppressants, appropriate wound care, and treatment of associated disease.4

PG is a diagnosis of exclusion. Underlying infection, vasculitis, malignancy, and Sweet's syndrome should be considered in the differential. However, one must consider PG in the differential diagnosis of an ulcer in a patient with an underlying predisposing illness, when the ulcer has characteristics of pathergy and intense pain, and is not healing appropriately as illustrated in this case.

Observational studies in hospitalized patients with and without diabetes indicate that hyperglycemia is a predictor of poor clinical outcome and mortality.14 Early randomized controlled trials of intensified insulin therapy in patients with surgical and medical acute critical illness reported a reduction on the risk of multiorgan failure and systemic infections,35 as well as short‐ and long‐term mortality.1, 4 Recent randomized controlled trials, however, have failed to confirm the previously suggested benefits of intensive glucose control,6 and the large multicenter normoglycaemia in intensive care evaluation and survival using glucose algorithm regulation (NICE‐SUGAR) study reported an absolute increase in mortality rate with intensive glucose control.7 In addition, intensified insulin therapy in critically‐ill patients has been shown to be associated with a higher rate of severe hypoglycemic events than less aggressive glycemic control protocols.710 These results have led to a heightened interest in improving the quality and safety of the management of diabetes and hyperglycemia in the hospital.

The use of intravenous continuous insulin infusion (CII) is the preferred route of insulin administration for the management of hyperglycemia in the critical care setting.1, 11 Numerous examples of successful CII algorithms in achieving glycemic control are reported in the literature.4, 5, 12 Traditionally, order forms to titrate drip to achieve a target blood glucose (BG) range using an established algorithm or by the application of mathematical rules have been used in clinical practice. Recently, computer‐based algorithms aiming to direct the nursing staff adjusting insulin infusion rate have become commercially available.13, 14 It is not known, however, if computer‐based algorithms are superior to standard paper form‐based protocols in achieving glucose control and in reducing hypoglycemic events in critically‐ill patients. Accordingly, this multicenter randomized study aimed to determine differences in glycemic control and hypoglycemic events between treatment with a computer‐guided CII device and a standard column‐based paper algorithm in critically‐ill patients in the medical intensive care unit (ICU).

Research Design and Methods

In this multicenter, prospective, open‐label randomized study, 160 adult patients admitted to a medical ICU with new hyperglycemia or with a known history of diabetes treated with diet, insulin therapy or with any combination of oral antidiabetic agents were enrolled after written informed consent had been obtained from the patient or closest family member (Figure 1). Patients with known history of diabetes had 2 BG readings >120 mg/dL while subjects without a history of diabetes had 2 BG readings >140 mg/dL prior to enrollment. We excluded patients with acute hyperglycemic crises such as diabetic ketoacidosis (DKA) and hyperosmolar hyperglycemic state,15 patients with severely impaired renal function (serum creatinine 3.5 mg/dL), dementia, and pregnancy. This study was conducted at 4 hospital centers including Grady Memorial Hospital, Emory University Hospital, and Piedmont Hospital in Atlanta, Georgia and the Regional Medical Center in Memphis, Tennessee.

Figure 1

Figure 2

Patients were randomized using a computer randomization table to receive CII following a computer‐guided algorithm (Glucommander) or CII following a standard paper form insulin infusion algorithm. Both protocols used glulisine (Apidra) insulin and targeted a BG between 80 mg/dL and 120 mg/dL. Insulin management was directed by the specific assigned protocol and was carried out daily by the nursing staff and by members of the internal medicine residency program. The ICU physician and primary care team decided on the treatment for all other medical problem(s) for which patients were admitted. Data were collected during CII up to the first 10 days of ICU stay.

Results

The admission characteristics and clinical outcomes of interest of the study patients are shown in Table 1. A total of 160 adult patients admitted to a medical ICU with new hyperglycemia (47%) or with a known history of diabetes (53%) were randomized into the study. Of them, 7 patients were excluded due to withdrawal of consent, treatment with subcutaneous basal or supplemental short‐acting insulin during CII, or receiving less than 4 hours of CII. There were no differences in the mean age, gender, race, history of diabetes, or primary admitting diagnosis between treatment groups. The most common admitting diagnosis categories included pulmonary (22.1%), cardiovascular (21.4%), infectious (20.0%), and central nervous system (16.6%) disorders.

The mean admission glucose concentration for study patients was 190.6 58.2 mg/dL and the mean A1C was 7.0 1.7%. Glycemic control parameters achieved with the CII protocols are listed in Table 2. At the start of CII, the mean BG value was similar for the Glucommander and paper protocols (189.7 64.8 mg/dL and 188.4 54.8 mg/dL, P = 0.419). The mean time to reach the BG target was shorter in the Glucommander group (4.8 2.8 vs. 7.8 9.1 hours, P < 0.001). The Glucommander group had a lower mean glucose value during insulin infusion (115.5 20.7 vs. 131.0 24.6 mg/dL, P < 0.001) and once at target goal, in a lower mean BG values (103.3 8.8 vs. 117.3 16.5 mg/dL, P < 0.001) than the standard algorithm (Figure 2). The mean inpatient BG difference between treatment groups was 15.5 mg/dL (P < 0.001), with a mean daily BG difference ranging from 17.4 mg/dL to 24.4 mg/dL less for the Glucommander group during days 2 to 6 of therapy (P < 0.01).

The Glucommander algorithm was associated with tighter glycemic control and less glucose variability than the standard paper form protocol. Once patients achieved BG target, on average 71.1% of BG readings in the Glucommander and 51.3% in the standard group remained within the 80 mg/dL to 120 mg/dL target range (P < 0.001). In addition, the Glucommander was associated with a significantly lower rate of severe hyperglycemia during insulin infusion. The number of patients with 1 or more episodes of BG >200 mg/dL (11.7% vs. 25%, P = 0.057 before adjusting for potential site difference and P = 0.034 after adjusting for site difference) were less in the Glucommander group than in the standard paper regimen. In addition, 4 of these patients in spite of being on the highest insulin delivery column failed to achieve glucoses <180 and had an average in‐hospital glucose level of 204.5 32.2 mg/dL. These patients were transitioned to the Glucommander arm and withdrawn from the study. All episodes of hypoglycemia occurred after the patients achieved 1 glucose measurement within the target range. The number of patients who experienced one or more BG <40 mg/dL and <60 mg/dL was 3.9% and 42.9% in the Glucommander and 5.6% and 31.9% in the standard regimen, respectively (both, P = not significant [NS]). Similar results were obtained when site effect was accommodated (both, P = NS). Based on repeated measures analyses, the probabilities of BG reading <40 mg/dL or <60 mg/dL were not significantly different between groups (P = 0.969, P = 0.084) after accounting for within‐patient correlations with or without adjusting for time effect. None of these episodes resulted in seizures or were otherwise judged to be associated with deterioration of clinical status.

The mean insulin infusion rate was slightly higher in the Glucommander regimen but the difference was not statistically significant between groups. Patients treated with the Glucommander protocol received a mean infusion rate of 0.035 0.024 unit/kg/hour for a total of 2.85 1.93 units per hour, and those treated with the paper protocol received a 0.028 0.021 units/kg/hour for a total of 2.50 2.28 units per hour, P = 0.12 and P = 0.09, respectively.

The numbers of BG measurements were similar between the Glucommander and standard paper algorithms (44.2 39.8 and 41.2 34.5 respectively, P = NS) with the number of glucose testing per patient ranging from 6 to 175 in the Glucommander and 3 to 168 in the standard group. Similarly, when normalized to the duration of insulin infusion, the frequency of BG monitoring was not different with the protocols (0.68 0.18 and 0.62 0.22 tests/hour respectively, P = NS).

Compared to the standard paper insulin infusion algorithm, patients treated with the Glucommander device had a similar mean ICU LOS (13.4 13.8 vs. 8.5 7.6 days, P = 0.145), mean hospital LOS (17.5 15.0 days vs. 23.9 26.3 days, P = 0.704) and hospital mortality (26.0% vs. 21.9%, P = 0.561).

Discussion

This study is the first to compare the safety and efficacy of a CII via a computer‐guided algorithm and a standard paper form protocol in nonsurgical patients in the ICU. Both treatment algorithms resulted in significant improvement in glycemic control with the Glucommander achieving glycemic glucose target in a shorter time of treatment, a lower mean glucose concentration, and in greater percentage of glucose measurements maintained within target range, without an increased risk of severe hypoglycemia compared to the standard paper protocol.

Hyperglycemia in hospitalized patients is a common, serious, and costly health care problem. Evidence from observational and interventional studies indicate that hyperglycemia in critical illness is associated with an increased risk of complications and mortality.25 There is ongoing debate, however, about the optimal glucose level in hospitalized patients with critical illness. Although, several cohort studies as well as early randomized trials in ICU patients reported that intensified insulin treatment to achieve a target glucose between 80 mg/dL to 110 mg/dL reported a reduction in short‐term and long‐term mortality and rates of multiorgan failure and systemic infections compared with conventionally treated patients.3, 4, 17 More recent randomized controlled trials and meta‐analyses, however, have shown that this low BG target has been difficult to achieve without increasing the risk for severe hypoglycemia.710 In addition, recent multicenter trials have failed to show significant improvement in clinical outcome or have even shown increased mortality risk with intensive glycemic control.610 Based on these reports, the American Association of Clinical Endocrinologist (AACE) and American Diabetes Association (ADA) task force on inpatient glycemic control recommended different glycemic targets in the ICU setting. Current guidelines suggest targeting a BG level between 140 mg/dL and 180 mg/dL (7.8 and 10.0 mmol/L) for the majority of ICU patients and a lower glucose targets between 110 mg/dL and 140 mg/dL (6.1 and 7.8 mmol/L) in selected ICU patients (ie, centers with extensive experience and appropriate nursing support, cardiac surgical patients, patients with stable glycemic control without hypoglycemia). Glucose targets >180 mg/dL or <110 mg/dL are no longer recommended in ICU patients.

The rate of severe hypoglycemic events (<40 mg/dL) observed in both arms of our trial was significantly lower than those reported in recent international trials of intensive glycemic control.3, 4, 8 The overall rate of severe hypoglycemic events in international trials ranged between 5% to 28.6%.3, 4, 7, 8, 18, 19 In this trial, the number of patients with severe hypoglycemia was 3.9% in the computer‐based and 5.6% in the standard paper algorithm. Repeated measures analyses show the probabilities of BG readings <40 mg/dL were similar and not significantly different between groups (P = 0.969). We observed, however, a high rate of mild hypoglycemic events in patients treated with both insulin algorithms. The number of patients with BG <60 mg/dL was 42.9% in the Glucommander and 31.9% in the standard (P = NS). Minimizing the rate of hypoglycemia events is of major importance in hospitalized patients because it has been shown that hypoglycemia may be an independent risk factor of poor clinical outcome and mortality.20 Hypoglycemia may increase the risk of ventricular arrhythmias, in part due to the prolongation QT interval21 and can impair cerebral glucose metabolism resulting in brain metabolic dysfunction, as suggested by recent clinical studies.22 Moreover, insulin‐induced hypoglycemia is also associated with increased proinflammatory cytokines (tumor necrosis factor [TNF]‐alpha, interleukin [IL]‐1beta, IL‐6, and IL‐8) and oxidative stress23 that correlate with elevations of counterregulatory hormones (catecholamines, cortisol).

The Glucommander was associated with lower glycemic variability and with a higher percentage of BG readings within target range than patients treated with the standard paper form regimen. The clinical importance of the degree of variability and rapidity of fluctuations in glucose levels in critically ill patients is a topic of recent interest. Glycemic variability has been identified as a strong independent contributor to the risk of mortality in critically ill and surgical patients.24 Low levels of glycemic variability (standard deviation [SD] <10 mg/dL or 10‐20 mg/dL) have been shown to have a statistically significant lower risk of mortality, even after adjustment for severity of illness. Further studies are needed to determine benefits on clinical outcomes from the more consistent BG control from computer‐based titration protocols.

We acknowledge the following limitations in this multicenter open label study. First, this study was conducted in the medical ICU and excluded postsurgical patients and subjects expected to undergo a major surgical procedure during the hospital stay. Although a recent meta‐analysis9 of 26 studies involving 13,567 patients reported no benefits in the general ICU population, it found a favorable effect of intensive glycemic control on mortality in surgical ICU patients (relative risk [RR], 0.63; confidence interval [CI], 0.44‐0.91). We also excluded patients with severe renal insufficiency and patients with a history of hyperglycemic crises. In addition, our study was not powered to demonstrate differences in mortality or clinical outcome between treatment groups, and the BG targets used in this study were lower than glycemic targets recently recommended by the AACE and ADA inpatient glycemic control task force.25 Raising the BG targets is likely to reduce or prevent the rate of mild and severe hypoglycemic events in the ICU.

In conclusion, the computer‐guided algorithm resulted in a more rapid and tighter glycemic control with a similar rate of hypoglycemic events than the standard paper form protocol in medical ICU patients. Our study suggests that, both treatment algorithms are appropriate alternatives for the management of hyperglycemia in critically ill patients, and the choice depends on a physician's preferences, cost considerations, and the availability of the computer guided algorithm. Large randomized clinical trials are needed to test the impact of the new AACE/ADA recommended BG targets in reducing hypoglycemic events, hospital complications, and hospital mortality in critically ill patients in the ICU.

Observational studies in hospitalized patients with and without diabetes indicate that hyperglycemia is a predictor of poor clinical outcome and mortality.14 Early randomized controlled trials of intensified insulin therapy in patients with surgical and medical acute critical illness reported a reduction on the risk of multiorgan failure and systemic infections,35 as well as short‐ and long‐term mortality.1, 4 Recent randomized controlled trials, however, have failed to confirm the previously suggested benefits of intensive glucose control,6 and the large multicenter normoglycaemia in intensive care evaluation and survival using glucose algorithm regulation (NICE‐SUGAR) study reported an absolute increase in mortality rate with intensive glucose control.7 In addition, intensified insulin therapy in critically‐ill patients has been shown to be associated with a higher rate of severe hypoglycemic events than less aggressive glycemic control protocols.710 These results have led to a heightened interest in improving the quality and safety of the management of diabetes and hyperglycemia in the hospital.

The use of intravenous continuous insulin infusion (CII) is the preferred route of insulin administration for the management of hyperglycemia in the critical care setting.1, 11 Numerous examples of successful CII algorithms in achieving glycemic control are reported in the literature.4, 5, 12 Traditionally, order forms to titrate drip to achieve a target blood glucose (BG) range using an established algorithm or by the application of mathematical rules have been used in clinical practice. Recently, computer‐based algorithms aiming to direct the nursing staff adjusting insulin infusion rate have become commercially available.13, 14 It is not known, however, if computer‐based algorithms are superior to standard paper form‐based protocols in achieving glucose control and in reducing hypoglycemic events in critically‐ill patients. Accordingly, this multicenter randomized study aimed to determine differences in glycemic control and hypoglycemic events between treatment with a computer‐guided CII device and a standard column‐based paper algorithm in critically‐ill patients in the medical intensive care unit (ICU).

Research Design and Methods

In this multicenter, prospective, open‐label randomized study, 160 adult patients admitted to a medical ICU with new hyperglycemia or with a known history of diabetes treated with diet, insulin therapy or with any combination of oral antidiabetic agents were enrolled after written informed consent had been obtained from the patient or closest family member (Figure 1). Patients with known history of diabetes had 2 BG readings >120 mg/dL while subjects without a history of diabetes had 2 BG readings >140 mg/dL prior to enrollment. We excluded patients with acute hyperglycemic crises such as diabetic ketoacidosis (DKA) and hyperosmolar hyperglycemic state,15 patients with severely impaired renal function (serum creatinine 3.5 mg/dL), dementia, and pregnancy. This study was conducted at 4 hospital centers including Grady Memorial Hospital, Emory University Hospital, and Piedmont Hospital in Atlanta, Georgia and the Regional Medical Center in Memphis, Tennessee.

Figure 1

Figure 2

Patients were randomized using a computer randomization table to receive CII following a computer‐guided algorithm (Glucommander) or CII following a standard paper form insulin infusion algorithm. Both protocols used glulisine (Apidra) insulin and targeted a BG between 80 mg/dL and 120 mg/dL. Insulin management was directed by the specific assigned protocol and was carried out daily by the nursing staff and by members of the internal medicine residency program. The ICU physician and primary care team decided on the treatment for all other medical problem(s) for which patients were admitted. Data were collected during CII up to the first 10 days of ICU stay.

