User login
A-fib and rate control: Don’t go too low
Aim for a heart rate of <110 beats per minute (bpm) in patients with permanent atrial fibrillation. Maintaining this rate requires less medication than more stringent rate control, resulting in fewer side effects and no increased risk of cardiovascular events.1
STRENGTH OF RECOMMENDATION
B: Based on 1 long-term randomized controlled trial (RCT).
Van Gelder IC, Groenveld HF, Crijns HJ, et al. Lenient versus strict rate control in patients with atrial fibrillation. N Engl J Med. 2010;362: 1363-1373.
Illustrative case
A 67-year-old man comes in for a follow-up visit after being hospitalized for atrial fibrillation with a rapid ventricular rate. Before being discharged, he was put on warfarin and metoprolol, and his heart rate today is 96 bpm. You consider increasing the dose of his beta-blocker. What should his target heart rate be?
Atrial fibrillation, the most common sustained arrhythmia,2 can lead to life-threatening events such as heart failure and stroke. Studies, including the Atrial Fibrillation Follow-Up Investigation of Rhythm Management (AFFIRM) and Rate Control versus Electrical Cardioversion (RACE) trials, have found no difference in morbidity or mortality between rate control and rhythm control strategies.2,3 Thus, rate control is usually preferred for patients with atrial fibrillation because of adverse effects associated with antiarrhythmic drugs.
Guidelines cite stringent targets
The American College of Cardiology/American Heart Association Task Force/European Society of Cardiology (ACC/AHA/ESC) guidelines make no definite recommendations about heart rate targets. The guidelines do indicate, however, that rate control criteria vary based on age, “but usually involve achieving ventricular rates between 60 and 80 [bpm] at rest and between 90 and 115 [bpm] during moderate exercise.”4
This guidance is based on data from epidemiologic studies suggesting that faster heart rates in sinus rhythm may increase mortality from cardiovascular causes.5 However, strict control often requires higher doses of rate-controlling medications, which can lead to adverse events such as symptomatic bradycardia, dizziness, and syncope, as well as pacemaker implantation.
Pooled data suggest a more relaxed rate is better
A retrospective analysis of pooled data from the rate-control arms of the AFFIRM and RACE trials found no difference in all-cause mortality between the more stringent rate-control group in AFFIRM and the more lenient control in RACE.6 This finding suggested that more lenient heart rate targets may be preferred to avoid the adverse effects often associated with the higher doses of rate-controlling drugs needed to achieve strict control. The Rate Control Efficacy in Permanent Atrial Fibrillation: a Comparison between Lenient versus Strict Rate Control II (RACE II) study we report on here provides strong evidence in favor of lenient rate control.
STUDY SUMMARY: Lenient control is as effective, easier to achieve
RACE II was the first RCT to directly compare lenient rate control (resting heart rate <110 bpm) with strict rate control (resting heart rate <80 bpm, and <110 bpm during moderate exercise). This prospective, multi-center study in Holland randomized patients with permanent atrial fibrillation (N=614) to either a lenient or strict rate-control group. Eligibility criteria were (1) permanent atrial fibrillation for up to 12 months; (2) ≤80 years of age (3) mean resting heart rate >80 bpm; and (4) current use of oral anticoagulation therapy (or aspirin, in the absence of risk factors for thromboembolic complications).
Patients received various doses of beta-blockers, nondihydropyridine calcium-channel blockers, or digoxin, singly or in combination as needed to reach the target heart rate. In both groups, the resting heart rate was determined by 12-lead electrocardiogram after the patient remained in a supine position for 2 to 3 minutes. In the strict-control group, heart rate was also measured during moderate exercise on a stationary bicycle after the resting rate goal had been achieved. In addition, patients in the strict-control group wore a Holter monitor for 24 hours to check for bradycardia.
Participants in both groups were seen every 2 weeks until their heart rate goals were achieved, with follow-up at 1, 2, and 3 years. The primary composite outcome included death from cardiovascular causes; hospitalization for heart failure, stroke, systemic embolism, major bleeding, or life-threatening adverse effects of rate-control drugs; arrhythmic events, including sustained ventricular tachycardia, syncope, or cardiac arrest; and implantation of a pacemaker or cardioverter-defibrillator.
At the end of 3 years, the estimated cumulative incidence of the primary outcome was 12.9% in the lenient-control group vs 14.9% in the strict-control group. The absolute difference was -2.0 (90% confidence interval [CI], -7.6 to 3.5); a 90% CI was acceptable because the study only tested whether lenient control was worse than strict control. The frequency of reported symptoms and adverse events was similar between the 2 groups, but the lenient-control group had fewer visits for rate control (75 vs 684; P<.001), required fewer medications, and took lower doses of some medications.
Heart rate targets were met in 97.7% of patients in the lenient-control group, compared with 67% in the strict-control group (P<.001). Of those not meeting the strict control targets, 25% were due to an adverse medication event. There were no differences between the 2 groups in symptoms or in New York Heart Association functional class status.
WHAT'S NEW: Now we know: It doesn’t pay to go too low
A heart rate <80 at rest and <110 during exercise is difficult to maintain. This more stringent target often requires high dosages of drugs and/or multiple medications, which may lead to adverse effects. This RCT—the first to compare outcomes in patients with lenient vs strict heart rate control—found that morbidity and mortality were similar between the 2 groups. This means that, in many cases, patients will need less medication—leading to a reduction in risk of side effects and interactions.
CAVEATS: Unblinded study excluded very old, high risk
This was not a blinded study, so both patients and providers knew the target heart rates. However, the major outcomes were determined with relative objectivity and were not different between the 2 groups, so it is unlikely that this knowledge would have a major effect on the results. Nonetheless, this is a single study, and the findings are not yet supported by other large, prospective studies.
The researchers did not enroll patients >80 years, who have a higher incidence of atrial fibrillation and are less likely than younger patients to tolerate higher doses of rate-controlling medications. Also excluded were sedentary patients and those with a history of stroke, which resulted in a lower-risk study population. However, 40% of the subjects had a CHADS score ≥2 (an indication of high risk of stroke in patients with atrial fibrillation), and subgroup analysis found that the results applied to higher-risk groups.
Finally, it is possible that it may take longer than 3 years (the duration of study follow-up) for higher ventricular rates to result in adverse cardiovascular outcomes and that there could be a benefit of strict rate control over a longer period of time.
CHALLENGES TO IMPLEMENTATION: Guidelines do not reflect these findings
These findings are not yet incorporated into the ACC/AHA/ESC guidelines or those issued by other organizations. Clinical inertia may stop some physicians from reducing medications for patients with atrial fibrillation, but in general, both doctors and patients should welcome an easing of the drug burden.
Click here to view PURL METHODOLOGY
1. Van Gelder IC, Groenveld HF, Crijns HJ, et al. Lenient versus strict rate control in patients with atrial fibrillation. N Engl J Med. 2010;362:1363-1373.
2. Wyse DG, Waldo AL, DiMarco JP, et al. A comparison of rate control and rhythm control in patients with atrial fibrillation. N Engl J Med. 2002;347:1825-1833.
3. Hagens VE, Ranchor AV, Van SE, et al. Effect of rate or rhythm control on quality of life in persistent atrial fibrillation. Results from the Rate Control Versus Electrical Cardioversion (RACE) Study. J Am Coll Cardiol. 2004;43:241-247.
