Heterogeneity, evidence and salt

Commentary

CJEM 2004;6(3):165-166

See also page 147.

In this edition of the Canadian Journal of Emergency Medicine, Vaillancourt and colleagues present an important article on acute compartment syndrome.¹ While reading this article I was struck by the heterogeneity of the subjects. I realized that I'd been naively thinking of "compartment syndrome" as a homogeneous entity. But these authors demonstrate that penetrating trauma, blunt trauma, and even non-traumatic problems funnel into a final common pathway that leads to acute compartment syndrome and fasciotomy. This prompted me to consider other heterogeneous conditions that are inappropriately viewed as homogeneous and to ponder the role that heterogeneity plays in the "evidence" used to make clinical decisions.

Pediatric head trauma

Pediatric head trauma is heterogeneous in at least 2 fundamental ways. First, researchers frequently group "children" of all ages. In a recent observational study of traumatic brain injury,² Palchak and colleagues enrolled over 2000 "children," right up to18 years of age. It seems intuitive to me that a 17-day-old, a 17-month-old, and a 17-year-old are different physiological creatures. A 17-day-old feeds and cries, a 17-month-old runs and plays, and a 17-year-old can drive a car, work for a living and conceive children. The authors performed recursive partitioning on their data and produced a list of low-risk criteria for traumatic brain injury. But, given the underlying heterogeneity of their subjects, I am uneasy using their decision rule to help me make decisions in the emergency department.

Second, the process of developing "evidence-based" guidelines often involves more than just "evidence." In 2001, Schutzman and colleagues suggested guidelines for the evaluation and management of children under 2 years of age with apparently minor head trauma.³ In this proposed guideline the authors acknowledged that "available data do not demonstrate that either seizure, vomiting, or loss of consciousness are independent predictors for intracranial injury;" yet they proceeded to suggest that children with a seizure or loss of consciousness for longer than 1 minute should have computerized tomographic (CT) scanning of the head. With regard to vomiting, they arbitrarily suggested a threshold of 5 episodes or 6 hours of vomiting as an indication for a head CT. Why go to the trouble of reviewing many studies, which identify features not associated with intracranial injuries, and then recommend those exact features as indications for CT scanning? I believe the authors understand, on some level, that epidural hematomas, subdural hematomas, subarachnoid hemorrhages, depressed skull fractures, and other clinically important intracranial injuries are a heterogeneous group for which we should expect different clinical presentations. I remain open, however, to the possibility that this particular guideline might actually be eminence-based, vehemence-based, eloquence-based, providence-based, or nervousness-based.⁴

Bronchiolitis

"Bronchiolitis" is not a disease; it is a convenient term for wheezing infants presumed to have viral lung infections. And it is heterogeneous. Studies^5-7 have demonstrated highly variable rates of respiratory syncytial virus infection in children with "bronchiolitis." Over the last decade there have been at least 9 randomized controlled trials evaluating the use of nebulized epinephrine to treat infants with bronchiolitis. Four of these suggested that epinephrine was efficacious,^8-11 4 suggested it was not efficacious,^12-15 and 1 suggested that, although it might not be entirely efficacious, it might expedite discharge from the ED.¹⁶ Not surprisingly, a subsequent meta-analysis concluded that epinephrine may have some minimal short-term benefits, but doesn't help inpatients.¹⁷ The reality may be that bronchiolitis is a heterogeneous condition, that epinephrine works for some patients and doesn't work for others, and that when you study a heterogeneous clinical condition, the conclusions become muddled.

Conclusion

In considering these examples, heterogeneity is probably pervasive and under-appreciated in clinical research. Unrecognized heterogeneity may have a profound impact on the conclusions of studies we use to guide clinical decision-making. "Evidence-based" medicine is wonderful fare, but when served with heterogeneity, readers should consider adding a "grain of salt."

References

1. Vaillancourt C, Shrier I, Vandal A, Falk M, Rossignol M, Vernec A, et al. Acute compartment syndrome: How long before muscle necrosis occurs? Can J Emerg Med 2004;6(3):147-54.
2. Palchak MJ, Holmes JF, Vance CW, Gelber RE, Schauer BA, Harrison MJ, et al. A decision rule for identifying children at low risk for brain injuries after blunt head trauma. Ann Emerg Med 2003;42:492-506.
3. Schutzman SA, Barnes P, Duhaime AC, Greenes D, Homer C, Jaffe D, et al. Evaluation and management of children younger than two years old with apparently minor head trauma: Proposed guidelines. Pediatrics 2001;107:983-93.
4. Isaacs D, Fitzgerald D. Seven alternatives to evidence based medicine. BMJ 1999;319:1618.
5. Brown L, Reiley DG, Jeng A, Green SM. Bronchiolitis: Can objective criteria predict eligibility for brief hospitalization? Can J Emerg Med 2003;5(4):239-44.
6. Shaw KN, Bell LM, Sherman NH. Outpatient assessment of infants with bronchiolitis. Am J Dis Child 1991;145:151-5.
7. Mulholland EK, Olinsky A, Shann FA. Clinical findings and severity of acute bronchiolitis. Lancet 1990;335:1259-61.
8. Kristjansson S, Lodrup Carlsen KC, Wennergren G, Strannegard IL, Carlsen KH. Nebulised racemic adrenaline in the treatment of acute bronchiolitis in infants and toddlers. Arch Dis Child 1993;69: 650-4.
9. Reijonen T, Korppi M, Pitkakangas S, Tenhola S, Remes K. The clinical efficacy of nebulized racemic epinephrine and albuterol in acute bronchiolitis. Arch Pediatr Adolesc Med 1995;149:686-92.
10. Menon K, Sutcliffe T, Klassen TP. A randomized trial comparing the efficacy of epinephrine with salbutamol in the treatment of acute bronchiolitis. J Pediatr 1995;126:1004-7.
11. Bertrand P, Aranibar H, Castro E, Sanchez I. Efficacy of nebulized epinephrine versus salbutamol in hospitalized infants with bronchiolitis. Pediatr Pulmonol 2001;31:284-8.
12. Abul-Ainine A, Luyt D. Short term effects of adrenaline in bronchiolitis: a randomised controlled trial. Arch Dis Child 2002;86:276-9.
13. Patel H, Platt RW, Pekeles GS, Ducharme FM. A randomized, controlled trial of the effectiveness of nebulized therapy with epinephrine compared with albuterol and saline in infants hospitalized for acute viral bronchiolitis. J Pediatr 2002;141:818-24.
14. Hariprakash S, Alexander J, Carroll W, Ramesh P, Randell T, Turnbull F, et al. Randomized controlled trial of nebulized adrenaline in acute bronchiolitis. Pedaitr Allergy Immunol 2003;14:134-9.
15. Wainwright C, Altamirano L, Cheney M, Cheney J, Barber S, Price D, et al. A multicenter, randomized, double-blind, controlled trial of nebulized epinephrine in infants with acute bronchiolitis. N Engl J Med 2003;349:27-35.
16. Mull CC, Scarfone RJ, Ferri LR, Carlin T, Salvaggio C, Bechtel KA, et al. A randomized trial of nebulized epinephrine vs. albuterol in the emergency department treatment of bronchiolitis. Arch Pediatr Adolesc Med 2004;158:113-8.
17. Hartling L, Wiebe N, Russell K, Patel H, Klassen TP. A meta-analysis of randomized controlled trials evaluating the efficacy of epinephrine for the treatment of acute viral bronchiolitis. Arch Pediatr Adolesc Med 2003;157:957-64.