Results

The admission characteristics and clinical outcomes of interest of the study patients are shown in Table 1. A total of 160 adult patients admitted to a medical ICU with new hyperglycemia (47%) or with a known history of diabetes (53%) were randomized into the study. Of them, 7 patients were excluded due to withdrawal of consent, treatment with subcutaneous basal or supplemental short‐acting insulin during CII, or receiving less than 4 hours of CII. There were no differences in the mean age, gender, race, history of diabetes, or primary admitting diagnosis between treatment groups. The most common admitting diagnosis categories included pulmonary (22.1%), cardiovascular (21.4%), infectious (20.0%), and central nervous system (16.6%) disorders.

The mean admission glucose concentration for study patients was 190.6 58.2 mg/dL and the mean A1C was 7.0 1.7%. Glycemic control parameters achieved with the CII protocols are listed in Table 2. At the start of CII, the mean BG value was similar for the Glucommander and paper protocols (189.7 64.8 mg/dL and 188.4 54.8 mg/dL, P = 0.419). The mean time to reach the BG target was shorter in the Glucommander group (4.8 2.8 vs. 7.8 9.1 hours, P < 0.001). The Glucommander group had a lower mean glucose value during insulin infusion (115.5 20.7 vs. 131.0 24.6 mg/dL, P < 0.001) and once at target goal, in a lower mean BG values (103.3 8.8 vs. 117.3 16.5 mg/dL, P < 0.001) than the standard algorithm (Figure 2). The mean inpatient BG difference between treatment groups was 15.5 mg/dL (P < 0.001), with a mean daily BG difference ranging from 17.4 mg/dL to 24.4 mg/dL less for the Glucommander group during days 2 to 6 of therapy (P < 0.01).

The Glucommander algorithm was associated with tighter glycemic control and less glucose variability than the standard paper form protocol. Once patients achieved BG target, on average 71.1% of BG readings in the Glucommander and 51.3% in the standard group remained within the 80 mg/dL to 120 mg/dL target range (P < 0.001). In addition, the Glucommander was associated with a significantly lower rate of severe hyperglycemia during insulin infusion. The number of patients with 1 or more episodes of BG >200 mg/dL (11.7% vs. 25%, P = 0.057 before adjusting for potential site difference and P = 0.034 after adjusting for site difference) were less in the Glucommander group than in the standard paper regimen. In addition, 4 of these patients in spite of being on the highest insulin delivery column failed to achieve glucoses <180 and had an average in‐hospital glucose level of 204.5 32.2 mg/dL. These patients were transitioned to the Glucommander arm and withdrawn from the study. All episodes of hypoglycemia occurred after the patients achieved 1 glucose measurement within the target range. The number of patients who experienced one or more BG <40 mg/dL and <60 mg/dL was 3.9% and 42.9% in the Glucommander and 5.6% and 31.9% in the standard regimen, respectively (both, P = not significant [NS]). Similar results were obtained when site effect was accommodated (both, P = NS). Based on repeated measures analyses, the probabilities of BG reading <40 mg/dL or <60 mg/dL were not significantly different between groups (P = 0.969, P = 0.084) after accounting for within‐patient correlations with or without adjusting for time effect. None of these episodes resulted in seizures or were otherwise judged to be associated with deterioration of clinical status.

The mean insulin infusion rate was slightly higher in the Glucommander regimen but the difference was not statistically significant between groups. Patients treated with the Glucommander protocol received a mean infusion rate of 0.035 0.024 unit/kg/hour for a total of 2.85 1.93 units per hour, and those treated with the paper protocol received a 0.028 0.021 units/kg/hour for a total of 2.50 2.28 units per hour, P = 0.12 and P = 0.09, respectively.

The numbers of BG measurements were similar between the Glucommander and standard paper algorithms (44.2 39.8 and 41.2 34.5 respectively, P = NS) with the number of glucose testing per patient ranging from 6 to 175 in the Glucommander and 3 to 168 in the standard group. Similarly, when normalized to the duration of insulin infusion, the frequency of BG monitoring was not different with the protocols (0.68 0.18 and 0.62 0.22 tests/hour respectively, P = NS).

Compared to the standard paper insulin infusion algorithm, patients treated with the Glucommander device had a similar mean ICU LOS (13.4 13.8 vs. 8.5 7.6 days, P = 0.145), mean hospital LOS (17.5 15.0 days vs. 23.9 26.3 days, P = 0.704) and hospital mortality (26.0% vs. 21.9%, P = 0.561).

Discussion

This study is the first to compare the safety and efficacy of a CII via a computer‐guided algorithm and a standard paper form protocol in nonsurgical patients in the ICU. Both treatment algorithms resulted in significant improvement in glycemic control with the Glucommander achieving glycemic glucose target in a shorter time of treatment, a lower mean glucose concentration, and in greater percentage of glucose measurements maintained within target range, without an increased risk of severe hypoglycemia compared to the standard paper protocol.

Hyperglycemia in hospitalized patients is a common, serious, and costly health care problem. Evidence from observational and interventional studies indicate that hyperglycemia in critical illness is associated with an increased risk of complications and mortality.25 There is ongoing debate, however, about the optimal glucose level in hospitalized patients with critical illness. Although, several cohort studies as well as early randomized trials in ICU patients reported that intensified insulin treatment to achieve a target glucose between 80 mg/dL to 110 mg/dL reported a reduction in short‐term and long‐term mortality and rates of multiorgan failure and systemic infections compared with conventionally treated patients.3, 4, 17 More recent randomized controlled trials and meta‐analyses, however, have shown that this low BG target has been difficult to achieve without increasing the risk for severe hypoglycemia.710 In addition, recent multicenter trials have failed to show significant improvement in clinical outcome or have even shown increased mortality risk with intensive glycemic control.610 Based on these reports, the American Association of Clinical Endocrinologist (AACE) and American Diabetes Association (ADA) task force on inpatient glycemic control recommended different glycemic targets in the ICU setting. Current guidelines suggest targeting a BG level between 140 mg/dL and 180 mg/dL (7.8 and 10.0 mmol/L) for the majority of ICU patients and a lower glucose targets between 110 mg/dL and 140 mg/dL (6.1 and 7.8 mmol/L) in selected ICU patients (ie, centers with extensive experience and appropriate nursing support, cardiac surgical patients, patients with stable glycemic control without hypoglycemia). Glucose targets >180 mg/dL or <110 mg/dL are no longer recommended in ICU patients.

The rate of severe hypoglycemic events (<40 mg/dL) observed in both arms of our trial was significantly lower than those reported in recent international trials of intensive glycemic control.3, 4, 8 The overall rate of severe hypoglycemic events in international trials ranged between 5% to 28.6%.3, 4, 7, 8, 18, 19 In this trial, the number of patients with severe hypoglycemia was 3.9% in the computer‐based and 5.6% in the standard paper algorithm. Repeated measures analyses show the probabilities of BG readings <40 mg/dL were similar and not significantly different between groups (P = 0.969). We observed, however, a high rate of mild hypoglycemic events in patients treated with both insulin algorithms. The number of patients with BG <60 mg/dL was 42.9% in the Glucommander and 31.9% in the standard (P = NS). Minimizing the rate of hypoglycemia events is of major importance in hospitalized patients because it has been shown that hypoglycemia may be an independent risk factor of poor clinical outcome and mortality.20 Hypoglycemia may increase the risk of ventricular arrhythmias, in part due to the prolongation QT interval21 and can impair cerebral glucose metabolism resulting in brain metabolic dysfunction, as suggested by recent clinical studies.22 Moreover, insulin‐induced hypoglycemia is also associated with increased proinflammatory cytokines (tumor necrosis factor [TNF]‐alpha, interleukin [IL]‐1beta, IL‐6, and IL‐8) and oxidative stress23 that correlate with elevations of counterregulatory hormones (catecholamines, cortisol).

The Glucommander was associated with lower glycemic variability and with a higher percentage of BG readings within target range than patients treated with the standard paper form regimen. The clinical importance of the degree of variability and rapidity of fluctuations in glucose levels in critically ill patients is a topic of recent interest. Glycemic variability has been identified as a strong independent contributor to the risk of mortality in critically ill and surgical patients.24 Low levels of glycemic variability (standard deviation [SD] <10 mg/dL or 10‐20 mg/dL) have been shown to have a statistically significant lower risk of mortality, even after adjustment for severity of illness. Further studies are needed to determine benefits on clinical outcomes from the more consistent BG control from computer‐based titration protocols.

We acknowledge the following limitations in this multicenter open label study. First, this study was conducted in the medical ICU and excluded postsurgical patients and subjects expected to undergo a major surgical procedure during the hospital stay. Although a recent meta‐analysis9 of 26 studies involving 13,567 patients reported no benefits in the general ICU population, it found a favorable effect of intensive glycemic control on mortality in surgical ICU patients (relative risk [RR], 0.63; confidence interval [CI], 0.44‐0.91). We also excluded patients with severe renal insufficiency and patients with a history of hyperglycemic crises. In addition, our study was not powered to demonstrate differences in mortality or clinical outcome between treatment groups, and the BG targets used in this study were lower than glycemic targets recently recommended by the AACE and ADA inpatient glycemic control task force.25 Raising the BG targets is likely to reduce or prevent the rate of mild and severe hypoglycemic events in the ICU.

In conclusion, the computer‐guided algorithm resulted in a more rapid and tighter glycemic control with a similar rate of hypoglycemic events than the standard paper form protocol in medical ICU patients. Our study suggests that, both treatment algorithms are appropriate alternatives for the management of hyperglycemia in critically ill patients, and the choice depends on a physician's preferences, cost considerations, and the availability of the computer guided algorithm. Large randomized clinical trials are needed to test the impact of the new AACE/ADA recommended BG targets in reducing hypoglycemic events, hospital complications, and hospital mortality in critically ill patients in the ICU.

Observational studies in hospitalized patients with and without diabetes indicate that hyperglycemia is a predictor of poor clinical outcome and mortality.14 Early randomized controlled trials of intensified insulin therapy in patients with surgical and medical acute critical illness reported a reduction on the risk of multiorgan failure and systemic infections,35 as well as short‐ and long‐term mortality.1, 4 Recent randomized controlled trials, however, have failed to confirm the previously suggested benefits of intensive glucose control,6 and the large multicenter normoglycaemia in intensive care evaluation and survival using glucose algorithm regulation (NICE‐SUGAR) study reported an absolute increase in mortality rate with intensive glucose control.7 In addition, intensified insulin therapy in critically‐ill patients has been shown to be associated with a higher rate of severe hypoglycemic events than less aggressive glycemic control protocols.710 These results have led to a heightened interest in improving the quality and safety of the management of diabetes and hyperglycemia in the hospital.

The use of intravenous continuous insulin infusion (CII) is the preferred route of insulin administration for the management of hyperglycemia in the critical care setting.1, 11 Numerous examples of successful CII algorithms in achieving glycemic control are reported in the literature.4, 5, 12 Traditionally, order forms to titrate drip to achieve a target blood glucose (BG) range using an established algorithm or by the application of mathematical rules have been used in clinical practice. Recently, computer‐based algorithms aiming to direct the nursing staff adjusting insulin infusion rate have become commercially available.13, 14 It is not known, however, if computer‐based algorithms are superior to standard paper form‐based protocols in achieving glucose control and in reducing hypoglycemic events in critically‐ill patients. Accordingly, this multicenter randomized study aimed to determine differences in glycemic control and hypoglycemic events between treatment with a computer‐guided CII device and a standard column‐based paper algorithm in critically‐ill patients in the medical intensive care unit (ICU).

Research Design and Methods

In this multicenter, prospective, open‐label randomized study, 160 adult patients admitted to a medical ICU with new hyperglycemia or with a known history of diabetes treated with diet, insulin therapy or with any combination of oral antidiabetic agents were enrolled after written informed consent had been obtained from the patient or closest family member (Figure 1). Patients with known history of diabetes had 2 BG readings >120 mg/dL while subjects without a history of diabetes had 2 BG readings >140 mg/dL prior to enrollment. We excluded patients with acute hyperglycemic crises such as diabetic ketoacidosis (DKA) and hyperosmolar hyperglycemic state,15 patients with severely impaired renal function (serum creatinine 3.5 mg/dL), dementia, and pregnancy. This study was conducted at 4 hospital centers including Grady Memorial Hospital, Emory University Hospital, and Piedmont Hospital in Atlanta, Georgia and the Regional Medical Center in Memphis, Tennessee.

Figure 1

Figure 2

Patients were randomized using a computer randomization table to receive CII following a computer‐guided algorithm (Glucommander) or CII following a standard paper form insulin infusion algorithm. Both protocols used glulisine (Apidra) insulin and targeted a BG between 80 mg/dL and 120 mg/dL. Insulin management was directed by the specific assigned protocol and was carried out daily by the nursing staff and by members of the internal medicine residency program. The ICU physician and primary care team decided on the treatment for all other medical problem(s) for which patients were admitted. Data were collected during CII up to the first 10 days of ICU stay.

Results

The admission characteristics and clinical outcomes of interest of the study patients are shown in Table 1. A total of 160 adult patients admitted to a medical ICU with new hyperglycemia (47%) or with a known history of diabetes (53%) were randomized into the study. Of them, 7 patients were excluded due to withdrawal of consent, treatment with subcutaneous basal or supplemental short‐acting insulin during CII, or receiving less than 4 hours of CII. There were no differences in the mean age, gender, race, history of diabetes, or primary admitting diagnosis between treatment groups. The most common admitting diagnosis categories included pulmonary (22.1%), cardiovascular (21.4%), infectious (20.0%), and central nervous system (16.6%) disorders.

The mean admission glucose concentration for study patients was 190.6 58.2 mg/dL and the mean A1C was 7.0 1.7%. Glycemic control parameters achieved with the CII protocols are listed in Table 2. At the start of CII, the mean BG value was similar for the Glucommander and paper protocols (189.7 64.8 mg/dL and 188.4 54.8 mg/dL, P = 0.419). The mean time to reach the BG target was shorter in the Glucommander group (4.8 2.8 vs. 7.8 9.1 hours, P < 0.001). The Glucommander group had a lower mean glucose value during insulin infusion (115.5 20.7 vs. 131.0 24.6 mg/dL, P < 0.001) and once at target goal, in a lower mean BG values (103.3 8.8 vs. 117.3 16.5 mg/dL, P < 0.001) than the standard algorithm (Figure 2). The mean inpatient BG difference between treatment groups was 15.5 mg/dL (P < 0.001), with a mean daily BG difference ranging from 17.4 mg/dL to 24.4 mg/dL less for the Glucommander group during days 2 to 6 of therapy (P < 0.01).

The Glucommander algorithm was associated with tighter glycemic control and less glucose variability than the standard paper form protocol. Once patients achieved BG target, on average 71.1% of BG readings in the Glucommander and 51.3% in the standard group remained within the 80 mg/dL to 120 mg/dL target range (P < 0.001). In addition, the Glucommander was associated with a significantly lower rate of severe hyperglycemia during insulin infusion. The number of patients with 1 or more episodes of BG >200 mg/dL (11.7% vs. 25%, P = 0.057 before adjusting for potential site difference and P = 0.034 after adjusting for site difference) were less in the Glucommander group than in the standard paper regimen. In addition, 4 of these patients in spite of being on the highest insulin delivery column failed to achieve glucoses <180 and had an average in‐hospital glucose level of 204.5 32.2 mg/dL. These patients were transitioned to the Glucommander arm and withdrawn from the study. All episodes of hypoglycemia occurred after the patients achieved 1 glucose measurement within the target range. The number of patients who experienced one or more BG <40 mg/dL and <60 mg/dL was 3.9% and 42.9% in the Glucommander and 5.6% and 31.9% in the standard regimen, respectively (both, P = not significant [NS]). Similar results were obtained when site effect was accommodated (both, P = NS). Based on repeated measures analyses, the probabilities of BG reading <40 mg/dL or <60 mg/dL were not significantly different between groups (P = 0.969, P = 0.084) after accounting for within‐patient correlations with or without adjusting for time effect. None of these episodes resulted in seizures or were otherwise judged to be associated with deterioration of clinical status.