4. Fuster V, Ryden LE, Cannom DS, et al. ACC/AHA/ESC 2006 Guidelines for the Management of Patients with Atrial Fibrillation: a report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines and the European Society of Cardiology Committee for Practice Guidelines (Writing Committee to Revise the 2001 Guidelines for the Management of Patients With Atrial Fibrillation): developed in collaboration with the European Heart Rhythm Association and the Heart Rhythm Society. Circulation. 2006;114:e257-e354.
5. Dorian P. Rate control in atrial fibrillation. N Engl J Med. 2010;362:1439-1441.
6. Van Gelder IC, Wyse DG, Chandler ML, et al. Does intensity of rate-control influence outcome in atrial fibrillation? An analysis of pooled data from the RACE and AFFIRM studies. Europace. 2006;8:935-942.
 |
Kristen Deane, MD
James J. Stevermer, MD, MSPH
Department of Family and Community Medicine, University of Missouri, Columbia
Kohar Jones, MD
Department of Family Medicine, University of Chicago
PURLs EDITOR
John Hickner, MD, MSc
Department of Family Medicine, Cleveland Clinic
|
Kristen Deane, MD
James J. Stevermer, MD, MSPH
Department of Family and Community Medicine, University of Missouri, Columbia
Kohar Jones, MD
Department of Family Medicine, University of Chicago
PURLs EDITOR
John Hickner, MD, MSc
Department of Family Medicine, Cleveland Clinic
|
Kristen Deane, MD
James J. Stevermer, MD, MSPH
Department of Family and Community Medicine, University of Missouri, Columbia
Kohar Jones, MD
Department of Family Medicine, University of Chicago
PURLs EDITOR
John Hickner, MD, MSc
Department of Family Medicine, Cleveland Clinic
Aim for a heart rate of <110 beats per minute (bpm) in patients with permanent atrial fibrillation. Maintaining this rate requires less medication than more stringent rate control, resulting in fewer side effects and no increased risk of cardiovascular events.1
STRENGTH OF RECOMMENDATION
B: Based on 1 long-term randomized controlled trial (RCT).
Van Gelder IC, Groenveld HF, Crijns HJ, et al. Lenient versus strict rate control in patients with atrial fibrillation. N Engl J Med. 2010;362: 1363-1373.
Illustrative case
A 67-year-old man comes in for a follow-up visit after being hospitalized for atrial fibrillation with a rapid ventricular rate. Before being discharged, he was put on warfarin and metoprolol, and his heart rate today is 96 bpm. You consider increasing the dose of his beta-blocker. What should his target heart rate be?
Atrial fibrillation, the most common sustained arrhythmia,2 can lead to life-threatening events such as heart failure and stroke. Studies, including the Atrial Fibrillation Follow-Up Investigation of Rhythm Management (AFFIRM) and Rate Control versus Electrical Cardioversion (RACE) trials, have found no difference in morbidity or mortality between rate control and rhythm control strategies.2,3 Thus, rate control is usually preferred for patients with atrial fibrillation because of adverse effects associated with antiarrhythmic drugs.
Guidelines cite stringent targets
The American College of Cardiology/American Heart Association Task Force/European Society of Cardiology (ACC/AHA/ESC) guidelines make no definite recommendations about heart rate targets. The guidelines do indicate, however, that rate control criteria vary based on age, “but usually involve achieving ventricular rates between 60 and 80 [bpm] at rest and between 90 and 115 [bpm] during moderate exercise.”4
This guidance is based on data from epidemiologic studies suggesting that faster heart rates in sinus rhythm may increase mortality from cardiovascular causes.5 However, strict control often requires higher doses of rate-controlling medications, which can lead to adverse events such as symptomatic bradycardia, dizziness, and syncope, as well as pacemaker implantation.
Pooled data suggest a more relaxed rate is better
A retrospective analysis of pooled data from the rate-control arms of the AFFIRM and RACE trials found no difference in all-cause mortality between the more stringent rate-control group in AFFIRM and the more lenient control in RACE.6 This finding suggested that more lenient heart rate targets may be preferred to avoid the adverse effects often associated with the higher doses of rate-controlling drugs needed to achieve strict control. The Rate Control Efficacy in Permanent Atrial Fibrillation: a Comparison between Lenient versus Strict Rate Control II (RACE II) study we report on here provides strong evidence in favor of lenient rate control.
STUDY SUMMARY: Lenient control is as effective, easier to achieve
RACE II was the first RCT to directly compare lenient rate control (resting heart rate <110 bpm) with strict rate control (resting heart rate <80 bpm, and <110 bpm during moderate exercise). This prospective, multi-center study in Holland randomized patients with permanent atrial fibrillation (N=614) to either a lenient or strict rate-control group. Eligibility criteria were (1) permanent atrial fibrillation for up to 12 months; (2) ≤80 years of age (3) mean resting heart rate >80 bpm; and (4) current use of oral anticoagulation therapy (or aspirin, in the absence of risk factors for thromboembolic complications).
Patients received various doses of beta-blockers, nondihydropyridine calcium-channel blockers, or digoxin, singly or in combination as needed to reach the target heart rate. In both groups, the resting heart rate was determined by 12-lead electrocardiogram after the patient remained in a supine position for 2 to 3 minutes. In the strict-control group, heart rate was also measured during moderate exercise on a stationary bicycle after the resting rate goal had been achieved. In addition, patients in the strict-control group wore a Holter monitor for 24 hours to check for bradycardia.
Participants in both groups were seen every 2 weeks until their heart rate goals were achieved, with follow-up at 1, 2, and 3 years. The primary composite outcome included death from cardiovascular causes; hospitalization for heart failure, stroke, systemic embolism, major bleeding, or life-threatening adverse effects of rate-control drugs; arrhythmic events, including sustained ventricular tachycardia, syncope, or cardiac arrest; and implantation of a pacemaker or cardioverter-defibrillator.
At the end of 3 years, the estimated cumulative incidence of the primary outcome was 12.9% in the lenient-control group vs 14.9% in the strict-control group. The absolute difference was -2.0 (90% confidence interval [CI], -7.6 to 3.5); a 90% CI was acceptable because the study only tested whether lenient control was worse than strict control. The frequency of reported symptoms and adverse events was similar between the 2 groups, but the lenient-control group had fewer visits for rate control (75 vs 684; P<.001), required fewer medications, and took lower doses of some medications.
Heart rate targets were met in 97.7% of patients in the lenient-control group, compared with 67% in the strict-control group (P<.001). Of those not meeting the strict control targets, 25% were due to an adverse medication event. There were no differences between the 2 groups in symptoms or in New York Heart Association functional class status.
WHAT'S NEW: Now we know: It doesn’t pay to go too low
A heart rate <80 at rest and <110 during exercise is difficult to maintain. This more stringent target often requires high dosages of drugs and/or multiple medications, which may lead to adverse effects. This RCT—the first to compare outcomes in patients with lenient vs strict heart rate control—found that morbidity and mortality were similar between the 2 groups. This means that, in many cases, patients will need less medication—leading to a reduction in risk of side effects and interactions.
CAVEATS: Unblinded study excluded very old, high risk
This was not a blinded study, so both patients and providers knew the target heart rates. However, the major outcomes were determined with relative objectivity and were not different between the 2 groups, so it is unlikely that this knowledge would have a major effect on the results. Nonetheless, this is a single study, and the findings are not yet supported by other large, prospective studies.
