Although Schrenck et al state that concurrent diltiazem and digoxin therapy has not been noted to cause increased digoxin levels, the authors of several studies have noted significant digoxin increase.8 Admittedly, reported findings of digoxin and verapamil or diltiazem interactio:qs are inconsistent. It is important to note that intersubject and intrasubject variability for adverse effects is high and findings that report only mean value changes in digoxin levels can fail to identify significant adverse effects among individual patients. Perhaps a more worthy area of discussion by Schrenck et al would have been the finding by Ellenbogen5 in which the on[y variables predictive of symptomatic hypotension during acute rate control following diltiazem therapy were diastolic pressure of less than 75 mm Hg and digoxin administration. When the authors of a study sponsored by a pharmaceutical company attack a competitor's medication, great care should be taken to render a comprehensive and balanced analysis. Misrepresentation of the true science of the BOhm study is found in several such studies.2-5 Also, a study by Chew et al 9 is cited in many studies to demonstrate the occurrence of marked pulmonary wedge pressure increase in some patients with impaired cardiac function following verapamil infusion; however, the authors of these studies often fail to cite important studies that demonstrate marked pulmonary wedge pressure increase, with a similar incidence, following the infusion of diltiazem 1° and studies that demonstrate favorable findings following use of a more appropriate dose of verapamil) 1 These are important and seemingly biased omissions. I do not know whether verapamil or diltiazem is the more efficacious medication. However, ! do know that at my hospital, the cost of diltiazero is almost four times that of verapamil. I can best conclude by quoting one of the more balanced statements I have read among the diltiazem studies sponsored by pharmaceutical companies. In the discussion section of a study published by the Diltiazem-Atrial Fibrillation/ Flutter Study Group,12 it is stated, "It is possible that intravenous diltiazem may be a better agent than intravenous verapamil for treatment of atrial fibrillation and flutter. However, support for this hypothesis',requires a comparative trial."
R Ben Zemenick, DO Departmentof EmergencyMedicine St Paul Medical Center Dallas, TX i. Bohm M, 5chwinger RHG, Erdmann E: Different eardiodepressant potency of various calcium antagonists in human myocardium. Am J Cardiol 1990;65:1039- i04i. 2. EIknbogen KA, Dins VC, Plumb VJ, et al: A placebocontrolled trial of continuous intravenous diltiazem infusion for 24-hour heart rate control during atdaI fibrilIation and atrial flutter: A muIticenter study. J Am Coil Cardiol 199i; 18:1046-1051. 3. Heywood yT, Graham B, Marais GE, et aI: Effects of intravenous diltiazem on rapid atrial fibrfllation accompanied by congestive heart failure. AmJ Cardio1199I;67:1150-1 I52. 4. Goldenberg IF, Lewis WR, Dins VC, et al: Intravenous diItiazem for the treatment of patients with atrial flbrillation offlutter and moderate to severe congestive heart failure. Am J Cardio11994;74:884-889. 5. Ellenbogen KA, Dins VC, Cardello FP, et al: Safety and efficacy of intravenous diltiazem in atriaZfibrillation or atrial flutter. AmJ Cardiol 1995;75:45-49. 6. Godfraind 1", 5alomone 5, Dessy C, et al: Selectivity scale of calcium antagonist in the human cardiovascular system based on in vitro studies. J Cardiovasc Pharmacol 1992;20: 534-542. 7. Schwinger RHG, BOhm M, Erdmann E: Negative inotropic properties of isradipine, nifedipine, diltiazem, and verapamil in diseased human myocardial tissue. J Cardiovasc Pharmacol 1990;I5:892-899. 8. Lessem JN: Interaction between Cae+ antagonists and digtalis. Cardiovasc Drug Tiler i988;1:441-446. 9. Chew CYC, Heeht HS, Collett dT, et ah Influence of severity of ventricutar dysfunction on hemodynamic responses to intravenously administered verapamil in ischemic heart disease. AmJ Cardio11981;47:917 922 10. Ferlinz j Litron PD: Hemodynamic and myocardial performance characteristics @er verapamit use in congestive heart failure. AmJ Cardiol 1983;61:1339-1345. 1 i. Materne P, Legrand V, Vandormalel M, et al: Hemodynamic effects of intravenous diltiazem with impaired left ventricular function. Am J Cardiol 1984;54: 733- 73 7. 12. Salerno DM, Dins VC, KIeiger RE, et aI: Efficacy and safety of intravenous diltiazem for treatment of atrial flbriIlation and atrial flutter. Am J Cardio11989;63:2046-I051.