The mean insulin infusion rate was slightly higher in the Glucommander regimen but the difference was not statistically significant between groups. Patients treated with the Glucommander protocol received a mean infusion rate of 0.035 0.024 unit/kg/hour for a total of 2.85 1.93 units per hour, and those treated with the paper protocol received a 0.028 0.021 units/kg/hour for a total of 2.50 2.28 units per hour, P = 0.12 and P = 0.09, respectively.

The numbers of BG measurements were similar between the Glucommander and standard paper algorithms (44.2 39.8 and 41.2 34.5 respectively, P = NS) with the number of glucose testing per patient ranging from 6 to 175 in the Glucommander and 3 to 168 in the standard group. Similarly, when normalized to the duration of insulin infusion, the frequency of BG monitoring was not different with the protocols (0.68 0.18 and 0.62 0.22 tests/hour respectively, P = NS).

Compared to the standard paper insulin infusion algorithm, patients treated with the Glucommander device had a similar mean ICU LOS (13.4 13.8 vs. 8.5 7.6 days, P = 0.145), mean hospital LOS (17.5 15.0 days vs. 23.9 26.3 days, P = 0.704) and hospital mortality (26.0% vs. 21.9%, P = 0.561).

Discussion

This study is the first to compare the safety and efficacy of a CII via a computer‐guided algorithm and a standard paper form protocol in nonsurgical patients in the ICU. Both treatment algorithms resulted in significant improvement in glycemic control with the Glucommander achieving glycemic glucose target in a shorter time of treatment, a lower mean glucose concentration, and in greater percentage of glucose measurements maintained within target range, without an increased risk of severe hypoglycemia compared to the standard paper protocol.

Hyperglycemia in hospitalized patients is a common, serious, and costly health care problem. Evidence from observational and interventional studies indicate that hyperglycemia in critical illness is associated with an increased risk of complications and mortality.25 There is ongoing debate, however, about the optimal glucose level in hospitalized patients with critical illness. Although, several cohort studies as well as early randomized trials in ICU patients reported that intensified insulin treatment to achieve a target glucose between 80 mg/dL to 110 mg/dL reported a reduction in short‐term and long‐term mortality and rates of multiorgan failure and systemic infections compared with conventionally treated patients.3, 4, 17 More recent randomized controlled trials and meta‐analyses, however, have shown that this low BG target has been difficult to achieve without increasing the risk for severe hypoglycemia.710 In addition, recent multicenter trials have failed to show significant improvement in clinical outcome or have even shown increased mortality risk with intensive glycemic control.610 Based on these reports, the American Association of Clinical Endocrinologist (AACE) and American Diabetes Association (ADA) task force on inpatient glycemic control recommended different glycemic targets in the ICU setting. Current guidelines suggest targeting a BG level between 140 mg/dL and 180 mg/dL (7.8 and 10.0 mmol/L) for the majority of ICU patients and a lower glucose targets between 110 mg/dL and 140 mg/dL (6.1 and 7.8 mmol/L) in selected ICU patients (ie, centers with extensive experience and appropriate nursing support, cardiac surgical patients, patients with stable glycemic control without hypoglycemia). Glucose targets >180 mg/dL or <110 mg/dL are no longer recommended in ICU patients.

The rate of severe hypoglycemic events (<40 mg/dL) observed in both arms of our trial was significantly lower than those reported in recent international trials of intensive glycemic control.3, 4, 8 The overall rate of severe hypoglycemic events in international trials ranged between 5% to 28.6%.3, 4, 7, 8, 18, 19 In this trial, the number of patients with severe hypoglycemia was 3.9% in the computer‐based and 5.6% in the standard paper algorithm. Repeated measures analyses show the probabilities of BG readings <40 mg/dL were similar and not significantly different between groups (P = 0.969). We observed, however, a high rate of mild hypoglycemic events in patients treated with both insulin algorithms. The number of patients with BG <60 mg/dL was 42.9% in the Glucommander and 31.9% in the standard (P = NS). Minimizing the rate of hypoglycemia events is of major importance in hospitalized patients because it has been shown that hypoglycemia may be an independent risk factor of poor clinical outcome and mortality.20 Hypoglycemia may increase the risk of ventricular arrhythmias, in part due to the prolongation QT interval21 and can impair cerebral glucose metabolism resulting in brain metabolic dysfunction, as suggested by recent clinical studies.22 Moreover, insulin‐induced hypoglycemia is also associated with increased proinflammatory cytokines (tumor necrosis factor [TNF]‐alpha, interleukin [IL]‐1beta, IL‐6, and IL‐8) and oxidative stress23 that correlate with elevations of counterregulatory hormones (catecholamines, cortisol).

The Glucommander was associated with lower glycemic variability and with a higher percentage of BG readings within target range than patients treated with the standard paper form regimen. The clinical importance of the degree of variability and rapidity of fluctuations in glucose levels in critically ill patients is a topic of recent interest. Glycemic variability has been identified as a strong independent contributor to the risk of mortality in critically ill and surgical patients.24 Low levels of glycemic variability (standard deviation [SD] <10 mg/dL or 10‐20 mg/dL) have been shown to have a statistically significant lower risk of mortality, even after adjustment for severity of illness. Further studies are needed to determine benefits on clinical outcomes from the more consistent BG control from computer‐based titration protocols.

We acknowledge the following limitations in this multicenter open label study. First, this study was conducted in the medical ICU and excluded postsurgical patients and subjects expected to undergo a major surgical procedure during the hospital stay. Although a recent meta‐analysis9 of 26 studies involving 13,567 patients reported no benefits in the general ICU population, it found a favorable effect of intensive glycemic control on mortality in surgical ICU patients (relative risk [RR], 0.63; confidence interval [CI], 0.44‐0.91). We also excluded patients with severe renal insufficiency and patients with a history of hyperglycemic crises. In addition, our study was not powered to demonstrate differences in mortality or clinical outcome between treatment groups, and the BG targets used in this study were lower than glycemic targets recently recommended by the AACE and ADA inpatient glycemic control task force.25 Raising the BG targets is likely to reduce or prevent the rate of mild and severe hypoglycemic events in the ICU.

In conclusion, the computer‐guided algorithm resulted in a more rapid and tighter glycemic control with a similar rate of hypoglycemic events than the standard paper form protocol in medical ICU patients. Our study suggests that, both treatment algorithms are appropriate alternatives for the management of hyperglycemia in critically ill patients, and the choice depends on a physician's preferences, cost considerations, and the availability of the computer guided algorithm. Large randomized clinical trials are needed to test the impact of the new AACE/ADA recommended BG targets in reducing hypoglycemic events, hospital complications, and hospital mortality in critically ill patients in the ICU.

Shock has been defined as failure to deliver and/or utilize adequate amounts of oxygen1 and is a common cause of critical illness. Few studies have examined the predictive utility of bedside clinical examination in predicting the category of shock. Scholars have suggested a bedside approach that uses simple examination techniques and applied physiology to rapidly identify a patients' circulation as high vs. low cardiac output. Those with a high‐output examination are designated as high‐output, most often septic shock. Low‐output patients are further categorized as heart full or heart empty to distinguish cardiogenic from hypovolemic categories of shock, respectively.2 The predictive characteristics of this simple algorithm have not been studied. In this study, we examine the operating characteristics of selected elements of this algorithm when administered at the bedside by trainees in Internal Medicine.

Methods

This study was performed after approval of the Institutional Review Board; informed consent was waived. Patients with nonsurgical problems who present to the hospital or who develop sustained hypotension are managed by medical house officers on the intensive care and/or rapid response team with the supervision of patients' attending physicians. All house officers were asked to document explicitly in their assessment notes the following examination findings: finger capillary refill (same/quicker vs. slower than examiner's), hand skin temperature (same/warmer vs. cooler than examiner's) and pulse pressure (ie, same/wider vs. thinner than examiner's), presence or absence of crackles >1/3 from base on bilateral lung examination and jugular venous pressure (JVP) vs. <8 cmH₂O. The documented examinations of either the rapid response team (PGY2; n = 14) or intensive care unit (ICU) resident (PGY3; n = 14) for patients evaluated between September 2008 and February 2009 were used for this study. Resuscitation was administered entirely by house officers, occasionally guided in person, but always supervised by attending physicians.

In May 2009, clinical data, including electrocardiograms/echocardiograms and laboratory (eg, cardiac enzymes, culture) results were abstracted from medical records of subjects. These were presented to a blinded senior clinician (DK) to review and apply evidence‐based or consensus criteria,36 whenever possible, to categorize the type of shock (septic vs. cardiogenic vs. hypovolemic) based on data acquired after the onset of shock. For example, patients with microbiologic and/or radiologic evidence of infection were classified as septic shock,1, 3, 4 those with acute left or right ventricular dysfunction on echocardiogram were classified as cardiogenic shock,1, 6 and those with clinical evidence of acute hemorrhage with hypovolemic shock.1, 5 While some of the patients were examined by DK as part of clinical care, he was blinded to the identity of patients and their algorithm‐related physical examination findings when he reviewed the abstracted data (>2 months after study closure) to adjudicate the final diagnosis of shock. These diagnoses were considered the reference standard for this study. The operating characteristics (sensitivity = true positive/true positive + false negative; specificity = true negative/true negative + false positive; negative predictive value (NPV) = true negative/all negatives; positive predictive value (PPV) = true positive/all positives; accuracy = true results/all results) were calculated for combinations of physical examination findings and correct final diagnosis (Figure 1).

Figure 1

Results

A total of 68 patients, averaging 71 16 years, were studied; 57% were male, and 66% were White, and 20% were Black. Table 1 lists characteristics of patients. A total of 37 patients were diagnosed as having septic shock, 11 had cardiogenic shock and 10 hypovolemic shock. Operating characteristics of the bedside examination techniques for predicting mechanism of shock are listed in Table 2. Capillary refill and skin temperature were 100% concordant yielding sensitivity of 89% (95% confidence interval [CI], 75‐97%), specificity of 68% (95% CI, 46‐83%), PPV of 77% (95% CI, 61‐88%), NPV of 84% (95% CI, 64‐96%) and overall accuracy of 79% (95% CI, 68‐88%) for diagnosis of high output (ie, septic shock). JVP 8 cmH₂O was more accurate than crackles for predicting cardiogenic shock in low‐output patients with sensitivity of 82% (95% CI, 48‐98%), specificity of 79% (95% CI, 41‐95%), PPV of 75% (95% CI, 43‐95%), NPV of 85% (95% CI, 55‐98%), and overall accuracy of 80% (95%CI, 59‐93%). Using just skin temperature and JVP, the bedside approach misdiagnosed 19 of 75 cases (overall accuracy, 75%; 95% CI, 16‐37%).

Discussion

This is the first study to examine the predictive characteristics of simple bedside physical examination techniques in correctly predicting the category/mechanism of shock. Overall, the algorithm performed well, and accurately predicted the category of shock in three‐quarters of patients. It also has the benefit of being very rapid, taking only seconds to complete, using bedside techniques that even inexperienced clinicians can apply.

Very few studies have examined the accuracy of examination techniques specifically for diagnosis of shock. In humans injected with endotoxin, body temperature and cardiac output increased, but skin temperature and capillary refill times are not well described.79 Schriger and Baraff10 reported that capillary refill >2 seconds was only 59% sensitive for diagnosing hypovolemia in patients with hypovolemic shock or orthostatic changes in blood pressure. Sensitivity was 77% in 13 patients with hypovolemic shock.10 However, some studies have demonstrated that age, sex, external temperature11 and fever12 can affect capillary refill times. Otieno et al.13 demonstrated a kappa statistic value of 0.49 for capillary refill 4 seconds, suggesting that reproducibility of this technique could be a major drawback. McGee et al.14 reviewed examination techniques for diagnosing hypovolemic states and concluded that postural changes in heart rate and blood pressure were the most accurate; capillary refill was not recommended. Stevenson and Perloff15 demonstrated that crackles and elevated JVP were absent in 18 of 43 patients with pulmonary capillary wedge pressures >22 mmHg. Butman et al.16 showed that elevated JVP was 82% accurate for predicting a wedge pressure >18 mmHg. Connors et al.17 demonstrated that clinicians' predictions of heart filling pressures and cardiac output were accurate (relative to pulmonary artery catheter measurements) in less than 50% of cases, though the examination techniques used were not qualified or quantified. No previous study has combined simple, semiobjective physical examination techniques for the purpose of distinguishing categories of shock.

Since identification of the pathogenesis of shock has important treatment/prognostic implications (eg, fluid and vasopressor therapies, early search for drainable focus of infection in sepsis, reestablishing vessel patency in myocardial infarction and pulmonary embolus), we believe that this simple, rapidly administered algorithm will prove useful in clinical medicine. In some clinical situations, the approach can lead to timely identification of the causative mechanism, allowing prompt definitive treatment. For example, a patient presenting with high‐output hypotension is so often sepsis/septic shock that treatment with antibiotics is justified (since success is time‐sensitive) even when the exact site/microbe has not yet been identified. Acute right heart overfilled low‐output hypotension should be considered pulmonary embolism (which also requires time‐sensitive therapies) until proven otherwise. Yet, a sizeable number of cases do not fit neatly into a single category. For example, 11% of patients with septic shock presented with cool extremities in the early phases of illness. In clinical decision‐making, 2 diagnostic‐therapeutic paradigms are common. In the first, the diagnosis is relatively certain and narrowly‐directed, mechanism‐specific treatment is appropriate. The second paradigm is 1 of significant uncertainty, when clinicians must treat empirically the most likely causes until more data become available to permit safe narrowing of therapies. For example, a patient presenting with hypotension, cool extremities, leukocytosis and apparent pneumonia should be treated empirically for septic shock while exploring explanations for the incongruous low‐output state (eg, profound hypovolemia, adrenal insufficiency, concurrent or precedent myocardial dysfunction). Patients often have several mechanisms contributing to hypotension. Since patients are not ideal forms, there can be no perfect decision‐tool; clinicians would be fool‐hardy to prematurely close decision‐making prior to definitive diagnosis. In the case of shock, such diagnostic arrogance would delay time‐sensitive therapies and thus contribute to morbidity and mortality. Nonetheless, this physical examination algorithmunderstanding its operating characteristics and limitationsmay add to the bedside clinician's diagnostic armamentarium.

Our study has several notable limitations. First, bedside examinations were performed by multiple observers who had limited (1 electronic mail) instruction on how to perform and document the data gathered for this study. So these results should be generalized cautiously until reproduced at other centers with greater numbers of observers (than the 28 of this study). The central supposition, that skin cooler, capillary refill longer, and pulse pressure more narrow than theirs, presupposes reasonable homogeneity of the normal state which is not necessarily true.11 Interobserver variability of physical examination further compromises the fidelity of findings recorded for this study.13 Since we conducted a retrospective review, and because of the emergency nature of the clinical problem, it would be difficult to conduct a study in which multiple examiners performed the same physical examinations to quantify interobserver variability. Irrespective, we would expect interobserver variability to systematically reduce accuracy; it is all‐the‐more impressive that trainees' examination results correctly diagnosed mechanism of shock in three‐quarters of cases. Also, examiners were not blinded to clinical history, so results of their examination could have been biased by their pre‐examination hypotheses of pathogenesis. Of course, they were not aware of the expert's final categorization of mechanism performed much later in time. Since there is no absolute reference standard for classification of the pathogenesis of shock, we depended upon careful review of selected data (same parameters for each patient) by a single senior investigatoralbeit armed with evidence‐based or consensus‐based standards of diagnosing shock. Finally, it can be argued that all forms of shock are mixed (with hypovolemia) early in the course; sepsis requires refilling of a leaky and dilated vasculature and the noncompliant ischemic ventricle often requires a higher filling pressure than normal to empty. To complicate even more, patients may have preexistent conditions (eg, chronic congestive heart failure, cirrhosis) that limit cardiovascular responses to acute shock. Our diagnostic approach was to identify the principal cause of the acute decompensation, assuming that many patients will have more than 1 single mechanism accounting for hypotension.

In conclusion, this is the first study to examine the utility of this simple physical examination algorithm to diagnose the mechanism of shock. Some have discounted or underemphasized examination techniques in favor of more time‐intensive and labor‐intensive diagnostic modalities, such as bedside echocardiography, which may waste precious time and resources. The simple physical examination algorithm assessed in this study has favorable operating characteristics and can be performed readily by even novice clinicians. If replicated at other centers and by greater numbers of observers, this approach could assist clinicians and teachers who train clinicians to rapidly diagnose and manage patients with shock.