The researchers did not enroll patients >80 years, who have a higher incidence of atrial fibrillation and are less likely than younger patients to tolerate higher doses of rate-controlling medications. Also excluded were sedentary patients and those with a history of stroke, which resulted in a lower-risk study population. However, 40% of the subjects had a CHADS score ≥2 (an indication of high risk of stroke in patients with atrial fibrillation), and subgroup analysis found that the results applied to higher-risk groups.
Finally, it is possible that it may take longer than 3 years (the duration of study follow-up) for higher ventricular rates to result in adverse cardiovascular outcomes and that there could be a benefit of strict rate control over a longer period of time.
CHALLENGES TO IMPLEMENTATION: Guidelines do not reflect these findings
These findings are not yet incorporated into the ACC/AHA/ESC guidelines or those issued by other organizations. Clinical inertia may stop some physicians from reducing medications for patients with atrial fibrillation, but in general, both doctors and patients should welcome an easing of the drug burden.
Click here to view PURL METHODOLOGY
Aim for a heart rate of <110 beats per minute (bpm) in patients with permanent atrial fibrillation. Maintaining this rate requires less medication than more stringent rate control, resulting in fewer side effects and no increased risk of cardiovascular events.1
STRENGTH OF RECOMMENDATION
B: Based on 1 long-term randomized controlled trial (RCT).
Van Gelder IC, Groenveld HF, Crijns HJ, et al. Lenient versus strict rate control in patients with atrial fibrillation. N Engl J Med. 2010;362: 1363-1373.
Illustrative case
A 67-year-old man comes in for a follow-up visit after being hospitalized for atrial fibrillation with a rapid ventricular rate. Before being discharged, he was put on warfarin and metoprolol, and his heart rate today is 96 bpm. You consider increasing the dose of his beta-blocker. What should his target heart rate be?
Atrial fibrillation, the most common sustained arrhythmia,2 can lead to life-threatening events such as heart failure and stroke. Studies, including the Atrial Fibrillation Follow-Up Investigation of Rhythm Management (AFFIRM) and Rate Control versus Electrical Cardioversion (RACE) trials, have found no difference in morbidity or mortality between rate control and rhythm control strategies.2,3 Thus, rate control is usually preferred for patients with atrial fibrillation because of adverse effects associated with antiarrhythmic drugs.
Guidelines cite stringent targets
The American College of Cardiology/American Heart Association Task Force/European Society of Cardiology (ACC/AHA/ESC) guidelines make no definite recommendations about heart rate targets. The guidelines do indicate, however, that rate control criteria vary based on age, “but usually involve achieving ventricular rates between 60 and 80 [bpm] at rest and between 90 and 115 [bpm] during moderate exercise.”4
This guidance is based on data from epidemiologic studies suggesting that faster heart rates in sinus rhythm may increase mortality from cardiovascular causes.5 However, strict control often requires higher doses of rate-controlling medications, which can lead to adverse events such as symptomatic bradycardia, dizziness, and syncope, as well as pacemaker implantation.
Pooled data suggest a more relaxed rate is better
A retrospective analysis of pooled data from the rate-control arms of the AFFIRM and RACE trials found no difference in all-cause mortality between the more stringent rate-control group in AFFIRM and the more lenient control in RACE.6 This finding suggested that more lenient heart rate targets may be preferred to avoid the adverse effects often associated with the higher doses of rate-controlling drugs needed to achieve strict control. The Rate Control Efficacy in Permanent Atrial Fibrillation: a Comparison between Lenient versus Strict Rate Control II (RACE II) study we report on here provides strong evidence in favor of lenient rate control.
STUDY SUMMARY: Lenient control is as effective, easier to achieve
RACE II was the first RCT to directly compare lenient rate control (resting heart rate <110 bpm) with strict rate control (resting heart rate <80 bpm, and <110 bpm during moderate exercise). This prospective, multi-center study in Holland randomized patients with permanent atrial fibrillation (N=614) to either a lenient or strict rate-control group. Eligibility criteria were (1) permanent atrial fibrillation for up to 12 months; (2) ≤80 years of age (3) mean resting heart rate >80 bpm; and (4) current use of oral anticoagulation therapy (or aspirin, in the absence of risk factors for thromboembolic complications).
Patients received various doses of beta-blockers, nondihydropyridine calcium-channel blockers, or digoxin, singly or in combination as needed to reach the target heart rate. In both groups, the resting heart rate was determined by 12-lead electrocardiogram after the patient remained in a supine position for 2 to 3 minutes. In the strict-control group, heart rate was also measured during moderate exercise on a stationary bicycle after the resting rate goal had been achieved. In addition, patients in the strict-control group wore a Holter monitor for 24 hours to check for bradycardia.
Participants in both groups were seen every 2 weeks until their heart rate goals were achieved, with follow-up at 1, 2, and 3 years. The primary composite outcome included death from cardiovascular causes; hospitalization for heart failure, stroke, systemic embolism, major bleeding, or life-threatening adverse effects of rate-control drugs; arrhythmic events, including sustained ventricular tachycardia, syncope, or cardiac arrest; and implantation of a pacemaker or cardioverter-defibrillator.
At the end of 3 years, the estimated cumulative incidence of the primary outcome was 12.9% in the lenient-control group vs 14.9% in the strict-control group. The absolute difference was -2.0 (90% confidence interval [CI], -7.6 to 3.5); a 90% CI was acceptable because the study only tested whether lenient control was worse than strict control. The frequency of reported symptoms and adverse events was similar between the 2 groups, but the lenient-control group had fewer visits for rate control (75 vs 684; P<.001), required fewer medications, and took lower doses of some medications.
Heart rate targets were met in 97.7% of patients in the lenient-control group, compared with 67% in the strict-control group (P<.001). Of those not meeting the strict control targets, 25% were due to an adverse medication event. There were no differences between the 2 groups in symptoms or in New York Heart Association functional class status.
WHAT'S NEW: Now we know: It doesn’t pay to go too low
A heart rate <80 at rest and <110 during exercise is difficult to maintain. This more stringent target often requires high dosages of drugs and/or multiple medications, which may lead to adverse effects. This RCT—the first to compare outcomes in patients with lenient vs strict heart rate control—found that morbidity and mortality were similar between the 2 groups. This means that, in many cases, patients will need less medication—leading to a reduction in risk of side effects and interactions.
CAVEATS: Unblinded study excluded very old, high risk
This was not a blinded study, so both patients and providers knew the target heart rates. However, the major outcomes were determined with relative objectivity and were not different between the 2 groups, so it is unlikely that this knowledge would have a major effect on the results. Nonetheless, this is a single study, and the findings are not yet supported by other large, prospective studies.
The researchers did not enroll patients >80 years, who have a higher incidence of atrial fibrillation and are less likely than younger patients to tolerate higher doses of rate-controlling medications. Also excluded were sedentary patients and those with a history of stroke, which resulted in a lower-risk study population. However, 40% of the subjects had a CHADS score ≥2 (an indication of high risk of stroke in patients with atrial fibrillation), and subgroup analysis found that the results applied to higher-risk groups.
Finally, it is possible that it may take longer than 3 years (the duration of study follow-up) for higher ventricular rates to result in adverse cardiovascular outcomes and that there could be a benefit of strict rate control over a longer period of time.