Shock has been defined as failure to deliver and/or utilize adequate amounts of oxygen1 and is a common cause of critical illness. Few studies have examined the predictive utility of bedside clinical examination in predicting the category of shock. Scholars have suggested a bedside approach that uses simple examination techniques and applied physiology to rapidly identify a patients' circulation as high vs. low cardiac output. Those with a high‐output examination are designated as high‐output, most often septic shock. Low‐output patients are further categorized as heart full or heart empty to distinguish cardiogenic from hypovolemic categories of shock, respectively.2 The predictive characteristics of this simple algorithm have not been studied. In this study, we examine the operating characteristics of selected elements of this algorithm when administered at the bedside by trainees in Internal Medicine.

Methods

This study was performed after approval of the Institutional Review Board; informed consent was waived. Patients with nonsurgical problems who present to the hospital or who develop sustained hypotension are managed by medical house officers on the intensive care and/or rapid response team with the supervision of patients' attending physicians. All house officers were asked to document explicitly in their assessment notes the following examination findings: finger capillary refill (same/quicker vs. slower than examiner's), hand skin temperature (same/warmer vs. cooler than examiner's) and pulse pressure (ie, same/wider vs. thinner than examiner's), presence or absence of crackles >1/3 from base on bilateral lung examination and jugular venous pressure (JVP) vs. <8 cmH₂O. The documented examinations of either the rapid response team (PGY2; n = 14) or intensive care unit (ICU) resident (PGY3; n = 14) for patients evaluated between September 2008 and February 2009 were used for this study. Resuscitation was administered entirely by house officers, occasionally guided in person, but always supervised by attending physicians.

In May 2009, clinical data, including electrocardiograms/echocardiograms and laboratory (eg, cardiac enzymes, culture) results were abstracted from medical records of subjects. These were presented to a blinded senior clinician (DK) to review and apply evidence‐based or consensus criteria,36 whenever possible, to categorize the type of shock (septic vs. cardiogenic vs. hypovolemic) based on data acquired after the onset of shock. For example, patients with microbiologic and/or radiologic evidence of infection were classified as septic shock,1, 3, 4 those with acute left or right ventricular dysfunction on echocardiogram were classified as cardiogenic shock,1, 6 and those with clinical evidence of acute hemorrhage with hypovolemic shock.1, 5 While some of the patients were examined by DK as part of clinical care, he was blinded to the identity of patients and their algorithm‐related physical examination findings when he reviewed the abstracted data (>2 months after study closure) to adjudicate the final diagnosis of shock. These diagnoses were considered the reference standard for this study. The operating characteristics (sensitivity = true positive/true positive + false negative; specificity = true negative/true negative + false positive; negative predictive value (NPV) = true negative/all negatives; positive predictive value (PPV) = true positive/all positives; accuracy = true results/all results) were calculated for combinations of physical examination findings and correct final diagnosis (Figure 1).

Figure 1

Results

A total of 68 patients, averaging 71 16 years, were studied; 57% were male, and 66% were White, and 20% were Black. Table 1 lists characteristics of patients. A total of 37 patients were diagnosed as having septic shock, 11 had cardiogenic shock and 10 hypovolemic shock. Operating characteristics of the bedside examination techniques for predicting mechanism of shock are listed in Table 2. Capillary refill and skin temperature were 100% concordant yielding sensitivity of 89% (95% confidence interval [CI], 75‐97%), specificity of 68% (95% CI, 46‐83%), PPV of 77% (95% CI, 61‐88%), NPV of 84% (95% CI, 64‐96%) and overall accuracy of 79% (95% CI, 68‐88%) for diagnosis of high output (ie, septic shock). JVP 8 cmH₂O was more accurate than crackles for predicting cardiogenic shock in low‐output patients with sensitivity of 82% (95% CI, 48‐98%), specificity of 79% (95% CI, 41‐95%), PPV of 75% (95% CI, 43‐95%), NPV of 85% (95% CI, 55‐98%), and overall accuracy of 80% (95%CI, 59‐93%). Using just skin temperature and JVP, the bedside approach misdiagnosed 19 of 75 cases (overall accuracy, 75%; 95% CI, 16‐37%).

Discussion

This is the first study to examine the predictive characteristics of simple bedside physical examination techniques in correctly predicting the category/mechanism of shock. Overall, the algorithm performed well, and accurately predicted the category of shock in three‐quarters of patients. It also has the benefit of being very rapid, taking only seconds to complete, using bedside techniques that even inexperienced clinicians can apply.

Very few studies have examined the accuracy of examination techniques specifically for diagnosis of shock. In humans injected with endotoxin, body temperature and cardiac output increased, but skin temperature and capillary refill times are not well described.79 Schriger and Baraff10 reported that capillary refill >2 seconds was only 59% sensitive for diagnosing hypovolemia in patients with hypovolemic shock or orthostatic changes in blood pressure. Sensitivity was 77% in 13 patients with hypovolemic shock.10 However, some studies have demonstrated that age, sex, external temperature11 and fever12 can affect capillary refill times. Otieno et al.13 demonstrated a kappa statistic value of 0.49 for capillary refill 4 seconds, suggesting that reproducibility of this technique could be a major drawback. McGee et al.14 reviewed examination techniques for diagnosing hypovolemic states and concluded that postural changes in heart rate and blood pressure were the most accurate; capillary refill was not recommended. Stevenson and Perloff15 demonstrated that crackles and elevated JVP were absent in 18 of 43 patients with pulmonary capillary wedge pressures >22 mmHg. Butman et al.16 showed that elevated JVP was 82% accurate for predicting a wedge pressure >18 mmHg. Connors et al.17 demonstrated that clinicians' predictions of heart filling pressures and cardiac output were accurate (relative to pulmonary artery catheter measurements) in less than 50% of cases, though the examination techniques used were not qualified or quantified. No previous study has combined simple, semiobjective physical examination techniques for the purpose of distinguishing categories of shock.

Since identification of the pathogenesis of shock has important treatment/prognostic implications (eg, fluid and vasopressor therapies, early search for drainable focus of infection in sepsis, reestablishing vessel patency in myocardial infarction and pulmonary embolus), we believe that this simple, rapidly administered algorithm will prove useful in clinical medicine. In some clinical situations, the approach can lead to timely identification of the causative mechanism, allowing prompt definitive treatment. For example, a patient presenting with high‐output hypotension is so often sepsis/septic shock that treatment with antibiotics is justified (since success is time‐sensitive) even when the exact site/microbe has not yet been identified. Acute right heart overfilled low‐output hypotension should be considered pulmonary embolism (which also requires time‐sensitive therapies) until proven otherwise. Yet, a sizeable number of cases do not fit neatly into a single category. For example, 11% of patients with septic shock presented with cool extremities in the early phases of illness. In clinical decision‐making, 2 diagnostic‐therapeutic paradigms are common. In the first, the diagnosis is relatively certain and narrowly‐directed, mechanism‐specific treatment is appropriate. The second paradigm is 1 of significant uncertainty, when clinicians must treat empirically the most likely causes until more data become available to permit safe narrowing of therapies. For example, a patient presenting with hypotension, cool extremities, leukocytosis and apparent pneumonia should be treated empirically for septic shock while exploring explanations for the incongruous low‐output state (eg, profound hypovolemia, adrenal insufficiency, concurrent or precedent myocardial dysfunction). Patients often have several mechanisms contributing to hypotension. Since patients are not ideal forms, there can be no perfect decision‐tool; clinicians would be fool‐hardy to prematurely close decision‐making prior to definitive diagnosis. In the case of shock, such diagnostic arrogance would delay time‐sensitive therapies and thus contribute to morbidity and mortality. Nonetheless, this physical examination algorithmunderstanding its operating characteristics and limitationsmay add to the bedside clinician's diagnostic armamentarium.

Our study has several notable limitations. First, bedside examinations were performed by multiple observers who had limited (1 electronic mail) instruction on how to perform and document the data gathered for this study. So these results should be generalized cautiously until reproduced at other centers with greater numbers of observers (than the 28 of this study). The central supposition, that skin cooler, capillary refill longer, and pulse pressure more narrow than theirs, presupposes reasonable homogeneity of the normal state which is not necessarily true.11 Interobserver variability of physical examination further compromises the fidelity of findings recorded for this study.13 Since we conducted a retrospective review, and because of the emergency nature of the clinical problem, it would be difficult to conduct a study in which multiple examiners performed the same physical examinations to quantify interobserver variability. Irrespective, we would expect interobserver variability to systematically reduce accuracy; it is all‐the‐more impressive that trainees' examination results correctly diagnosed mechanism of shock in three‐quarters of cases. Also, examiners were not blinded to clinical history, so results of their examination could have been biased by their pre‐examination hypotheses of pathogenesis. Of course, they were not aware of the expert's final categorization of mechanism performed much later in time. Since there is no absolute reference standard for classification of the pathogenesis of shock, we depended upon careful review of selected data (same parameters for each patient) by a single senior investigatoralbeit armed with evidence‐based or consensus‐based standards of diagnosing shock. Finally, it can be argued that all forms of shock are mixed (with hypovolemia) early in the course; sepsis requires refilling of a leaky and dilated vasculature and the noncompliant ischemic ventricle often requires a higher filling pressure than normal to empty. To complicate even more, patients may have preexistent conditions (eg, chronic congestive heart failure, cirrhosis) that limit cardiovascular responses to acute shock. Our diagnostic approach was to identify the principal cause of the acute decompensation, assuming that many patients will have more than 1 single mechanism accounting for hypotension.

In conclusion, this is the first study to examine the utility of this simple physical examination algorithm to diagnose the mechanism of shock. Some have discounted or underemphasized examination techniques in favor of more time‐intensive and labor‐intensive diagnostic modalities, such as bedside echocardiography, which may waste precious time and resources. The simple physical examination algorithm assessed in this study has favorable operating characteristics and can be performed readily by even novice clinicians. If replicated at other centers and by greater numbers of observers, this approach could assist clinicians and teachers who train clinicians to rapidly diagnose and manage patients with shock.

Shock has been defined as failure to deliver and/or utilize adequate amounts of oxygen1 and is a common cause of critical illness. Few studies have examined the predictive utility of bedside clinical examination in predicting the category of shock. Scholars have suggested a bedside approach that uses simple examination techniques and applied physiology to rapidly identify a patients' circulation as high vs. low cardiac output. Those with a high‐output examination are designated as high‐output, most often septic shock. Low‐output patients are further categorized as heart full or heart empty to distinguish cardiogenic from hypovolemic categories of shock, respectively.2 The predictive characteristics of this simple algorithm have not been studied. In this study, we examine the operating characteristics of selected elements of this algorithm when administered at the bedside by trainees in Internal Medicine.

Methods

This study was performed after approval of the Institutional Review Board; informed consent was waived. Patients with nonsurgical problems who present to the hospital or who develop sustained hypotension are managed by medical house officers on the intensive care and/or rapid response team with the supervision of patients' attending physicians. All house officers were asked to document explicitly in their assessment notes the following examination findings: finger capillary refill (same/quicker vs. slower than examiner's), hand skin temperature (same/warmer vs. cooler than examiner's) and pulse pressure (ie, same/wider vs. thinner than examiner's), presence or absence of crackles >1/3 from base on bilateral lung examination and jugular venous pressure (JVP) vs. <8 cmH₂O. The documented examinations of either the rapid response team (PGY2; n = 14) or intensive care unit (ICU) resident (PGY3; n = 14) for patients evaluated between September 2008 and February 2009 were used for this study. Resuscitation was administered entirely by house officers, occasionally guided in person, but always supervised by attending physicians.

In May 2009, clinical data, including electrocardiograms/echocardiograms and laboratory (eg, cardiac enzymes, culture) results were abstracted from medical records of subjects. These were presented to a blinded senior clinician (DK) to review and apply evidence‐based or consensus criteria,36 whenever possible, to categorize the type of shock (septic vs. cardiogenic vs. hypovolemic) based on data acquired after the onset of shock. For example, patients with microbiologic and/or radiologic evidence of infection were classified as septic shock,1, 3, 4 those with acute left or right ventricular dysfunction on echocardiogram were classified as cardiogenic shock,1, 6 and those with clinical evidence of acute hemorrhage with hypovolemic shock.1, 5 While some of the patients were examined by DK as part of clinical care, he was blinded to the identity of patients and their algorithm‐related physical examination findings when he reviewed the abstracted data (>2 months after study closure) to adjudicate the final diagnosis of shock. These diagnoses were considered the reference standard for this study. The operating characteristics (sensitivity = true positive/true positive + false negative; specificity = true negative/true negative + false positive; negative predictive value (NPV) = true negative/all negatives; positive predictive value (PPV) = true positive/all positives; accuracy = true results/all results) were calculated for combinations of physical examination findings and correct final diagnosis (Figure 1).

Figure 1

Results

A total of 68 patients, averaging 71 16 years, were studied; 57% were male, and 66% were White, and 20% were Black. Table 1 lists characteristics of patients. A total of 37 patients were diagnosed as having septic shock, 11 had cardiogenic shock and 10 hypovolemic shock. Operating characteristics of the bedside examination techniques for predicting mechanism of shock are listed in Table 2. Capillary refill and skin temperature were 100% concordant yielding sensitivity of 89% (95% confidence interval [CI], 75‐97%), specificity of 68% (95% CI, 46‐83%), PPV of 77% (95% CI, 61‐88%), NPV of 84% (95% CI, 64‐96%) and overall accuracy of 79% (95% CI, 68‐88%) for diagnosis of high output (ie, septic shock). JVP 8 cmH₂O was more accurate than crackles for predicting cardiogenic shock in low‐output patients with sensitivity of 82% (95% CI, 48‐98%), specificity of 79% (95% CI, 41‐95%), PPV of 75% (95% CI, 43‐95%), NPV of 85% (95% CI, 55‐98%), and overall accuracy of 80% (95%CI, 59‐93%). Using just skin temperature and JVP, the bedside approach misdiagnosed 19 of 75 cases (overall accuracy, 75%; 95% CI, 16‐37%).

Discussion

This is the first study to examine the predictive characteristics of simple bedside physical examination techniques in correctly predicting the category/mechanism of shock. Overall, the algorithm performed well, and accurately predicted the category of shock in three‐quarters of patients. It also has the benefit of being very rapid, taking only seconds to complete, using bedside techniques that even inexperienced clinicians can apply.

Very few studies have examined the accuracy of examination techniques specifically for diagnosis of shock. In humans injected with endotoxin, body temperature and cardiac output increased, but skin temperature and capillary refill times are not well described.79 Schriger and Baraff10 reported that capillary refill >2 seconds was only 59% sensitive for diagnosing hypovolemia in patients with hypovolemic shock or orthostatic changes in blood pressure. Sensitivity was 77% in 13 patients with hypovolemic shock.10 However, some studies have demonstrated that age, sex, external temperature11 and fever12 can affect capillary refill times. Otieno et al.13 demonstrated a kappa statistic value of 0.49 for capillary refill 4 seconds, suggesting that reproducibility of this technique could be a major drawback. McGee et al.14 reviewed examination techniques for diagnosing hypovolemic states and concluded that postural changes in heart rate and blood pressure were the most accurate; capillary refill was not recommended. Stevenson and Perloff15 demonstrated that crackles and elevated JVP were absent in 18 of 43 patients with pulmonary capillary wedge pressures >22 mmHg. Butman et al.16 showed that elevated JVP was 82% accurate for predicting a wedge pressure >18 mmHg. Connors et al.17 demonstrated that clinicians' predictions of heart filling pressures and cardiac output were accurate (relative to pulmonary artery catheter measurements) in less than 50% of cases, though the examination techniques used were not qualified or quantified. No previous study has combined simple, semiobjective physical examination techniques for the purpose of distinguishing categories of shock.