CHALLENGES TO IMPLEMENTATION: Guidelines do not reflect these findings
These findings are not yet incorporated into the ACC/AHA/ESC guidelines or those issued by other organizations. Clinical inertia may stop some physicians from reducing medications for patients with atrial fibrillation, but in general, both doctors and patients should welcome an easing of the drug burden.
Click here to view PURL METHODOLOGY
1. Van Gelder IC, Groenveld HF, Crijns HJ, et al. Lenient versus strict rate control in patients with atrial fibrillation. N Engl J Med. 2010;362:1363-1373.
2. Wyse DG, Waldo AL, DiMarco JP, et al. A comparison of rate control and rhythm control in patients with atrial fibrillation. N Engl J Med. 2002;347:1825-1833.
3. Hagens VE, Ranchor AV, Van SE, et al. Effect of rate or rhythm control on quality of life in persistent atrial fibrillation. Results from the Rate Control Versus Electrical Cardioversion (RACE) Study. J Am Coll Cardiol. 2004;43:241-247.
4. Fuster V, Ryden LE, Cannom DS, et al. ACC/AHA/ESC 2006 Guidelines for the Management of Patients with Atrial Fibrillation: a report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines and the European Society of Cardiology Committee for Practice Guidelines (Writing Committee to Revise the 2001 Guidelines for the Management of Patients With Atrial Fibrillation): developed in collaboration with the European Heart Rhythm Association and the Heart Rhythm Society. Circulation. 2006;114:e257-e354.
5. Dorian P. Rate control in atrial fibrillation. N Engl J Med. 2010;362:1439-1441.
6. Van Gelder IC, Wyse DG, Chandler ML, et al. Does intensity of rate-control influence outcome in atrial fibrillation? An analysis of pooled data from the RACE and AFFIRM studies. Europace. 2006;8:935-942.
1. Van Gelder IC, Groenveld HF, Crijns HJ, et al. Lenient versus strict rate control in patients with atrial fibrillation. N Engl J Med. 2010;362:1363-1373.
2. Wyse DG, Waldo AL, DiMarco JP, et al. A comparison of rate control and rhythm control in patients with atrial fibrillation. N Engl J Med. 2002;347:1825-1833.
3. Hagens VE, Ranchor AV, Van SE, et al. Effect of rate or rhythm control on quality of life in persistent atrial fibrillation. Results from the Rate Control Versus Electrical Cardioversion (RACE) Study. J Am Coll Cardiol. 2004;43:241-247.
4. Fuster V, Ryden LE, Cannom DS, et al. ACC/AHA/ESC 2006 Guidelines for the Management of Patients with Atrial Fibrillation: a report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines and the European Society of Cardiology Committee for Practice Guidelines (Writing Committee to Revise the 2001 Guidelines for the Management of Patients With Atrial Fibrillation): developed in collaboration with the European Heart Rhythm Association and the Heart Rhythm Society. Circulation. 2006;114:e257-e354.
5. Dorian P. Rate control in atrial fibrillation. N Engl J Med. 2010;362:1439-1441.
6. Van Gelder IC, Wyse DG, Chandler ML, et al. Does intensity of rate-control influence outcome in atrial fibrillation? An analysis of pooled data from the RACE and AFFIRM studies. Europace. 2006;8:935-942.
Copyright © 2010 The Family Physicians Inquiries Network. All rights reserved.
When is it safe to forego a CT in kids with head trauma?
Use these newly derived and validated clinical prediction rules to decide which kids need a CT scan after head injury.1
STRENGTH OF RECOMMENDATION
A: Based on consistent, good-quality patient-oriented evidence.
Kuppermann N, Holmes JF, Dayan PS, et al. Identification of children at very low risk of clinically-important brain injuries after head trauma: a prospective cohort study. Lancet. 2009;374:1160-1170.
ILLUSTRATIVE CASE
An anxious mother rushes into your office carrying her 22-month-old son, who fell and hit his head an hour ago. The child has an egg-sized lump on his forehead. Upon questioning his mom about the incident, you learn that the boy fell from a seated position on a chair, which was about 2 feet off the ground. He did not lose consciousness and has no palpable skull fracture—and has been behaving normally ever since. Nonetheless, his mother wants to know if she should take the boy to the emergency department (ED) for a computed tomography (CT) head scan, “just to be safe.” What should you tell her?
Traumatic brain injury (TBI) is a leading cause of childhood morbidity and mortality. In the United States, pediatric head trauma is responsible for 7200 deaths, 60,000 hospitalizations, and more than 600,000 ED visits annually. 2 CT is the diagnostic standard when significant injury from head trauma is suspected, and more than half of all children brought to EDs as a result of head trauma undergo CT scanning. 3
CT is not risk free
CT scans are not benign, however. In addition to the risks associated with sedation, diagnostic radiation is a carcinogen. It is estimated that between 1 in 1000 and 1 in 5000 head CT scans results in a lethal malignancy, and the younger the child, the greater the risk. 4 Thus, when a child incurs a head injury, it is vital to weigh the potential benefit of imaging (discovering a serious, but treatable, injury) and the risk (CT-induced cancer).
Clinical prediction rules for head imaging in children have traditionally been less reliable than those for adults, especially for preverbal children. Guidelines agree that for children with moderate or severe head injury or with a Glasgow Coma Scale (GCS) score ≤13, CT is definitely recommended. 5 The guidelines are less clear regarding the necessity of CT imaging for children with a GCS of 14 or 15.
Eight head trauma clinical prediction rules for kids existed as of December 2008, and they differed considerably in population characteristics, predictors, outcomes, and performance. Only 2 of the 8 prediction rules were derived from high-quality studies, and none were validated in a population separate from their derivation group. 6 A high-quality, high-performing, validated rule was needed to identify children at low risk for serious, treatable head injury—for whom head CT would be unnecessary.
STUDY SUMMARY: Large study yields 2 validated age-based rules
Researchers from the Pediatric Emergency Care Applied Research Network (PECARN) conducted a prospective cohort study to first derive, and then to validate, clinical prediction rules to identify children at very low risk for clinically important traumatic brain injury (ciTBI). They defined ciTBI as death as a result of TBI, need for neurosurgical intervention, intubation of >24 hours, or hospitalization for >2 nights for TBI.
Twenty-five North American EDs enrolled patients younger than 18 years with GCS scores of 14 or 15 who presented within 24 hours of head trauma. Patients were excluded if the mechanism of injury was trivial (ie, ground-level falls or walking or running into stationary objects with no signs or symptoms of head trauma other than scalp abrasions or lacerations). Also excluded were children who had incurred a penetrating trauma, had a known brain tumor or preexisting neurologic disorder that complicated assessment, or had undergone imaging for the head injury at an outside facility. Of 57,030 potential participants, 42,412 patients qualified for the study.
Because the researchers set out to develop 2 pediatric clinical prediction rules—1 for children <2 years of age (preverbal) and 1 for kids ≥2—they divided participants into these age groups. Both groups were further divided into derivation cohorts (8502 preverbal patients and 25,283 patients ≥2 years) and validation cohorts (2216 and 6411 patients, respectively).
Based on their clinical assessment, emergency physicians obtained CT scans for a total of 14,969 children and found ciTBIs in 376—35% and 0.9% of the 42,412 study participants, respectively. Sixty patients required neurosurgery. Investigators ascertained outcomes for the 65% of participants who did not undergo CT imaging via telephone, medical record, and morgue record follow-up; 96 patients returned to a participating health care facility for subsequent care and CT scanning as a result. Of those 96, 5 patients were found to have a TBI. One child had a ciTBI and was hospitalized for 2 nights for a cerebral contusion.
The investigators used established prediction rule methods and Standards for the Reporting of Diagnostic Accuracy Studies (STARD) guidelines to derive the rules. They assigned a relative cost of 500 to 1 for failure to identify a patient with ciTBI vs incorrect classification of a patient who did not have a ciTBI.
Negative finding=0 of 6 predictors
The rules that were derived and validated on the basis of this study are more detailed than previous pediatric prediction rules. For children <2 years, the new standard features 6 factors: altered mental status, palpable skull fracture, loss of consciousness (LOC) for ≥5 seconds, nonfrontal scalp hematoma, severe injury mechanism, and acting abnormally (according to the parents).
The prediction rule for children ≥2 years has 6 criteria, as well, with some key differences. While it, too, includes altered mental status and severe injury mechanism, it also includes clinical signs of basilar skull fracture, any LOC, a history of vomiting, and severe headache. The criteria are further defined, as follows:
Altered mental status: GCS <15, agitation, somnolence, repetitive questions, or slow response to verbal communication.
Severe injury mechanism: Motor vehicle crash with patient ejection, death of another passenger, or vehicle rollover; pedestrian or bicyclist without a helmet struck by a motor vehicle; falls of >3 feet for children <2 years and >5 feet for children ≥2; or head struck by a high-impact object.
Clinical signs of basilar skull fracture: Retroauricular bruising—Battle’s sign (peri-orbital bruising)—raccoon eyes, hemotympanum, or cerebrospinal fluid otorrhea or rhinorrhea.
In both prediction rules, a child is considered negative and, therefore, not in need of a CT scan, only if he or she has none of the 6 clinical predictors of ciTBI.
New rules are highly predictive
In the validation cohorts, the rule for children <2 years had a 100% negative predictive value for ciTBI (95% confidence interval [CI], 99.7-100) and a sensitivity of 100% (95% CI, 86.3-100). The rule for the older children had a negative predictive value of 99.95% (95% CI, 99.81-99.99) and a sensitivity of 96.8% (95% CI, 89-99.6).
In a child who has no clinical predictors, the risk of ciTBI is negligible—and, considering the risk of malignancy from CT scanning, imaging is not recommended. Recommendations for how to proceed if a child has any predictive factors depend on the clinical scenario and age of the patient. In children with a GCS score of 14 or with other signs of altered mental status or palpable skull fracture in those <2 years, or signs of basilar skull fracture in kids ≥2, the risk of ciTBI is slightly greater than 4%. CT is definitely recommended.
In children with a GCS score of 15 and a severe mechanism of injury or any other isolated prediction factor (LOC >5 seconds, non-frontal hematoma, or not acting normally according to a parent in kids <2; any history of LOC, severe headache, or history of vomiting in patients ≥2), the risk of ciTBI is less than 1%. For these children, either CT or observation may be appropriate, as determined by other factors, including clinician experience and patient/parent preference. CT scanning should be given greater consideration in patients who have multiple findings, worsening symptoms, or are <3 months old.
WHAT’S NEW: Rules shed light on hazy areas
These new PECARN rules perform much better than previous pediatric clinical predictors and differ in several ways from the 8 older pediatric head CT imaging rules. The key provisions are the same—if a child has a change in mental status with palpable or visible signs of skull fracture, proceed to imaging. However, this study clarifies which of the other predictors are most important. A severe mechanism of injury is important for all ages. For younger, preverbal children, a nonfrontal hematoma and a parental report of abnormal behavior are important predictors; vomiting or a LOC for <5 seconds is not. For children ≥2 years, vomiting, headache, and any LOC are important; a hematoma is not.
CAVEATS: Clinical decision making is still key
The PECARN rules should guide, rather than dictate, clinical decision making. They use a narrow definition of “clinically important” TBI outcomes—basically death, neurosurgery to prevent death, or prolonged observation to prevent neurosurgery. There are other important, albeit less dire, clinical decisions associated with TBI for which a brain CT may be useful—determining if a high school athlete can safely complete the football season or whether a child should receive anticonvulsant medication to decrease the likelihood of posttraumatic seizures.
We worry, too, that some providers may be tempted to use the rules for after-hours telephone triage. However, clinical assessment of the presence of signs of skull fracture, basilar or otherwise, requires in-person assessment by an experienced clinician.
CHALLENGES TO IMPLEMENTATION: Over- (or under-) reliance on the rules
The PECARN decision rules should simplify head trauma assessment in children. Physicians should first check for altered mental status and signs of skull fracture and immediately send the patient for imaging if either is present. Otherwise, physicians should continue the assessment—looking for the other clinical predictors and ordering a brain CT if 1 or more are found. However, risk of ciTBI is only 1% when only 1 prediction criterion is present. These cases require careful consideration of the potential benefit and risk.
Some emergency physicians may resist using a checklist approach, even one as useful as the PECARN decision guide, and continue to rely solely on their clinical judgment. And some parents are likely to insist on a CT scan for reassurance that there is no TBI, despite the absence of any clinical predictors.
Acknowledgements
The PURLs Surveillance System is supported in part by Grant Number UL1RR024999 from the National Center for Research Resources; the grant is a Clinical Translational Science Award to the University of Chicago. The content is solely the responsibility of the authors and does not necessarily represent the official views of either the National Center for Research Resources or the National Institutes of Health.
The authors wish to thank Sarah-Anne Schumann, MD, Department of Medicine, University of Chicago, for her guidance in the preparation of this manuscript.
PURLs methodology
This study was selected and evaluated using FPIN’s Priority Updates from the Research Literature (PURL) Surveillance System methodology. The criteria and findings leading to the selection of this study as a PURL can be accessed at www.jfponline.com/purls
Click here to view PURL METHODOLOGY
1. Kuppermann N, Holmes JF, Dayan PS, et al. Identification of children at very low risk of clinically-important brain injuries after head trauma: a prospective cohort study. Lancet. 2009;374:1160-1170.
2. National Center for Injury Prevention and Control. Traumatic brain injury in the United States: assessing outcomes in children. CDC; 2006. Available at: http://www.cdc.gov/ncipc/tbi/tbi_report/index.htm . Accessed December 3, 2009.
3. Klassen TP, Reed MH, Stiell IG, et al. Variation in utilization of computed tomography scanning for the investigation of minor head trauma in children: a Canadian experience. Acad Emerg Med. 2000;7:739-744.
4. Brenner DJ. Estimating cancer risks from pediatric CT: going from the qualitative to the quantitative. Pediatr Radiol. 2002;32:228-231.
5. National Guideline Clearing House, ACR Appropriateness Criteria, 2008. Available at: www.guidelines.gov/summary/summary.aspx?doc_id=13670&nbr=007004&string=head+AND+trauma . Accessed December 3, 2009.
6. Maguire JL, Boutis K, Uleryk EM, et al. Should a head-injured child receive a head CT scan? A systematic review of clinical prediction rules. Pediatrics. 2009;124:e145-e154.
Use these newly derived and validated clinical prediction rules to decide which kids need a CT scan after head injury.1
STRENGTH OF RECOMMENDATION
A: Based on consistent, good-quality patient-oriented evidence.
Kuppermann N, Holmes JF, Dayan PS, et al. Identification of children at very low risk of clinically-important brain injuries after head trauma: a prospective cohort study. Lancet. 2009;374:1160-1170.