Since identification of the pathogenesis of shock has important treatment/prognostic implications (eg, fluid and vasopressor therapies, early search for drainable focus of infection in sepsis, reestablishing vessel patency in myocardial infarction and pulmonary embolus), we believe that this simple, rapidly administered algorithm will prove useful in clinical medicine. In some clinical situations, the approach can lead to timely identification of the causative mechanism, allowing prompt definitive treatment. For example, a patient presenting with high‐output hypotension is so often sepsis/septic shock that treatment with antibiotics is justified (since success is time‐sensitive) even when the exact site/microbe has not yet been identified. Acute right heart overfilled low‐output hypotension should be considered pulmonary embolism (which also requires time‐sensitive therapies) until proven otherwise. Yet, a sizeable number of cases do not fit neatly into a single category. For example, 11% of patients with septic shock presented with cool extremities in the early phases of illness. In clinical decision‐making, 2 diagnostic‐therapeutic paradigms are common. In the first, the diagnosis is relatively certain and narrowly‐directed, mechanism‐specific treatment is appropriate. The second paradigm is 1 of significant uncertainty, when clinicians must treat empirically the most likely causes until more data become available to permit safe narrowing of therapies. For example, a patient presenting with hypotension, cool extremities, leukocytosis and apparent pneumonia should be treated empirically for septic shock while exploring explanations for the incongruous low‐output state (eg, profound hypovolemia, adrenal insufficiency, concurrent or precedent myocardial dysfunction). Patients often have several mechanisms contributing to hypotension. Since patients are not ideal forms, there can be no perfect decision‐tool; clinicians would be fool‐hardy to prematurely close decision‐making prior to definitive diagnosis. In the case of shock, such diagnostic arrogance would delay time‐sensitive therapies and thus contribute to morbidity and mortality. Nonetheless, this physical examination algorithmunderstanding its operating characteristics and limitationsmay add to the bedside clinician's diagnostic armamentarium.

Our study has several notable limitations. First, bedside examinations were performed by multiple observers who had limited (1 electronic mail) instruction on how to perform and document the data gathered for this study. So these results should be generalized cautiously until reproduced at other centers with greater numbers of observers (than the 28 of this study). The central supposition, that skin cooler, capillary refill longer, and pulse pressure more narrow than theirs, presupposes reasonable homogeneity of the normal state which is not necessarily true.11 Interobserver variability of physical examination further compromises the fidelity of findings recorded for this study.13 Since we conducted a retrospective review, and because of the emergency nature of the clinical problem, it would be difficult to conduct a study in which multiple examiners performed the same physical examinations to quantify interobserver variability. Irrespective, we would expect interobserver variability to systematically reduce accuracy; it is all‐the‐more impressive that trainees' examination results correctly diagnosed mechanism of shock in three‐quarters of cases. Also, examiners were not blinded to clinical history, so results of their examination could have been biased by their pre‐examination hypotheses of pathogenesis. Of course, they were not aware of the expert's final categorization of mechanism performed much later in time. Since there is no absolute reference standard for classification of the pathogenesis of shock, we depended upon careful review of selected data (same parameters for each patient) by a single senior investigatoralbeit armed with evidence‐based or consensus‐based standards of diagnosing shock. Finally, it can be argued that all forms of shock are mixed (with hypovolemia) early in the course; sepsis requires refilling of a leaky and dilated vasculature and the noncompliant ischemic ventricle often requires a higher filling pressure than normal to empty. To complicate even more, patients may have preexistent conditions (eg, chronic congestive heart failure, cirrhosis) that limit cardiovascular responses to acute shock. Our diagnostic approach was to identify the principal cause of the acute decompensation, assuming that many patients will have more than 1 single mechanism accounting for hypotension.

In conclusion, this is the first study to examine the utility of this simple physical examination algorithm to diagnose the mechanism of shock. Some have discounted or underemphasized examination techniques in favor of more time‐intensive and labor‐intensive diagnostic modalities, such as bedside echocardiography, which may waste precious time and resources. The simple physical examination algorithm assessed in this study has favorable operating characteristics and can be performed readily by even novice clinicians. If replicated at other centers and by greater numbers of observers, this approach could assist clinicians and teachers who train clinicians to rapidly diagnose and manage patients with shock.

Medication reconciliation is integral to reducing medication errors surrounding hospitalizations.1, 2 The practice of medication reconciliation requires a systematic and comprehensive review of all the medications a patient is currently taking to ensure that medications being added, changed, or discontinued are carefully evaluated with the goal of maintaining an accurate list; that this process is undertaken at every transition along the continuum of care; and that an accurate list of medications is available to the patient or family/caregiver and all providers involved in the patient's care, especially when a care handoff takes place. With regulators, payers and the public increasingly demanding action to reduce medication errors in hospitals, all health care providers must support efforts to achieve accurate medication reconciliation.1, 3

While conceptually straightforward, implementing medication reconciliation has proved to be very difficult in the myriad healthcare settings that exist. The disjointed nature of the American health care system and a conglomeration of paper and electronic systems for tracking medications synergize to thwart efforts to maintain an accurate, up‐to‐date medication list at every step along the care continuum. Although The Joint Commission defines medication for the purpose of its accreditation standards (see box), the healthcare community lacks a common understanding or agreement regarding what constitutes a medication. There is also confusion about who should ultimately be responsible for obtaining the patient's medication information, for performing the various steps in the reconciliation process, and for managing the multiple providers who alter the medication list but may not feel competent to perform reconciliation of medications outside their area of expertise safely. Importantly, there is also a lack of clarity around how patients and family/caregivers should be involved in the process.

Despite these challenges, medication reconciliation remains a critical patient safety activity that is supported by the organizations signing this consensus statement, (Table 1). Although medication reconciliation has an impact on medication safety in all care settings, this paper focuses on issues most germane to the continuum of care involving the hospital setting. The themes and issues discussed will likely apply to other care settings as well. In this paper, we also recommend several concrete steps that we believe should be initiated immediately to begin to reach the goal of optimizing the medication safety achievable through effective medication reconciliation.

Background

Medication reconciliation is intended to be a systematic extension of the medication history‐taking process that has been used by health care providers for decades. Its recent iteration was developed to ensure that medications were not added, omitted, or changed inadvertently during care transitions. It became codified, refined, and tested over the past decade through the efforts of a number of groups focused on medication safety including the Institute for Healthcare Improvement (IHI) and the Institute for Safe Medication Practices (ISMP). With the reinforcing adoption of medication reconciliation as National Patient Safety Goal (NPSG) No. 8 in 2005 by The Joint Commission, efforts to implement it became widespread in both hospital‐based and ambulatory settings.

Medication reconciliation has three steps, as described by IHI4:

• Verification (collection of the patient's medication history);

• Clarification (ensuring that the medications and doses are appropriate); and

• Reconciliation (documentation of changes in the orders).

The details of the process vary by setting and by the availability of paper or electronic medical records. However, the essential steps remain the same, as does the need to perform reconciliation each time the patient transfers to a new setting or level of care. Table 2 lists the most common points at which medication reconciliation occurs in hospitalized patients.

Because of their complexity, organizations must take care to design their medication reconciliation processes systematically. IHI lists elements of a well‐designed medication reconciliation process as part of its 5 Million Lives Campaign How‐to Guide.4 Such a process:

• Uses a patient centered approach.

• Makes it easy to complete the process for all involved. Staff members recognize the what's‐in‐it‐for‐me aspect of the change.

• Minimizes the opportunity for drug interactions and therapeutic duplications by making the patient's list of current medications available when clinicians prescribe new medications.

• Provides the patient with an up‐to‐date list of medications.

• Ensures that other providers who need to know have information about changes in a patient's medication plan.

Research on how adverse drug events (ADE) occur supports the need for tight control of medication orders at transitions in care. For instance:

• In a study conducted at Mayo Health System in Wisconsin, poor communication of medical information at transition points was responsible for as many as 50% of all medication errors in the hospital and up to 20% of ADEs.5

• Variances between the medications patients were taking prior to admission and their admission orders ranged from 30% to 70% in 2 literature reviews.1, 6

• The largest study of medication reconciliation errors and risk factors at hospital admission documented that 36% of patients had errors in their admission orders.7

When The Joint Commission adopted medication reconciliation as NPSG No. 8 in 2005 it had 2 parts: Requirement 8Aa process must exist for comparing the patient's current medications with those ordered for the patient while under the care of the organization; and requirement 8Ba complete list of the patient's medications must be communicated to the next provider of service on transfer within or outside the organization and a complete list of medications must be provided to the patient on discharge.8

However, many hospitals found it difficult to implement medication reconciliation in a systematic way. There was also confusion among hospital staff and administration about the exact definition of medication reconciliation in terms of what it should entail.9 Given these difficulties, The Joint Commission announced that effective January 1, 2009, medication reconciliation would no longer be factored into an organization's accreditation decision or be considered for Requirements for Improvement. Additionally, The Joint Commission stated it is reviewing and revising the NPSG so that it will be ready to be released in January 2011 for implementation later that year.10

Recognizing the difficulty hospitals were having with meaningfully implementing medication reconciliation, the Society of Hospital Medicine convened a 1‐day conference on March 6, 2009, to obtain input from key stakeholders and focus on several critical domains relevant to the success of hospital‐based medication reconciliation. The Agency for Healthcare Research and Quality provided funding support for this conference through grant 1R13HS017520‐01.

An overarching theme emerged from the meeting: the need to reorient the focus of medication reconciliation away from that of an accreditation mandate and toward a broader view of patient safety. Forcing medication reconciliation via a requirement for accreditation tended to limit an organization's efforts to specific process measures. Addressing it as a more global patient safety issue takes into account the entire patient care experience and then opens the door to leverage nonclinical venues (e.g., medical home, family home, community, religious, and other social organizations, as well as social networking platforms) and engage the patient and family/caregivers to reinforce the importance of medication safety.

This white paper evolved from discussions at the March 2009 conference,11 and subsequent structured communication among attendees. Formal endorsement of this document was obtained from the organizations listed in Table 1. In this document, we explore several key issues in implementing clinically meaningful and patient‐centered medication reconciliation. We focus on building common language and understanding of the processes of and participants in medication reconciliation; consider issues of implementation and risk stratification; emphasize the need for research to identify best practices and discusses how to disseminate the findings; promote health information technology platforms that will support interoperable medication information exchange; support the formation of partnerships between patient care sites and nonclinical sites as well as utilizing social marketing opportunities to enhance opportunities for transmitting messages about medication safety; and reinforce the ongoing healthcare reform discussion which aims to align financial incentives with patient safety efforts. After each section, we offer concrete first steps to address the issues discussed.

Methods

The invitation‐only meeting held on the Northwestern Medical Campus in Chicago, IL, brought together stakeholders representing professional, clinical, health care quality, consumer, and regulatory organizations (Table 3). The conference convened these participants with the goals of identifying barriers to meaningful implementation of medication reconciliation and developing a feasible plan toward its effective implementation in the hospital setting. At the meeting, all participants were divided into 1 of 4 groups, which held a facilitated discussion around 1 of 4 key relevant domains: (1) how to measure success in medication reconciliation; (2) key elements of successful strategies; (3) leveraging partnerships outside the hospital setting to support medication reconciliation; and (4) the roles of the patient and family/caregivers and health literacy. Individual group discussions were cofacilitated by experts in the content area. After each discussion, the small group then rotated to a different discussion. Ultimately, each group participated in all four discussions, which built iteratively on the content derived from the prior groups' insights. Key comments were then shared with the large group for further discussion. To help build consensus, these large group discussions were directed by professional facilitators.

After the meeting, attendees participated in 2 follow‐up conference calls to discuss issues raised at the conference and responses obtained from host organizations. They also subsequently participated in two focus groups with The Joint Commission, giving input on the revision of the medication reconciliation NPSG.

Results

Conclusion

Medication reconciliation involves highly complex processes and is hampered by the disjointed nature of the American health care system. It is, however, a vital part of reducing ADE. If employed more broadly, it has the added benefits of enhancing communication among all providers of care and engaging patients and families/caregivers more consistently and meaningfully in their overall care.

Despite the difficulty of maintaining an accurate medication record in real time across disparate settings, reconciliation is a goal to which our organizations are committed. Given the wide range of healthcare organizations involved in providing medications to patients and the many agencies evaluating those efforts, we believed it would be helpful to provide an overarching set of goals to move medication reconciliation forward.

Our main message is this: Patient safety and patient/family‐centered care must be the principal drivers in the development and implementation of medication reconciliation systems. Ultimately this process is about ensuring that patients are receiving the most appropriate medications no matter where they are treated. With this document, we hope to bring to light the importance of creating and implementing a medication reconciliation program, addressing some barriers to success, and identifying potential solutions that will ensure utility and sustainability of this critical patient safety issue.

Medication reconciliation is integral to reducing medication errors surrounding hospitalizations.1, 2 The practice of medication reconciliation requires a systematic and comprehensive review of all the medications a patient is currently taking to ensure that medications being added, changed, or discontinued are carefully evaluated with the goal of maintaining an accurate list; that this process is undertaken at every transition along the continuum of care; and that an accurate list of medications is available to the patient or family/caregiver and all providers involved in the patient's care, especially when a care handoff takes place. With regulators, payers and the public increasingly demanding action to reduce medication errors in hospitals, all health care providers must support efforts to achieve accurate medication reconciliation.1, 3

While conceptually straightforward, implementing medication reconciliation has proved to be very difficult in the myriad healthcare settings that exist. The disjointed nature of the American health care system and a conglomeration of paper and electronic systems for tracking medications synergize to thwart efforts to maintain an accurate, up‐to‐date medication list at every step along the care continuum. Although The Joint Commission defines medication for the purpose of its accreditation standards (see box), the healthcare community lacks a common understanding or agreement regarding what constitutes a medication. There is also confusion about who should ultimately be responsible for obtaining the patient's medication information, for performing the various steps in the reconciliation process, and for managing the multiple providers who alter the medication list but may not feel competent to perform reconciliation of medications outside their area of expertise safely. Importantly, there is also a lack of clarity around how patients and family/caregivers should be involved in the process.

Despite these challenges, medication reconciliation remains a critical patient safety activity that is supported by the organizations signing this consensus statement, (Table 1). Although medication reconciliation has an impact on medication safety in all care settings, this paper focuses on issues most germane to the continuum of care involving the hospital setting. The themes and issues discussed will likely apply to other care settings as well. In this paper, we also recommend several concrete steps that we believe should be initiated immediately to begin to reach the goal of optimizing the medication safety achievable through effective medication reconciliation.

Background

Medication reconciliation is intended to be a systematic extension of the medication history‐taking process that has been used by health care providers for decades. Its recent iteration was developed to ensure that medications were not added, omitted, or changed inadvertently during care transitions. It became codified, refined, and tested over the past decade through the efforts of a number of groups focused on medication safety including the Institute for Healthcare Improvement (IHI) and the Institute for Safe Medication Practices (ISMP). With the reinforcing adoption of medication reconciliation as National Patient Safety Goal (NPSG) No. 8 in 2005 by The Joint Commission, efforts to implement it became widespread in both hospital‐based and ambulatory settings.

Medication reconciliation has three steps, as described by IHI4:

• Verification (collection of the patient's medication history);

• Clarification (ensuring that the medications and doses are appropriate); and

• Reconciliation (documentation of changes in the orders).

The details of the process vary by setting and by the availability of paper or electronic medical records. However, the essential steps remain the same, as does the need to perform reconciliation each time the patient transfers to a new setting or level of care. Table 2 lists the most common points at which medication reconciliation occurs in hospitalized patients.

Because of their complexity, organizations must take care to design their medication reconciliation processes systematically. IHI lists elements of a well‐designed medication reconciliation process as part of its 5 Million Lives Campaign How‐to Guide.4 Such a process:

• Uses a patient centered approach.

• Makes it easy to complete the process for all involved. Staff members recognize the what's‐in‐it‐for‐me aspect of the change.

• Minimizes the opportunity for drug interactions and therapeutic duplications by making the patient's list of current medications available when clinicians prescribe new medications.

• Provides the patient with an up‐to‐date list of medications.

• Ensures that other providers who need to know have information about changes in a patient's medication plan.