ILLUSTRATIVE CASE
An anxious mother rushes into your office carrying her 22-month-old son, who fell and hit his head an hour ago. The child has an egg-sized lump on his forehead. Upon questioning his mom about the incident, you learn that the boy fell from a seated position on a chair, which was about 2 feet off the ground. He did not lose consciousness and has no palpable skull fracture—and has been behaving normally ever since. Nonetheless, his mother wants to know if she should take the boy to the emergency department (ED) for a computed tomography (CT) head scan, “just to be safe.” What should you tell her?
Traumatic brain injury (TBI) is a leading cause of childhood morbidity and mortality. In the United States, pediatric head trauma is responsible for 7200 deaths, 60,000 hospitalizations, and more than 600,000 ED visits annually. 2 CT is the diagnostic standard when significant injury from head trauma is suspected, and more than half of all children brought to EDs as a result of head trauma undergo CT scanning. 3
CT is not risk free
CT scans are not benign, however. In addition to the risks associated with sedation, diagnostic radiation is a carcinogen. It is estimated that between 1 in 1000 and 1 in 5000 head CT scans results in a lethal malignancy, and the younger the child, the greater the risk. 4 Thus, when a child incurs a head injury, it is vital to weigh the potential benefit of imaging (discovering a serious, but treatable, injury) and the risk (CT-induced cancer).
Clinical prediction rules for head imaging in children have traditionally been less reliable than those for adults, especially for preverbal children. Guidelines agree that for children with moderate or severe head injury or with a Glasgow Coma Scale (GCS) score ≤13, CT is definitely recommended. 5 The guidelines are less clear regarding the necessity of CT imaging for children with a GCS of 14 or 15.
Eight head trauma clinical prediction rules for kids existed as of December 2008, and they differed considerably in population characteristics, predictors, outcomes, and performance. Only 2 of the 8 prediction rules were derived from high-quality studies, and none were validated in a population separate from their derivation group. 6 A high-quality, high-performing, validated rule was needed to identify children at low risk for serious, treatable head injury—for whom head CT would be unnecessary.
STUDY SUMMARY: Large study yields 2 validated age-based rules
Researchers from the Pediatric Emergency Care Applied Research Network (PECARN) conducted a prospective cohort study to first derive, and then to validate, clinical prediction rules to identify children at very low risk for clinically important traumatic brain injury (ciTBI). They defined ciTBI as death as a result of TBI, need for neurosurgical intervention, intubation of >24 hours, or hospitalization for >2 nights for TBI.
Twenty-five North American EDs enrolled patients younger than 18 years with GCS scores of 14 or 15 who presented within 24 hours of head trauma. Patients were excluded if the mechanism of injury was trivial (ie, ground-level falls or walking or running into stationary objects with no signs or symptoms of head trauma other than scalp abrasions or lacerations). Also excluded were children who had incurred a penetrating trauma, had a known brain tumor or preexisting neurologic disorder that complicated assessment, or had undergone imaging for the head injury at an outside facility. Of 57,030 potential participants, 42,412 patients qualified for the study.
Because the researchers set out to develop 2 pediatric clinical prediction rules—1 for children <2 years of age (preverbal) and 1 for kids ≥2—they divided participants into these age groups. Both groups were further divided into derivation cohorts (8502 preverbal patients and 25,283 patients ≥2 years) and validation cohorts (2216 and 6411 patients, respectively).
Based on their clinical assessment, emergency physicians obtained CT scans for a total of 14,969 children and found ciTBIs in 376—35% and 0.9% of the 42,412 study participants, respectively. Sixty patients required neurosurgery. Investigators ascertained outcomes for the 65% of participants who did not undergo CT imaging via telephone, medical record, and morgue record follow-up; 96 patients returned to a participating health care facility for subsequent care and CT scanning as a result. Of those 96, 5 patients were found to have a TBI. One child had a ciTBI and was hospitalized for 2 nights for a cerebral contusion.
The investigators used established prediction rule methods and Standards for the Reporting of Diagnostic Accuracy Studies (STARD) guidelines to derive the rules. They assigned a relative cost of 500 to 1 for failure to identify a patient with ciTBI vs incorrect classification of a patient who did not have a ciTBI.
Negative finding=0 of 6 predictors
The rules that were derived and validated on the basis of this study are more detailed than previous pediatric prediction rules. For children <2 years, the new standard features 6 factors: altered mental status, palpable skull fracture, loss of consciousness (LOC) for ≥5 seconds, nonfrontal scalp hematoma, severe injury mechanism, and acting abnormally (according to the parents).
The prediction rule for children ≥2 years has 6 criteria, as well, with some key differences. While it, too, includes altered mental status and severe injury mechanism, it also includes clinical signs of basilar skull fracture, any LOC, a history of vomiting, and severe headache. The criteria are further defined, as follows:
Altered mental status: GCS <15, agitation, somnolence, repetitive questions, or slow response to verbal communication.
Severe injury mechanism: Motor vehicle crash with patient ejection, death of another passenger, or vehicle rollover; pedestrian or bicyclist without a helmet struck by a motor vehicle; falls of >3 feet for children <2 years and >5 feet for children ≥2; or head struck by a high-impact object.
Clinical signs of basilar skull fracture: Retroauricular bruising—Battle’s sign (peri-orbital bruising)—raccoon eyes, hemotympanum, or cerebrospinal fluid otorrhea or rhinorrhea.
In both prediction rules, a child is considered negative and, therefore, not in need of a CT scan, only if he or she has none of the 6 clinical predictors of ciTBI.
New rules are highly predictive
In the validation cohorts, the rule for children <2 years had a 100% negative predictive value for ciTBI (95% confidence interval [CI], 99.7-100) and a sensitivity of 100% (95% CI, 86.3-100). The rule for the older children had a negative predictive value of 99.95% (95% CI, 99.81-99.99) and a sensitivity of 96.8% (95% CI, 89-99.6).
In a child who has no clinical predictors, the risk of ciTBI is negligible—and, considering the risk of malignancy from CT scanning, imaging is not recommended. Recommendations for how to proceed if a child has any predictive factors depend on the clinical scenario and age of the patient. In children with a GCS score of 14 or with other signs of altered mental status or palpable skull fracture in those <2 years, or signs of basilar skull fracture in kids ≥2, the risk of ciTBI is slightly greater than 4%. CT is definitely recommended.
In children with a GCS score of 15 and a severe mechanism of injury or any other isolated prediction factor (LOC >5 seconds, non-frontal hematoma, or not acting normally according to a parent in kids <2; any history of LOC, severe headache, or history of vomiting in patients ≥2), the risk of ciTBI is less than 1%. For these children, either CT or observation may be appropriate, as determined by other factors, including clinician experience and patient/parent preference. CT scanning should be given greater consideration in patients who have multiple findings, worsening symptoms, or are <3 months old.
WHAT’S NEW: Rules shed light on hazy areas
These new PECARN rules perform much better than previous pediatric clinical predictors and differ in several ways from the 8 older pediatric head CT imaging rules. The key provisions are the same—if a child has a change in mental status with palpable or visible signs of skull fracture, proceed to imaging. However, this study clarifies which of the other predictors are most important. A severe mechanism of injury is important for all ages. For younger, preverbal children, a nonfrontal hematoma and a parental report of abnormal behavior are important predictors; vomiting or a LOC for <5 seconds is not. For children ≥2 years, vomiting, headache, and any LOC are important; a hematoma is not.