Research on how adverse drug events (ADE) occur supports the need for tight control of medication orders at transitions in care. For instance:

• In a study conducted at Mayo Health System in Wisconsin, poor communication of medical information at transition points was responsible for as many as 50% of all medication errors in the hospital and up to 20% of ADEs.5

• Variances between the medications patients were taking prior to admission and their admission orders ranged from 30% to 70% in 2 literature reviews.1, 6

• The largest study of medication reconciliation errors and risk factors at hospital admission documented that 36% of patients had errors in their admission orders.7

When The Joint Commission adopted medication reconciliation as NPSG No. 8 in 2005 it had 2 parts: Requirement 8Aa process must exist for comparing the patient's current medications with those ordered for the patient while under the care of the organization; and requirement 8Ba complete list of the patient's medications must be communicated to the next provider of service on transfer within or outside the organization and a complete list of medications must be provided to the patient on discharge.8

However, many hospitals found it difficult to implement medication reconciliation in a systematic way. There was also confusion among hospital staff and administration about the exact definition of medication reconciliation in terms of what it should entail.9 Given these difficulties, The Joint Commission announced that effective January 1, 2009, medication reconciliation would no longer be factored into an organization's accreditation decision or be considered for Requirements for Improvement. Additionally, The Joint Commission stated it is reviewing and revising the NPSG so that it will be ready to be released in January 2011 for implementation later that year.10

Recognizing the difficulty hospitals were having with meaningfully implementing medication reconciliation, the Society of Hospital Medicine convened a 1‐day conference on March 6, 2009, to obtain input from key stakeholders and focus on several critical domains relevant to the success of hospital‐based medication reconciliation. The Agency for Healthcare Research and Quality provided funding support for this conference through grant 1R13HS017520‐01.

An overarching theme emerged from the meeting: the need to reorient the focus of medication reconciliation away from that of an accreditation mandate and toward a broader view of patient safety. Forcing medication reconciliation via a requirement for accreditation tended to limit an organization's efforts to specific process measures. Addressing it as a more global patient safety issue takes into account the entire patient care experience and then opens the door to leverage nonclinical venues (e.g., medical home, family home, community, religious, and other social organizations, as well as social networking platforms) and engage the patient and family/caregivers to reinforce the importance of medication safety.

This white paper evolved from discussions at the March 2009 conference,11 and subsequent structured communication among attendees. Formal endorsement of this document was obtained from the organizations listed in Table 1. In this document, we explore several key issues in implementing clinically meaningful and patient‐centered medication reconciliation. We focus on building common language and understanding of the processes of and participants in medication reconciliation; consider issues of implementation and risk stratification; emphasize the need for research to identify best practices and discusses how to disseminate the findings; promote health information technology platforms that will support interoperable medication information exchange; support the formation of partnerships between patient care sites and nonclinical sites as well as utilizing social marketing opportunities to enhance opportunities for transmitting messages about medication safety; and reinforce the ongoing healthcare reform discussion which aims to align financial incentives with patient safety efforts. After each section, we offer concrete first steps to address the issues discussed.

Methods

The invitation‐only meeting held on the Northwestern Medical Campus in Chicago, IL, brought together stakeholders representing professional, clinical, health care quality, consumer, and regulatory organizations (Table 3). The conference convened these participants with the goals of identifying barriers to meaningful implementation of medication reconciliation and developing a feasible plan toward its effective implementation in the hospital setting. At the meeting, all participants were divided into 1 of 4 groups, which held a facilitated discussion around 1 of 4 key relevant domains: (1) how to measure success in medication reconciliation; (2) key elements of successful strategies; (3) leveraging partnerships outside the hospital setting to support medication reconciliation; and (4) the roles of the patient and family/caregivers and health literacy. Individual group discussions were cofacilitated by experts in the content area. After each discussion, the small group then rotated to a different discussion. Ultimately, each group participated in all four discussions, which built iteratively on the content derived from the prior groups' insights. Key comments were then shared with the large group for further discussion. To help build consensus, these large group discussions were directed by professional facilitators.

After the meeting, attendees participated in 2 follow‐up conference calls to discuss issues raised at the conference and responses obtained from host organizations. They also subsequently participated in two focus groups with The Joint Commission, giving input on the revision of the medication reconciliation NPSG.

Results

Conclusion

Medication reconciliation involves highly complex processes and is hampered by the disjointed nature of the American health care system. It is, however, a vital part of reducing ADE. If employed more broadly, it has the added benefits of enhancing communication among all providers of care and engaging patients and families/caregivers more consistently and meaningfully in their overall care.

Despite the difficulty of maintaining an accurate medication record in real time across disparate settings, reconciliation is a goal to which our organizations are committed. Given the wide range of healthcare organizations involved in providing medications to patients and the many agencies evaluating those efforts, we believed it would be helpful to provide an overarching set of goals to move medication reconciliation forward.

Our main message is this: Patient safety and patient/family‐centered care must be the principal drivers in the development and implementation of medication reconciliation systems. Ultimately this process is about ensuring that patients are receiving the most appropriate medications no matter where they are treated. With this document, we hope to bring to light the importance of creating and implementing a medication reconciliation program, addressing some barriers to success, and identifying potential solutions that will ensure utility and sustainability of this critical patient safety issue.

Medication reconciliation is integral to reducing medication errors surrounding hospitalizations.1, 2 The practice of medication reconciliation requires a systematic and comprehensive review of all the medications a patient is currently taking to ensure that medications being added, changed, or discontinued are carefully evaluated with the goal of maintaining an accurate list; that this process is undertaken at every transition along the continuum of care; and that an accurate list of medications is available to the patient or family/caregiver and all providers involved in the patient's care, especially when a care handoff takes place. With regulators, payers and the public increasingly demanding action to reduce medication errors in hospitals, all health care providers must support efforts to achieve accurate medication reconciliation.1, 3

While conceptually straightforward, implementing medication reconciliation has proved to be very difficult in the myriad healthcare settings that exist. The disjointed nature of the American health care system and a conglomeration of paper and electronic systems for tracking medications synergize to thwart efforts to maintain an accurate, up‐to‐date medication list at every step along the care continuum. Although The Joint Commission defines medication for the purpose of its accreditation standards (see box), the healthcare community lacks a common understanding or agreement regarding what constitutes a medication. There is also confusion about who should ultimately be responsible for obtaining the patient's medication information, for performing the various steps in the reconciliation process, and for managing the multiple providers who alter the medication list but may not feel competent to perform reconciliation of medications outside their area of expertise safely. Importantly, there is also a lack of clarity around how patients and family/caregivers should be involved in the process.

Despite these challenges, medication reconciliation remains a critical patient safety activity that is supported by the organizations signing this consensus statement, (Table 1). Although medication reconciliation has an impact on medication safety in all care settings, this paper focuses on issues most germane to the continuum of care involving the hospital setting. The themes and issues discussed will likely apply to other care settings as well. In this paper, we also recommend several concrete steps that we believe should be initiated immediately to begin to reach the goal of optimizing the medication safety achievable through effective medication reconciliation.

Background

Medication reconciliation is intended to be a systematic extension of the medication history‐taking process that has been used by health care providers for decades. Its recent iteration was developed to ensure that medications were not added, omitted, or changed inadvertently during care transitions. It became codified, refined, and tested over the past decade through the efforts of a number of groups focused on medication safety including the Institute for Healthcare Improvement (IHI) and the Institute for Safe Medication Practices (ISMP). With the reinforcing adoption of medication reconciliation as National Patient Safety Goal (NPSG) No. 8 in 2005 by The Joint Commission, efforts to implement it became widespread in both hospital‐based and ambulatory settings.

Medication reconciliation has three steps, as described by IHI4:

• Verification (collection of the patient's medication history);

• Clarification (ensuring that the medications and doses are appropriate); and

• Reconciliation (documentation of changes in the orders).

The details of the process vary by setting and by the availability of paper or electronic medical records. However, the essential steps remain the same, as does the need to perform reconciliation each time the patient transfers to a new setting or level of care. Table 2 lists the most common points at which medication reconciliation occurs in hospitalized patients.

Because of their complexity, organizations must take care to design their medication reconciliation processes systematically. IHI lists elements of a well‐designed medication reconciliation process as part of its 5 Million Lives Campaign How‐to Guide.4 Such a process:

• Uses a patient centered approach.

• Makes it easy to complete the process for all involved. Staff members recognize the what's‐in‐it‐for‐me aspect of the change.

• Minimizes the opportunity for drug interactions and therapeutic duplications by making the patient's list of current medications available when clinicians prescribe new medications.

• Provides the patient with an up‐to‐date list of medications.

• Ensures that other providers who need to know have information about changes in a patient's medication plan.

Research on how adverse drug events (ADE) occur supports the need for tight control of medication orders at transitions in care. For instance:

• In a study conducted at Mayo Health System in Wisconsin, poor communication of medical information at transition points was responsible for as many as 50% of all medication errors in the hospital and up to 20% of ADEs.5

• Variances between the medications patients were taking prior to admission and their admission orders ranged from 30% to 70% in 2 literature reviews.1, 6

• The largest study of medication reconciliation errors and risk factors at hospital admission documented that 36% of patients had errors in their admission orders.7

When The Joint Commission adopted medication reconciliation as NPSG No. 8 in 2005 it had 2 parts: Requirement 8Aa process must exist for comparing the patient's current medications with those ordered for the patient while under the care of the organization; and requirement 8Ba complete list of the patient's medications must be communicated to the next provider of service on transfer within or outside the organization and a complete list of medications must be provided to the patient on discharge.8

However, many hospitals found it difficult to implement medication reconciliation in a systematic way. There was also confusion among hospital staff and administration about the exact definition of medication reconciliation in terms of what it should entail.9 Given these difficulties, The Joint Commission announced that effective January 1, 2009, medication reconciliation would no longer be factored into an organization's accreditation decision or be considered for Requirements for Improvement. Additionally, The Joint Commission stated it is reviewing and revising the NPSG so that it will be ready to be released in January 2011 for implementation later that year.10

Recognizing the difficulty hospitals were having with meaningfully implementing medication reconciliation, the Society of Hospital Medicine convened a 1‐day conference on March 6, 2009, to obtain input from key stakeholders and focus on several critical domains relevant to the success of hospital‐based medication reconciliation. The Agency for Healthcare Research and Quality provided funding support for this conference through grant 1R13HS017520‐01.

An overarching theme emerged from the meeting: the need to reorient the focus of medication reconciliation away from that of an accreditation mandate and toward a broader view of patient safety. Forcing medication reconciliation via a requirement for accreditation tended to limit an organization's efforts to specific process measures. Addressing it as a more global patient safety issue takes into account the entire patient care experience and then opens the door to leverage nonclinical venues (e.g., medical home, family home, community, religious, and other social organizations, as well as social networking platforms) and engage the patient and family/caregivers to reinforce the importance of medication safety.

This white paper evolved from discussions at the March 2009 conference,11 and subsequent structured communication among attendees. Formal endorsement of this document was obtained from the organizations listed in Table 1. In this document, we explore several key issues in implementing clinically meaningful and patient‐centered medication reconciliation. We focus on building common language and understanding of the processes of and participants in medication reconciliation; consider issues of implementation and risk stratification; emphasize the need for research to identify best practices and discusses how to disseminate the findings; promote health information technology platforms that will support interoperable medication information exchange; support the formation of partnerships between patient care sites and nonclinical sites as well as utilizing social marketing opportunities to enhance opportunities for transmitting messages about medication safety; and reinforce the ongoing healthcare reform discussion which aims to align financial incentives with patient safety efforts. After each section, we offer concrete first steps to address the issues discussed.

Methods

The invitation‐only meeting held on the Northwestern Medical Campus in Chicago, IL, brought together stakeholders representing professional, clinical, health care quality, consumer, and regulatory organizations (Table 3). The conference convened these participants with the goals of identifying barriers to meaningful implementation of medication reconciliation and developing a feasible plan toward its effective implementation in the hospital setting. At the meeting, all participants were divided into 1 of 4 groups, which held a facilitated discussion around 1 of 4 key relevant domains: (1) how to measure success in medication reconciliation; (2) key elements of successful strategies; (3) leveraging partnerships outside the hospital setting to support medication reconciliation; and (4) the roles of the patient and family/caregivers and health literacy. Individual group discussions were cofacilitated by experts in the content area. After each discussion, the small group then rotated to a different discussion. Ultimately, each group participated in all four discussions, which built iteratively on the content derived from the prior groups' insights. Key comments were then shared with the large group for further discussion. To help build consensus, these large group discussions were directed by professional facilitators.

After the meeting, attendees participated in 2 follow‐up conference calls to discuss issues raised at the conference and responses obtained from host organizations. They also subsequently participated in two focus groups with The Joint Commission, giving input on the revision of the medication reconciliation NPSG.

Results

Conclusion

Medication reconciliation involves highly complex processes and is hampered by the disjointed nature of the American health care system. It is, however, a vital part of reducing ADE. If employed more broadly, it has the added benefits of enhancing communication among all providers of care and engaging patients and families/caregivers more consistently and meaningfully in their overall care.

Despite the difficulty of maintaining an accurate medication record in real time across disparate settings, reconciliation is a goal to which our organizations are committed. Given the wide range of healthcare organizations involved in providing medications to patients and the many agencies evaluating those efforts, we believed it would be helpful to provide an overarching set of goals to move medication reconciliation forward.

Our main message is this: Patient safety and patient/family‐centered care must be the principal drivers in the development and implementation of medication reconciliation systems. Ultimately this process is about ensuring that patients are receiving the most appropriate medications no matter where they are treated. With this document, we hope to bring to light the importance of creating and implementing a medication reconciliation program, addressing some barriers to success, and identifying potential solutions that will ensure utility and sustainability of this critical patient safety issue.

The decision to transfer a patient to the intensive care unit (ICU) from a general care setting is complex and based not only on clinical findings and patient wishes but also on the understanding that ICU resources are limited and costly.1 Adding to the decision‐making complexity is the knowledge that patients who transfer to an ICU from a general medical unit comprise the highest mortality group of ICU patients, with the mortality rate directly proportional to both the time spent on the general medical unit1, 2 and the number of physiologic abnormalities before ICU admission.3, 4

Prior studies have shown that cardiac arrest and unplanned (unexpected) transfers to the ICU are preceded by a period of physiologic instability reflected in the vital signs.510 However, vital signs alone may not accurately indicate clinical condition. For example, a person may be able to maintain normal blood pressure and heart rate despite severe illness or may have abnormal vital signs at baseline, which may be the case for an otherwise healthy young woman who has baseline low systolic blood pressure. Also, noncritical conditions commonly seen in hospitalized medical or surgical patients, such as anxiety or pain, may increase the respiratory rate or heart rate. Conversely, certain common medications, such as ‐blockers, may mask or blunt the normal physiologic response to illness. Overall, the prevalence of abnormal physiologic variables is high among hospitalized adult patients irrespective of the presence of serious adverse events.11 This prevalence may be a reason why 2 recent studies of inpatient medical emergency teams (METs) or rapid response teams (RRTs), which generally rely on vital signs for activation, failed to show a decrease in adult mortality rates.12, 13

Given the complexity of interpreting single vital sign readings, we evaluated a simple and clinically intuitive variable, the shock index (SI) (heart rate/systolic blood pressure, a noninvasive indication of left ventricular function),14 as a potential marker of the need for intensive care. Allgwer and Burri14 first developed the SI in studies of patients with acute blood loss, intraabdominal bleeding, fat emboli, and severe infections. They observed that a healthy adult had a mean SI of 0.54 (standard deviation [SD], 0.021), while an index of 1.0 indicated threatened shock and indices greater than 1.5 were seen in volume‐deficient shock.

We hypothesized that an elevated SI is a differentiating factor between a patient who had an unplanned ICU transfer and a general medical patient who did not require this higher level of care. To our knowledge, the SI has not been studied for this application previously.

Patients and Methods

Results

A total of 50 pairs of matching cases and controls was included in this study. Table 1 lists the source of admission, demographic characteristics, and numbers of deaths for cases and controls. There were no statistically significant differences in admission source, age, sex, ethnicity, admission care unit, or Charlson Comorbidity Index. Mean length of stay was 14.8 days (SD, 9.7 days) for the cases and 5.7 days (SD, 6.3 days; P < 0.001) for the controls. Admission diagnoses were classified on the basis of the organ system of involvement (Table 2). In 30 of 50 cases, the admission diagnosis and the reason for ICU transfer were related.

We reviewed the vital signs and shock indices for the 24 hours before ICU transfer for each case and over the entire hospitalization for each control, to determine the worst set (the lowest systolic blood pressure and the highest heart rate, respiratory rate, and SI). The cases had 1 to 22 complete sets of vitals for the 24 hours before ICU transfer; the median number of sets was 3 and the mean was 4. The controls had 1 to 12 complete sets for the 24 hours before the worst SI: the median was 3 sets and the mean was 3. In 26 of 50 controls, the worst SI occurred within the first 24 hours after admission. There was a significant difference between the median values of the worst shock indices of the cases and the controls (0.87 vs. 0.72; P < 0.005).