CAVEATS: Clinical decision making is still key
The PECARN rules should guide, rather than dictate, clinical decision making. They use a narrow definition of “clinically important” TBI outcomes—basically death, neurosurgery to prevent death, or prolonged observation to prevent neurosurgery. There are other important, albeit less dire, clinical decisions associated with TBI for which a brain CT may be useful—determining if a high school athlete can safely complete the football season or whether a child should receive anticonvulsant medication to decrease the likelihood of posttraumatic seizures.
We worry, too, that some providers may be tempted to use the rules for after-hours telephone triage. However, clinical assessment of the presence of signs of skull fracture, basilar or otherwise, requires in-person assessment by an experienced clinician.
CHALLENGES TO IMPLEMENTATION: Over- (or under-) reliance on the rules
The PECARN decision rules should simplify head trauma assessment in children. Physicians should first check for altered mental status and signs of skull fracture and immediately send the patient for imaging if either is present. Otherwise, physicians should continue the assessment—looking for the other clinical predictors and ordering a brain CT if 1 or more are found. However, risk of ciTBI is only 1% when only 1 prediction criterion is present. These cases require careful consideration of the potential benefit and risk.
Some emergency physicians may resist using a checklist approach, even one as useful as the PECARN decision guide, and continue to rely solely on their clinical judgment. And some parents are likely to insist on a CT scan for reassurance that there is no TBI, despite the absence of any clinical predictors.
Acknowledgements
The PURLs Surveillance System is supported in part by Grant Number UL1RR024999 from the National Center for Research Resources; the grant is a Clinical Translational Science Award to the University of Chicago. The content is solely the responsibility of the authors and does not necessarily represent the official views of either the National Center for Research Resources or the National Institutes of Health.
The authors wish to thank Sarah-Anne Schumann, MD, Department of Medicine, University of Chicago, for her guidance in the preparation of this manuscript.
PURLs methodology
This study was selected and evaluated using FPIN’s Priority Updates from the Research Literature (PURL) Surveillance System methodology. The criteria and findings leading to the selection of this study as a PURL can be accessed at www.jfponline.com/purls
Click here to view PURL METHODOLOGY
Use these newly derived and validated clinical prediction rules to decide which kids need a CT scan after head injury.1
STRENGTH OF RECOMMENDATION
A: Based on consistent, good-quality patient-oriented evidence.
Kuppermann N, Holmes JF, Dayan PS, et al. Identification of children at very low risk of clinically-important brain injuries after head trauma: a prospective cohort study. Lancet. 2009;374:1160-1170.
ILLUSTRATIVE CASE
An anxious mother rushes into your office carrying her 22-month-old son, who fell and hit his head an hour ago. The child has an egg-sized lump on his forehead. Upon questioning his mom about the incident, you learn that the boy fell from a seated position on a chair, which was about 2 feet off the ground. He did not lose consciousness and has no palpable skull fracture—and has been behaving normally ever since. Nonetheless, his mother wants to know if she should take the boy to the emergency department (ED) for a computed tomography (CT) head scan, “just to be safe.” What should you tell her?
Traumatic brain injury (TBI) is a leading cause of childhood morbidity and mortality. In the United States, pediatric head trauma is responsible for 7200 deaths, 60,000 hospitalizations, and more than 600,000 ED visits annually. 2 CT is the diagnostic standard when significant injury from head trauma is suspected, and more than half of all children brought to EDs as a result of head trauma undergo CT scanning. 3
CT is not risk free
CT scans are not benign, however. In addition to the risks associated with sedation, diagnostic radiation is a carcinogen. It is estimated that between 1 in 1000 and 1 in 5000 head CT scans results in a lethal malignancy, and the younger the child, the greater the risk. 4 Thus, when a child incurs a head injury, it is vital to weigh the potential benefit of imaging (discovering a serious, but treatable, injury) and the risk (CT-induced cancer).
Clinical prediction rules for head imaging in children have traditionally been less reliable than those for adults, especially for preverbal children. Guidelines agree that for children with moderate or severe head injury or with a Glasgow Coma Scale (GCS) score ≤13, CT is definitely recommended. 5 The guidelines are less clear regarding the necessity of CT imaging for children with a GCS of 14 or 15.
Eight head trauma clinical prediction rules for kids existed as of December 2008, and they differed considerably in population characteristics, predictors, outcomes, and performance. Only 2 of the 8 prediction rules were derived from high-quality studies, and none were validated in a population separate from their derivation group. 6 A high-quality, high-performing, validated rule was needed to identify children at low risk for serious, treatable head injury—for whom head CT would be unnecessary.
STUDY SUMMARY: Large study yields 2 validated age-based rules
Researchers from the Pediatric Emergency Care Applied Research Network (PECARN) conducted a prospective cohort study to first derive, and then to validate, clinical prediction rules to identify children at very low risk for clinically important traumatic brain injury (ciTBI). They defined ciTBI as death as a result of TBI, need for neurosurgical intervention, intubation of >24 hours, or hospitalization for >2 nights for TBI.
Twenty-five North American EDs enrolled patients younger than 18 years with GCS scores of 14 or 15 who presented within 24 hours of head trauma. Patients were excluded if the mechanism of injury was trivial (ie, ground-level falls or walking or running into stationary objects with no signs or symptoms of head trauma other than scalp abrasions or lacerations). Also excluded were children who had incurred a penetrating trauma, had a known brain tumor or preexisting neurologic disorder that complicated assessment, or had undergone imaging for the head injury at an outside facility. Of 57,030 potential participants, 42,412 patients qualified for the study.
Because the researchers set out to develop 2 pediatric clinical prediction rules—1 for children <2 years of age (preverbal) and 1 for kids ≥2—they divided participants into these age groups. Both groups were further divided into derivation cohorts (8502 preverbal patients and 25,283 patients ≥2 years) and validation cohorts (2216 and 6411 patients, respectively).
Based on their clinical assessment, emergency physicians obtained CT scans for a total of 14,969 children and found ciTBIs in 376—35% and 0.9% of the 42,412 study participants, respectively. Sixty patients required neurosurgery. Investigators ascertained outcomes for the 65% of participants who did not undergo CT imaging via telephone, medical record, and morgue record follow-up; 96 patients returned to a participating health care facility for subsequent care and CT scanning as a result. Of those 96, 5 patients were found to have a TBI. One child had a ciTBI and was hospitalized for 2 nights for a cerebral contusion.
The investigators used established prediction rule methods and Standards for the Reporting of Diagnostic Accuracy Studies (STARD) guidelines to derive the rules. They assigned a relative cost of 500 to 1 for failure to identify a patient with ciTBI vs incorrect classification of a patient who did not have a ciTBI.
Negative finding=0 of 6 predictors
The rules that were derived and validated on the basis of this study are more detailed than previous pediatric prediction rules. For children <2 years, the new standard features 6 factors: altered mental status, palpable skull fracture, loss of consciousness (LOC) for ≥5 seconds, nonfrontal scalp hematoma, severe injury mechanism, and acting abnormally (according to the parents).
The prediction rule for children ≥2 years has 6 criteria, as well, with some key differences. While it, too, includes altered mental status and severe injury mechanism, it also includes clinical signs of basilar skull fracture, any LOC, a history of vomiting, and severe headache. The criteria are further defined, as follows:
Altered mental status: GCS <15, agitation, somnolence, repetitive questions, or slow response to verbal communication.