Table 3 shows the different values of the SI and the corresponding odds ratio of unplanned ICU transfer for cases compared with controls. The difference was significant at an SI of 0.85 and greater, indicating a strong association with unplanned ICU transfer.

We also found that the patients who transferred to the ICU had a greater number of inpatient deaths (9 cases vs. 1 control; P = 0.008), which would be expected, but there was no difference in 30‐day or 6‐month mortality rate (Table 1). One patient died after 30 days and while still hospitalized.

Comparison between the temporal trend of vital signs and the SI of the cases for the 24 hours before ICU transfer is shown in Figure 1. This graph shows the median of all the worst values (minimum systolic blood pressure and maximum SI, heart rate, and respiratory rate) over the four 6‐hour time periods (24 hours) preceding ICU transfer. Of note, the change in vital signs is subtle even while the SI increased to more than 0.8 as the patients clinically worsened before transfer.

Figure 1

Discussion

In our comparison of the SI of 50 patients who required unplanned (unexpected) transfer to the ICU with the SI of 50 matched controls who did not require this higher level of care, we found that a SI of 0.85 or greater was significantly associated with unplanned transfer to the ICU. The cases had a significantly higher worst SI than the controls, and they also had a significantly longer hospital stay and higher inpatient mortality rate, as would be expected for a sicker patient population. These findings are important given that the SI may be useful for assessing illness severity, for helping determine the need for transfer to the ICU, or for activating METs or RRTs.

A major problem with providing optimal care for hospitalized general medical patients is the inherent difficulty in determining illness severity and clinical decline, especially when the decline occurs gradually. Existing consensus recommendations for ICU admission include both specific diagnoses and arbitrary objective criteria based on abnormal vital signs and laboratory values.27 Also, individual institutions may have their own ICU admission requirements, which may differ from these or RRT criteria. Although vital signs are important as a snapshot of basic physiologic function, a number of noncritical conditions may lead to abnormal vital signs, and not all abnormal vital signs are associated with an adverse clinical event. By relying solely on vital signs, clinicians may not recognize critical illness and therefore not transfer a patient to the ICU or may inappropriately transfer a patient who does not need ICU‐level care.

Markers of illness severity other than vital signs, such as the Acute Physiology and Chronic Health Evaluation (APACHE) score, have been shown to predict the death of ICU patients17, 18 but have been rarely studied outside the ICU setting.19 Also, calculating the APACHE score is cumbersome, and there is no cutoff score that defines when a patient should be transferred to the ICU. Subbe et al.,20 in their study to identify critically ill patients, found that introduction of a physiological scoring system (including MET or RRT activation scores) would have identified only a small number of additional patients as critically ill. Another common marker of illness severity, the 4 criteria of the systemic inflammatory response syndrome (temperature <36C or >38C; heart rate >90 beats per minute; respiratory rate >20 breaths per minute or PCO₂ <32 mm Hg; and white blood cell count >12,000/L or <4000/L or with more than 10% band cells)21 may be too sensitive to use as a decision aid, since even a healthy person running after a bus could have 2 of the 4 criteria.22 Likewise, surgical patients may have transient leukocytosis due to a stress response independent of an infection.23

The SI may be more accurate than vital signs alone to determine illness severity and who is at risk for an unplanned transfer to the ICU. Birkhahn et al.24 concluded that the SI may be more useful in early hemorrhage than either heart rate or systolic blood pressure alone. Rady et al.25 showed that the SI used in the emergency department can identify critical illness with apparently stable vital signs, where an elevation of the SI above 0.9 was associated with an illness that was treated immediately with admission to the hospital and intensive therapy on admission. However, it is unclear whether the SI can be used to monitor ongoing treatment, because a previous study showed that the SI may be of limited value in the assessment of systemic oxygen transport and response to therapy in clinical septic shock.26 Of note, the SI is mostly independent of the effects of pain or anxiety, which cause a concurrent rise in heart rate and systolic blood pressure. Because the heart's left ventricular work is unchanged or may increase from the underlying catecholamine surge, the SI will be unchanged or may actually decrease.

Our study adds to the medical literature the findings that: (1) the SI may be useful as an indicator of illness severity and a triage tool in patients with no trauma but with various medical conditions, and (2) the SI showed a strong association with unplanned ICU transfer.

The main strength of our study is its case‐control design with matched controls. Also, by comparing groups from the same patient care unit, we sought to minimize the selection bias that can be inherent in case‐control studies. Limitations include the retrospective, nonrandomized study design and the fact that there may have been variations in vital sign measurements by the multiple caregivers. However, the vital signs were taken according to standard hospital practice and reflect real‐world conditions. Although generalizability may be somewhat limited because of our homogeneous patient population, our patients had a wide range of various medical illnesses, so our study should be applicable to other hospital settings, both academic and community‐based.

One of the main weaknesses of our study is that the results were not adjusted for the burden of comorbid conditions, although there were no statistically significant differences in the number of comorbid conditions among the cases and the controls (P = 0.96). Also, we did not directly compare the SI with vital signs alone to determine superiority.

The SI may be an important objective measure to help clinicians decide when patients need treatment that is more aggressive, assistance from a MET or an RRT, or a preemptive, rather than unplanned, transfer to an ICU. Although it is unlikely that a single measure will allow accurate triage of all medical or surgical patients, the SI may be a useful adjunct to clinical judgment and other objective measures in determining illness severity and clinical decline. Further prospective studies are needed to compare the role of the SI specifically with MET or RRT activation criteria, to clarify the role of comorbid conditions in unplanned transfers to the ICU, to validate the cut point for the SI in various disease states, and to assess its utility in patients with septic shock. Depending on these results, it may be beneficial to incorporate the SI into the electronic medical record as an automatic alert to identify patients at risk for ICU transfer.

Conclusions

The SI is an easily calculated composite index of heart rate and systolic blood pressure. An elevated SI of 0.85 can identify patients who are at risk for unplanned transfer to the ICU from general patient care units. Future studies will determine whether the SI is more accurate than simple vital signs as an indicator of clinical decline. If so, it may be useful as a trigger to activate METs or RRTs for treatment.

The decision to transfer a patient to the intensive care unit (ICU) from a general care setting is complex and based not only on clinical findings and patient wishes but also on the understanding that ICU resources are limited and costly.1 Adding to the decision‐making complexity is the knowledge that patients who transfer to an ICU from a general medical unit comprise the highest mortality group of ICU patients, with the mortality rate directly proportional to both the time spent on the general medical unit1, 2 and the number of physiologic abnormalities before ICU admission.3, 4

Prior studies have shown that cardiac arrest and unplanned (unexpected) transfers to the ICU are preceded by a period of physiologic instability reflected in the vital signs.510 However, vital signs alone may not accurately indicate clinical condition. For example, a person may be able to maintain normal blood pressure and heart rate despite severe illness or may have abnormal vital signs at baseline, which may be the case for an otherwise healthy young woman who has baseline low systolic blood pressure. Also, noncritical conditions commonly seen in hospitalized medical or surgical patients, such as anxiety or pain, may increase the respiratory rate or heart rate. Conversely, certain common medications, such as ‐blockers, may mask or blunt the normal physiologic response to illness. Overall, the prevalence of abnormal physiologic variables is high among hospitalized adult patients irrespective of the presence of serious adverse events.11 This prevalence may be a reason why 2 recent studies of inpatient medical emergency teams (METs) or rapid response teams (RRTs), which generally rely on vital signs for activation, failed to show a decrease in adult mortality rates.12, 13

Given the complexity of interpreting single vital sign readings, we evaluated a simple and clinically intuitive variable, the shock index (SI) (heart rate/systolic blood pressure, a noninvasive indication of left ventricular function),14 as a potential marker of the need for intensive care. Allgwer and Burri14 first developed the SI in studies of patients with acute blood loss, intraabdominal bleeding, fat emboli, and severe infections. They observed that a healthy adult had a mean SI of 0.54 (standard deviation [SD], 0.021), while an index of 1.0 indicated threatened shock and indices greater than 1.5 were seen in volume‐deficient shock.

We hypothesized that an elevated SI is a differentiating factor between a patient who had an unplanned ICU transfer and a general medical patient who did not require this higher level of care. To our knowledge, the SI has not been studied for this application previously.

Patients and Methods

Results

A total of 50 pairs of matching cases and controls was included in this study. Table 1 lists the source of admission, demographic characteristics, and numbers of deaths for cases and controls. There were no statistically significant differences in admission source, age, sex, ethnicity, admission care unit, or Charlson Comorbidity Index. Mean length of stay was 14.8 days (SD, 9.7 days) for the cases and 5.7 days (SD, 6.3 days; P < 0.001) for the controls. Admission diagnoses were classified on the basis of the organ system of involvement (Table 2). In 30 of 50 cases, the admission diagnosis and the reason for ICU transfer were related.

We reviewed the vital signs and shock indices for the 24 hours before ICU transfer for each case and over the entire hospitalization for each control, to determine the worst set (the lowest systolic blood pressure and the highest heart rate, respiratory rate, and SI). The cases had 1 to 22 complete sets of vitals for the 24 hours before ICU transfer; the median number of sets was 3 and the mean was 4. The controls had 1 to 12 complete sets for the 24 hours before the worst SI: the median was 3 sets and the mean was 3. In 26 of 50 controls, the worst SI occurred within the first 24 hours after admission. There was a significant difference between the median values of the worst shock indices of the cases and the controls (0.87 vs. 0.72; P < 0.005).

Table 3 shows the different values of the SI and the corresponding odds ratio of unplanned ICU transfer for cases compared with controls. The difference was significant at an SI of 0.85 and greater, indicating a strong association with unplanned ICU transfer.

We also found that the patients who transferred to the ICU had a greater number of inpatient deaths (9 cases vs. 1 control; P = 0.008), which would be expected, but there was no difference in 30‐day or 6‐month mortality rate (Table 1). One patient died after 30 days and while still hospitalized.

Comparison between the temporal trend of vital signs and the SI of the cases for the 24 hours before ICU transfer is shown in Figure 1. This graph shows the median of all the worst values (minimum systolic blood pressure and maximum SI, heart rate, and respiratory rate) over the four 6‐hour time periods (24 hours) preceding ICU transfer. Of note, the change in vital signs is subtle even while the SI increased to more than 0.8 as the patients clinically worsened before transfer.

Figure 1

Discussion

In our comparison of the SI of 50 patients who required unplanned (unexpected) transfer to the ICU with the SI of 50 matched controls who did not require this higher level of care, we found that a SI of 0.85 or greater was significantly associated with unplanned transfer to the ICU. The cases had a significantly higher worst SI than the controls, and they also had a significantly longer hospital stay and higher inpatient mortality rate, as would be expected for a sicker patient population. These findings are important given that the SI may be useful for assessing illness severity, for helping determine the need for transfer to the ICU, or for activating METs or RRTs.

A major problem with providing optimal care for hospitalized general medical patients is the inherent difficulty in determining illness severity and clinical decline, especially when the decline occurs gradually. Existing consensus recommendations for ICU admission include both specific diagnoses and arbitrary objective criteria based on abnormal vital signs and laboratory values.27 Also, individual institutions may have their own ICU admission requirements, which may differ from these or RRT criteria. Although vital signs are important as a snapshot of basic physiologic function, a number of noncritical conditions may lead to abnormal vital signs, and not all abnormal vital signs are associated with an adverse clinical event. By relying solely on vital signs, clinicians may not recognize critical illness and therefore not transfer a patient to the ICU or may inappropriately transfer a patient who does not need ICU‐level care.

Markers of illness severity other than vital signs, such as the Acute Physiology and Chronic Health Evaluation (APACHE) score, have been shown to predict the death of ICU patients17, 18 but have been rarely studied outside the ICU setting.19 Also, calculating the APACHE score is cumbersome, and there is no cutoff score that defines when a patient should be transferred to the ICU. Subbe et al.,20 in their study to identify critically ill patients, found that introduction of a physiological scoring system (including MET or RRT activation scores) would have identified only a small number of additional patients as critically ill. Another common marker of illness severity, the 4 criteria of the systemic inflammatory response syndrome (temperature <36C or >38C; heart rate >90 beats per minute; respiratory rate >20 breaths per minute or PCO₂ <32 mm Hg; and white blood cell count >12,000/L or <4000/L or with more than 10% band cells)21 may be too sensitive to use as a decision aid, since even a healthy person running after a bus could have 2 of the 4 criteria.22 Likewise, surgical patients may have transient leukocytosis due to a stress response independent of an infection.23

The SI may be more accurate than vital signs alone to determine illness severity and who is at risk for an unplanned transfer to the ICU. Birkhahn et al.24 concluded that the SI may be more useful in early hemorrhage than either heart rate or systolic blood pressure alone. Rady et al.25 showed that the SI used in the emergency department can identify critical illness with apparently stable vital signs, where an elevation of the SI above 0.9 was associated with an illness that was treated immediately with admission to the hospital and intensive therapy on admission. However, it is unclear whether the SI can be used to monitor ongoing treatment, because a previous study showed that the SI may be of limited value in the assessment of systemic oxygen transport and response to therapy in clinical septic shock.26 Of note, the SI is mostly independent of the effects of pain or anxiety, which cause a concurrent rise in heart rate and systolic blood pressure. Because the heart's left ventricular work is unchanged or may increase from the underlying catecholamine surge, the SI will be unchanged or may actually decrease.

Our study adds to the medical literature the findings that: (1) the SI may be useful as an indicator of illness severity and a triage tool in patients with no trauma but with various medical conditions, and (2) the SI showed a strong association with unplanned ICU transfer.

The main strength of our study is its case‐control design with matched controls. Also, by comparing groups from the same patient care unit, we sought to minimize the selection bias that can be inherent in case‐control studies. Limitations include the retrospective, nonrandomized study design and the fact that there may have been variations in vital sign measurements by the multiple caregivers. However, the vital signs were taken according to standard hospital practice and reflect real‐world conditions. Although generalizability may be somewhat limited because of our homogeneous patient population, our patients had a wide range of various medical illnesses, so our study should be applicable to other hospital settings, both academic and community‐based.

One of the main weaknesses of our study is that the results were not adjusted for the burden of comorbid conditions, although there were no statistically significant differences in the number of comorbid conditions among the cases and the controls (P = 0.96). Also, we did not directly compare the SI with vital signs alone to determine superiority.

The SI may be an important objective measure to help clinicians decide when patients need treatment that is more aggressive, assistance from a MET or an RRT, or a preemptive, rather than unplanned, transfer to an ICU. Although it is unlikely that a single measure will allow accurate triage of all medical or surgical patients, the SI may be a useful adjunct to clinical judgment and other objective measures in determining illness severity and clinical decline. Further prospective studies are needed to compare the role of the SI specifically with MET or RRT activation criteria, to clarify the role of comorbid conditions in unplanned transfers to the ICU, to validate the cut point for the SI in various disease states, and to assess its utility in patients with septic shock. Depending on these results, it may be beneficial to incorporate the SI into the electronic medical record as an automatic alert to identify patients at risk for ICU transfer.

Conclusions

The SI is an easily calculated composite index of heart rate and systolic blood pressure. An elevated SI of 0.85 can identify patients who are at risk for unplanned transfer to the ICU from general patient care units. Future studies will determine whether the SI is more accurate than simple vital signs as an indicator of clinical decline. If so, it may be useful as a trigger to activate METs or RRTs for treatment.

The decision to transfer a patient to the intensive care unit (ICU) from a general care setting is complex and based not only on clinical findings and patient wishes but also on the understanding that ICU resources are limited and costly.1 Adding to the decision‐making complexity is the knowledge that patients who transfer to an ICU from a general medical unit comprise the highest mortality group of ICU patients, with the mortality rate directly proportional to both the time spent on the general medical unit1, 2 and the number of physiologic abnormalities before ICU admission.3, 4

Prior studies have shown that cardiac arrest and unplanned (unexpected) transfers to the ICU are preceded by a period of physiologic instability reflected in the vital signs.510 However, vital signs alone may not accurately indicate clinical condition. For example, a person may be able to maintain normal blood pressure and heart rate despite severe illness or may have abnormal vital signs at baseline, which may be the case for an otherwise healthy young woman who has baseline low systolic blood pressure. Also, noncritical conditions commonly seen in hospitalized medical or surgical patients, such as anxiety or pain, may increase the respiratory rate or heart rate. Conversely, certain common medications, such as ‐blockers, may mask or blunt the normal physiologic response to illness. Overall, the prevalence of abnormal physiologic variables is high among hospitalized adult patients irrespective of the presence of serious adverse events.11 This prevalence may be a reason why 2 recent studies of inpatient medical emergency teams (METs) or rapid response teams (RRTs), which generally rely on vital signs for activation, failed to show a decrease in adult mortality rates.12, 13

Given the complexity of interpreting single vital sign readings, we evaluated a simple and clinically intuitive variable, the shock index (SI) (heart rate/systolic blood pressure, a noninvasive indication of left ventricular function),14 as a potential marker of the need for intensive care. Allgwer and Burri14 first developed the SI in studies of patients with acute blood loss, intraabdominal bleeding, fat emboli, and severe infections. They observed that a healthy adult had a mean SI of 0.54 (standard deviation [SD], 0.021), while an index of 1.0 indicated threatened shock and indices greater than 1.5 were seen in volume‐deficient shock.

We hypothesized that an elevated SI is a differentiating factor between a patient who had an unplanned ICU transfer and a general medical patient who did not require this higher level of care. To our knowledge, the SI has not been studied for this application previously.

Patients and Methods

Results

A total of 50 pairs of matching cases and controls was included in this study. Table 1 lists the source of admission, demographic characteristics, and numbers of deaths for cases and controls. There were no statistically significant differences in admission source, age, sex, ethnicity, admission care unit, or Charlson Comorbidity Index. Mean length of stay was 14.8 days (SD, 9.7 days) for the cases and 5.7 days (SD, 6.3 days; P < 0.001) for the controls. Admission diagnoses were classified on the basis of the organ system of involvement (Table 2). In 30 of 50 cases, the admission diagnosis and the reason for ICU transfer were related.

We reviewed the vital signs and shock indices for the 24 hours before ICU transfer for each case and over the entire hospitalization for each control, to determine the worst set (the lowest systolic blood pressure and the highest heart rate, respiratory rate, and SI). The cases had 1 to 22 complete sets of vitals for the 24 hours before ICU transfer; the median number of sets was 3 and the mean was 4. The controls had 1 to 12 complete sets for the 24 hours before the worst SI: the median was 3 sets and the mean was 3. In 26 of 50 controls, the worst SI occurred within the first 24 hours after admission. There was a significant difference between the median values of the worst shock indices of the cases and the controls (0.87 vs. 0.72; P < 0.005).

Table 3 shows the different values of the SI and the corresponding odds ratio of unplanned ICU transfer for cases compared with controls. The difference was significant at an SI of 0.85 and greater, indicating a strong association with unplanned ICU transfer.

We also found that the patients who transferred to the ICU had a greater number of inpatient deaths (9 cases vs. 1 control; P = 0.008), which would be expected, but there was no difference in 30‐day or 6‐month mortality rate (Table 1). One patient died after 30 days and while still hospitalized.

Comparison between the temporal trend of vital signs and the SI of the cases for the 24 hours before ICU transfer is shown in Figure 1. This graph shows the median of all the worst values (minimum systolic blood pressure and maximum SI, heart rate, and respiratory rate) over the four 6‐hour time periods (24 hours) preceding ICU transfer. Of note, the change in vital signs is subtle even while the SI increased to more than 0.8 as the patients clinically worsened before transfer.

Figure 1

Discussion

In our comparison of the SI of 50 patients who required unplanned (unexpected) transfer to the ICU with the SI of 50 matched controls who did not require this higher level of care, we found that a SI of 0.85 or greater was significantly associated with unplanned transfer to the ICU. The cases had a significantly higher worst SI than the controls, and they also had a significantly longer hospital stay and higher inpatient mortality rate, as would be expected for a sicker patient population. These findings are important given that the SI may be useful for assessing illness severity, for helping determine the need for transfer to the ICU, or for activating METs or RRTs.

A major problem with providing optimal care for hospitalized general medical patients is the inherent difficulty in determining illness severity and clinical decline, especially when the decline occurs gradually. Existing consensus recommendations for ICU admission include both specific diagnoses and arbitrary objective criteria based on abnormal vital signs and laboratory values.27 Also, individual institutions may have their own ICU admission requirements, which may differ from these or RRT criteria. Although vital signs are important as a snapshot of basic physiologic function, a number of noncritical conditions may lead to abnormal vital signs, and not all abnormal vital signs are associated with an adverse clinical event. By relying solely on vital signs, clinicians may not recognize critical illness and therefore not transfer a patient to the ICU or may inappropriately transfer a patient who does not need ICU‐level care.

Markers of illness severity other than vital signs, such as the Acute Physiology and Chronic Health Evaluation (APACHE) score, have been shown to predict the death of ICU patients17, 18 but have been rarely studied outside the ICU setting.19 Also, calculating the APACHE score is cumbersome, and there is no cutoff score that defines when a patient should be transferred to the ICU. Subbe et al.,20 in their study to identify critically ill patients, found that introduction of a physiological scoring system (including MET or RRT activation scores) would have identified only a small number of additional patients as critically ill. Another common marker of illness severity, the 4 criteria of the systemic inflammatory response syndrome (temperature <36C or >38C; heart rate >90 beats per minute; respiratory rate >20 breaths per minute or PCO₂ <32 mm Hg; and white blood cell count >12,000/L or <4000/L or with more than 10% band cells)21 may be too sensitive to use as a decision aid, since even a healthy person running after a bus could have 2 of the 4 criteria.22 Likewise, surgical patients may have transient leukocytosis due to a stress response independent of an infection.23

The SI may be more accurate than vital signs alone to determine illness severity and who is at risk for an unplanned transfer to the ICU. Birkhahn et al.24 concluded that the SI may be more useful in early hemorrhage than either heart rate or systolic blood pressure alone. Rady et al.25 showed that the SI used in the emergency department can identify critical illness with apparently stable vital signs, where an elevation of the SI above 0.9 was associated with an illness that was treated immediately with admission to the hospital and intensive therapy on admission. However, it is unclear whether the SI can be used to monitor ongoing treatment, because a previous study showed that the SI may be of limited value in the assessment of systemic oxygen transport and response to therapy in clinical septic shock.26 Of note, the SI is mostly independent of the effects of pain or anxiety, which cause a concurrent rise in heart rate and systolic blood pressure. Because the heart's left ventricular work is unchanged or may increase from the underlying catecholamine surge, the SI will be unchanged or may actually decrease.

Our study adds to the medical literature the findings that: (1) the SI may be useful as an indicator of illness severity and a triage tool in patients with no trauma but with various medical conditions, and (2) the SI showed a strong association with unplanned ICU transfer.

The main strength of our study is its case‐control design with matched controls. Also, by comparing groups from the same patient care unit, we sought to minimize the selection bias that can be inherent in case‐control studies. Limitations include the retrospective, nonrandomized study design and the fact that there may have been variations in vital sign measurements by the multiple caregivers. However, the vital signs were taken according to standard hospital practice and reflect real‐world conditions. Although generalizability may be somewhat limited because of our homogeneous patient population, our patients had a wide range of various medical illnesses, so our study should be applicable to other hospital settings, both academic and community‐based.

One of the main weaknesses of our study is that the results were not adjusted for the burden of comorbid conditions, although there were no statistically significant differences in the number of comorbid conditions among the cases and the controls (P = 0.96). Also, we did not directly compare the SI with vital signs alone to determine superiority.

The SI may be an important objective measure to help clinicians decide when patients need treatment that is more aggressive, assistance from a MET or an RRT, or a preemptive, rather than unplanned, transfer to an ICU. Although it is unlikely that a single measure will allow accurate triage of all medical or surgical patients, the SI may be a useful adjunct to clinical judgment and other objective measures in determining illness severity and clinical decline. Further prospective studies are needed to compare the role of the SI specifically with MET or RRT activation criteria, to clarify the role of comorbid conditions in unplanned transfers to the ICU, to validate the cut point for the SI in various disease states, and to assess its utility in patients with septic shock. Depending on these results, it may be beneficial to incorporate the SI into the electronic medical record as an automatic alert to identify patients at risk for ICU transfer.

Conclusions

The SI is an easily calculated composite index of heart rate and systolic blood pressure. An elevated SI of 0.85 can identify patients who are at risk for unplanned transfer to the ICU from general patient care units. Future studies will determine whether the SI is more accurate than simple vital signs as an indicator of clinical decline. If so, it may be useful as a trigger to activate METs or RRTs for treatment.

During their junior medicine rotation, our students are asked to post to Blackboard (an online student forum) an anonymous essay about an issue of professionalism or ethics, either inspiring or troubling. In many ways, these vignettes are like postcards, written by visitors describing foreign cultures and norms. They represent a way for the students to debrief, but also provide an opportunity for us, as faculty, to reflect upon the way we practice and teach medicine. Many postingslike postcards from exotic or historic placesare inspiring stories of residents and faculty extending themselves for their patients. Unfortunately, unlike typical postcards, there are also essays that are troubling or provoking and challenge us to consider how we could improve the professional and ethical environment on our teams.

In order to begin a learning process with our faculty and housestaff, we have presented a number of these anonymous essays at both faculty and housestaff Department of Medicine conferences as well as our monthly hospital Ethics conference. The goal of these conferences was to gather as a moral community to reflect on our students' experience and consider ways in which our day to day practice as attendings could be informed by what they tell us. In addition, the junior medicine site directors have a session each quarter with their junior students to review some of the most significant issues brought up by their essays.

Practically, these vignettes and conferences serve three main purposes:

• 1

  Raising Awareness: Many professional issues noted by our students occur under the radar. Attendings are often unaware of the issues of professionalism and/or ethics confronting our students and housestaff.

• 2

  Exploring Attitudes: Some attending may underemphasize the importance of specific issues of professionalism and/or ethics. Open discussions at faculty or resident conferences create opportunities for individuals to reflect upon their own reactions and for the group to create a norm.

• 3

  Sharing Skills: It is difficult to learn the practice of professionalism and ethics from a book. Skill in this area is gained primarily by experience. Conferences provide an excellent forum for seasoned physicians to share wisdom with less experienced physicians. In addition, important teaching points can be made: Students should not deliver bad news alone. Errors should be disclosed.

Following are 3 of the essays we presented, along with brief commentaries. At the end, we provide practical suggestions for individual attendings to improve the professional climate on their teams.

The Hospital Didn't Wait

Code. On 12, the surgical wards floor. Elise sprinted to the stairwell, dashed up to 12, and ran to the corner room as fast as she could. She could see the room before she got there. Instinctively, she started reviewing the steps she had memorized so many months ago. But when she finally arrived at the patient's bathroom, her thought process came to a jolting halt as she came upon the gruesome scene.

The 76‐year‐old patient had hanged himself with the cinching rope from his garment bag, and now dangled suspended from a high towel rack against the wall. Nurses from the floor started to file in, and without losing a beat Elise barked commands. Together they brought the man's body down to the floor, laid him on his back, and stripped off his hospital gown. Elise was in charge; deliberately but forcefully, she ordered a nurse to retrieve a defibrillator, and had another resident check for a pulse. There was none. Anesthesiology was here. Quickly and expertly, they shoved a plastic tube down his throat and began ventilation. The nurse placed on the electrodes between chest compressions then called to clear the body. Airway stepped back. The chest pumper stepped back. The body lurched forward as the defibrillator issued a long beep and discharged. Still no pulse. The cycle repeated.

Finally, Elise called a stop. Time of death, 19:37. By now there were about 20 people crammed in the patient room, all of whom had a separate role during the code. Some stayed behind, while the rest left to return to their interrupted work. The medical student didn't know what to think as he returned to the team room. His jaw was sorehad he been clenching it the whole time?and as he brought his hand up to rub his face, he saw that his knuckles were bloody. Somehow he had scraped them during the code. As he logged back into the computer to finish off his evening notes, he knew that he wouldn't have time to reflect until hours later when he returned home. Codes happened all the time. There was still work to be done in the hospital, and the hospital didn't wait.

The room had already been assigned to a patient waiting in the Emergency Department downstairs. That patient would be here in a few minutes. The hospital didn't wait.

When we presented this case in our conferences, there was universal agreement that such a traumatic event merits, even demands, team debriefing and processing. But in the real life aftermath of this traumatic event, the take‐home message for the medical student was that the hospital didn't wait for such discussions. We know this is not unique to our institution. In a study of 32 medical students who were asked to reflect on their most memorable patient death,1 debriefing sessions were rare and many students felt inadequately supported. While experienced clinicians may be accustomed to seeing patients die, students are new to the culture of the hospital, and have not had the chance to develop the defense mechanisms necessary to cope with this sort of experience. Angoff2 writes, As medical educators, we ought to ask our students how they are coping with long hours, fatigue, illness, suffering, and death. We ought to model and commend compassion and react to the deep feelings of our students in the same way we would teach them to react to the deep feelings of their patients.

I Told a Man Today That He Had Brain Cancer

The resident, intern, and I were huddled together in our team room when the report came back on the computer. New 3.5 2.3 1.7 cm contrast‐enhancing lesion seen anterior to genu of corpus callosum. Concerning for metastatic focus vs. lymphoma. Advise follow‐up. It wasn't unexpected but we had nevertheless been hoping for better.

The three of us went into his room and I was waiting to see how my resident would deliver the bad news, but she didn't. She simply said that we were continuing to do imaging studies and that a neurology team would be in touch. There were probably several reasons why she didn't tell him: not enough time, not her responsibility, or maybe she was just uncomfortable with it. Whatever the case, we left the room with my patient still oblivious to the awful mass now tangled in his head.

If my resident was taking a pass on this conversation, I knew it fell to me he needed to hear it from his primary team. I came back after rounds alone, sat down next to his bed, and told him that his MRI results had come back, and that I had unfortunate news.

I told him that the images showed that his lung cancer had spread to his brain.

I paused to give him a chance to let it sink in. He turned away and looked up at the ceiling.

Where is it? How big is it?

What now?

Reflecting on this case, our audiences were disturbed that a student would attempt this difficult conversation alone, while recognizing that the student clearly felt a sense of responsibility and desire to help his patient by sharing important information. We talked about how students may erroneously pick up a message that the team member who has spent the most time with a patient is the most obvious choice to have difficult conversations. We also noted that, unfortunately, sometimes students are directly asked by their team to shoulder this responsibility on their own. In this painful account, there is no mention of preparation, supervision, or support for the student before or after the encounter. The student perceived (rightly or wrongly) that the team leaders lacked comfort or skill to deliver the bad news, and stepped in. It is possible that the attending lacked the skill and ability to model an interaction, but more likely the deficit was in awareness and attitude. It is unlikely the attending knew that the student had this conversation alone. One of the major reasons we present these vignettes is to make attendings and housestaff more aware of issues that occur under their radar so that they can take preventative action. However, once the resident or attending found out that the student had this conversation alone, the student should be pulled aside for a 1:1 discussion. At the end of the day, the student should know that it was inappropriate to attempt this conversation alone

Rosenbaum3 reviewed a number of strategies to teach the skill of delivering bad news, from lecture and small group discussions to role play and standardized patients. When asked, students cited role‐modeling as the best way to learn how to deliver bad news.4 Observation of a veteran clinician provides a firm foundation for learning; but that is not enough. Unfortunately, we know from the literature (and our student vignettes suggest) that students and residents are unprepared to carry out these conversations properly, either because of misguided attitudes, lack of experience, or inadequate training.57 We conceptualize engaging in difficult conversations as a procedure, demanding a skill set. Mere observation of an expert executing this procedure is only a beginning. With any other skill, from successful completion of a lumbar puncture to initiating cardiopulmonary resuscitation (CPR), a student would never conclude that knowing the patient the best sufficiently credentials the student to undertake these procedures. We maintain that a difficult conversationbe it breaking bad news, discussing end‐of‐life care preferences, code status discussions, or prognosisis a clinical intervention, like any other procedure in medicine. If performed with skill and caution, it can bring about a stronger therapeutic relationship and increased support for the patient; if performed clumsily, it can lead to unintended adverse outcomes, including misunderstanding, mistrust, anxiety, and anger.