Severe injury mechanism: Motor vehicle crash with patient ejection, death of another passenger, or vehicle rollover; pedestrian or bicyclist without a helmet struck by a motor vehicle; falls of >3 feet for children <2 years and >5 feet for children ≥2; or head struck by a high-impact object.
Clinical signs of basilar skull fracture: Retroauricular bruising—Battle’s sign (peri-orbital bruising)—raccoon eyes, hemotympanum, or cerebrospinal fluid otorrhea or rhinorrhea.
In both prediction rules, a child is considered negative and, therefore, not in need of a CT scan, only if he or she has none of the 6 clinical predictors of ciTBI.
New rules are highly predictive
In the validation cohorts, the rule for children <2 years had a 100% negative predictive value for ciTBI (95% confidence interval [CI], 99.7-100) and a sensitivity of 100% (95% CI, 86.3-100). The rule for the older children had a negative predictive value of 99.95% (95% CI, 99.81-99.99) and a sensitivity of 96.8% (95% CI, 89-99.6).
In a child who has no clinical predictors, the risk of ciTBI is negligible—and, considering the risk of malignancy from CT scanning, imaging is not recommended. Recommendations for how to proceed if a child has any predictive factors depend on the clinical scenario and age of the patient. In children with a GCS score of 14 or with other signs of altered mental status or palpable skull fracture in those <2 years, or signs of basilar skull fracture in kids ≥2, the risk of ciTBI is slightly greater than 4%. CT is definitely recommended.
In children with a GCS score of 15 and a severe mechanism of injury or any other isolated prediction factor (LOC >5 seconds, non-frontal hematoma, or not acting normally according to a parent in kids <2; any history of LOC, severe headache, or history of vomiting in patients ≥2), the risk of ciTBI is less than 1%. For these children, either CT or observation may be appropriate, as determined by other factors, including clinician experience and patient/parent preference. CT scanning should be given greater consideration in patients who have multiple findings, worsening symptoms, or are <3 months old.
WHAT’S NEW: Rules shed light on hazy areas
These new PECARN rules perform much better than previous pediatric clinical predictors and differ in several ways from the 8 older pediatric head CT imaging rules. The key provisions are the same—if a child has a change in mental status with palpable or visible signs of skull fracture, proceed to imaging. However, this study clarifies which of the other predictors are most important. A severe mechanism of injury is important for all ages. For younger, preverbal children, a nonfrontal hematoma and a parental report of abnormal behavior are important predictors; vomiting or a LOC for <5 seconds is not. For children ≥2 years, vomiting, headache, and any LOC are important; a hematoma is not.
CAVEATS: Clinical decision making is still key
The PECARN rules should guide, rather than dictate, clinical decision making. They use a narrow definition of “clinically important” TBI outcomes—basically death, neurosurgery to prevent death, or prolonged observation to prevent neurosurgery. There are other important, albeit less dire, clinical decisions associated with TBI for which a brain CT may be useful—determining if a high school athlete can safely complete the football season or whether a child should receive anticonvulsant medication to decrease the likelihood of posttraumatic seizures.
We worry, too, that some providers may be tempted to use the rules for after-hours telephone triage. However, clinical assessment of the presence of signs of skull fracture, basilar or otherwise, requires in-person assessment by an experienced clinician.
CHALLENGES TO IMPLEMENTATION: Over- (or under-) reliance on the rules
The PECARN decision rules should simplify head trauma assessment in children. Physicians should first check for altered mental status and signs of skull fracture and immediately send the patient for imaging if either is present. Otherwise, physicians should continue the assessment—looking for the other clinical predictors and ordering a brain CT if 1 or more are found. However, risk of ciTBI is only 1% when only 1 prediction criterion is present. These cases require careful consideration of the potential benefit and risk.
Some emergency physicians may resist using a checklist approach, even one as useful as the PECARN decision guide, and continue to rely solely on their clinical judgment. And some parents are likely to insist on a CT scan for reassurance that there is no TBI, despite the absence of any clinical predictors.
Acknowledgements
The PURLs Surveillance System is supported in part by Grant Number UL1RR024999 from the National Center for Research Resources; the grant is a Clinical Translational Science Award to the University of Chicago. The content is solely the responsibility of the authors and does not necessarily represent the official views of either the National Center for Research Resources or the National Institutes of Health.
The authors wish to thank Sarah-Anne Schumann, MD, Department of Medicine, University of Chicago, for her guidance in the preparation of this manuscript.
PURLs methodology
This study was selected and evaluated using FPIN’s Priority Updates from the Research Literature (PURL) Surveillance System methodology. The criteria and findings leading to the selection of this study as a PURL can be accessed at www.jfponline.com/purls
Click here to view PURL METHODOLOGY
1. Kuppermann N, Holmes JF, Dayan PS, et al. Identification of children at very low risk of clinically-important brain injuries after head trauma: a prospective cohort study. Lancet. 2009;374:1160-1170.
2. National Center for Injury Prevention and Control. Traumatic brain injury in the United States: assessing outcomes in children. CDC; 2006. Available at: http://www.cdc.gov/ncipc/tbi/tbi_report/index.htm . Accessed December 3, 2009.
3. Klassen TP, Reed MH, Stiell IG, et al. Variation in utilization of computed tomography scanning for the investigation of minor head trauma in children: a Canadian experience. Acad Emerg Med. 2000;7:739-744.
4. Brenner DJ. Estimating cancer risks from pediatric CT: going from the qualitative to the quantitative. Pediatr Radiol. 2002;32:228-231.
5. National Guideline Clearing House, ACR Appropriateness Criteria, 2008. Available at: www.guidelines.gov/summary/summary.aspx?doc_id=13670&nbr=007004&string=head+AND+trauma . Accessed December 3, 2009.
6. Maguire JL, Boutis K, Uleryk EM, et al. Should a head-injured child receive a head CT scan? A systematic review of clinical prediction rules. Pediatrics. 2009;124:e145-e154.
1. Kuppermann N, Holmes JF, Dayan PS, et al. Identification of children at very low risk of clinically-important brain injuries after head trauma: a prospective cohort study. Lancet. 2009;374:1160-1170.
2. National Center for Injury Prevention and Control. Traumatic brain injury in the United States: assessing outcomes in children. CDC; 2006. Available at: http://www.cdc.gov/ncipc/tbi/tbi_report/index.htm . Accessed December 3, 2009.
3. Klassen TP, Reed MH, Stiell IG, et al. Variation in utilization of computed tomography scanning for the investigation of minor head trauma in children: a Canadian experience. Acad Emerg Med. 2000;7:739-744.
4. Brenner DJ. Estimating cancer risks from pediatric CT: going from the qualitative to the quantitative. Pediatr Radiol. 2002;32:228-231.
5. National Guideline Clearing House, ACR Appropriateness Criteria, 2008. Available at: www.guidelines.gov/summary/summary.aspx?doc_id=13670&nbr=007004&string=head+AND+trauma . Accessed December 3, 2009.
6. Maguire JL, Boutis K, Uleryk EM, et al. Should a head-injured child receive a head CT scan? A systematic review of clinical prediction rules. Pediatrics. 2009;124:e145-e154.
Copyright © 2010 The Family Physicians Inquiries Network.
All rights reserved.