In reply, We thank Dr Zemenick for his comments; however, we stand firmly by our conclusions. We believe it is important to emphasize that our study was a comparison of intravenous diNazem and intravenous digoxin in the emergency treatment of acute atrial fibrillation and flutter (AFF).We have never conducted a formal investigation of intravenous verapamil in this setting. However, because intravenous verapamil has been used by many emergency physicians in the emergency treatment of AFF, we believed it was important to briefly address "reported" comparisons of diltiazem and verapamil, We agree this is a controversial subject, as Dr Zemenick indicated in his own article, 1 but our study was clearly designed to test a different hypothesis. Although Dr Zemenick has de-
ANNALS OF EMERGENCY MEDICINE
scribed his interpretation of the results in the study performed by BOhm,2 we do not believe it was our obligation, or the intent of our study, to validate (or invalidate)the conclusions drawn by the authors of the BOhm study. Regardless of the viewpoint expressed by Dr Zemenick that our sponsored study attacks a competitor's medication, we only wished to reflect common teaching in comparing verapamil and diltiazem. We clarified our brief mention of this comparison by stating that it was only what had been commonly reported. Many textbooks refer to the negative inotropic effects of verapamil as stronger than those of diltiazem.3-7 We firmly believe our statements are fair and accurate, given their documentation in standard medical teaching, and we never intended them to be viewed as an attack. We did, however, want to provoke thought and controversy to further the investigations by emergency physicians in the emergency management of AFF. Again, we thank Dr Zemenick. We also thank our study "sponsors" for their unrestrictive and noncollaborative support of our work.
David M Schreck,MD, MS Alberto R River& MD Victor Tricarico, MD i. Zemenick RB: Verapamfl or diltiazem for acute rate control: Which is best? Am J Cardio[ 1995; 76:538-539. 2. BOhm M, 5chwinger RHG, Erdmann E: Different cardiodepressant potency of various calcium antagonists in human myocardium. Am J Cardiol 1990;65:10394 04i. 3. Epstein M, Oster JR: Role of caldum channel blockers in treatment of hypertension, in Epstein M, Oster JR (eds): Hypertension: Practical Management. Miami, FL: Battersea Medical Publications, t 988:115-116; 4. McMahon FG: Calcium channel blockers, in MeMahon FG (ed): Management of Essential Hypertension. Mount Kisco, N~" Futura Medical Publishing, i984:394. 5. Oparil 5: Arterial hypertension, in Wyngaarden JB, Smith LH, BennettJC (e&): Cecil Textbook of Medicine. Philadelphia, PA: 5aunders, 1992:264~265. 6. Rutherjord JD, Braunwald E: Chronic ischemic heart disease, in Braunwald E (ed): Heart Disease. Philadelphia, PA: Saunders, 1992:1314. 7. Murad F: Drugs used for the treatment of angina, in Gilman AG, RalI TW, et al (eds): The Pharmacologic Basis of Therapeutics. New York: Pergamon Press, 1990:775.
Corticosteroids and Asthma in Children To the Editor. Barnett et al are to be commended for their comparison of intravenous and oral corticosteroids in the management of acute asthma in children [February 1997;29:212-217]. They
concluded that for children with moderate to severe asthma, admission rates were the same regardless of whether oral or intravenous steroids were used. They also found that 4 hours after treatment, respiratory rate, oxygen saturation, Pulmonary Index Score, and 1-second forced expiratory volume were similar in children treated with oral steroids and those given intravenous steroids. One issue these authors did not address is whether administration of corticosteroids by the intravenous route results in more rapid clinical improvement than oral administration. Clearly, a more rapid clinical response would be advantageous to patients and ED staff alike because it would permit earlier discharge and more efficient use of ED personnel and resources. Do the authors have any data on the other time points in the study (1,2, and 3 hours after treatment) that might aid the determination of whether response rate differs between the two treatments?
Adam J Singer, MD Departmentof EmergencyMedicine SUNYat StonyBrook Stony Brook, NY In reply. We thank Dr Singer for his comments. We chose 4 hours as the time point at which to decide each patient's disposition because it seems to be the accepted length of stay in an ED without an observation unit. This point also provided information about the onset of action of methylprednisolone by both routes of administration. Evaluation of all the variables (eg, respiratory rate, oxygen saturation, Pulmonary Index Score, and 1-second forced expiratory volume) at 1,2, and 3 hours showed no differences. Thus it would appear that intravenous steroids did not have an effect sooner than did oral steroids in this period. We found significant differences in the pulmonary variables at 1 hour between admitted and discharged patients. Thus decisions regarding disposition may have been made earlier for some patients in each group.
PeterBarnett, MBBS Nate Kuppermann,MD Mark Baskin, MD Grace Caput& MD Departmentof EmergencyMedicine Children's Hospital Boston, MA
Spirometric Testing in Acute Asthma To the Editor. In their review of refractory asthma, Jagoda et al [February 1997;29:275-281] suggest that spirometry enables the physician to more accurately predict which patients will require hospital admission. The authors cite seven studies to support the statement, "Spirometric values on presentation, measurement after initial 13agonist treatment, and values after 1 to 6 hours of ED treatment each predicted 60% to 85% of those patients who were admitted or relapsed." This wording suggests that spirometric values are accurately predictive of outcome. However, this is not the case. Blinded prospective studies have not demonstrated spirometric values-at presentation, changes from presentation, or values at disposition--as accurately predictive of outcome. 1-3 Misclassification rates of any prediction rule should be measured in prospective studies in new clinical settings, and the use of data from nonblinded studies is an important deficiency because it introduces bias? In the only blinded study cited by the authors, Martin et al3 demonstrated that patients were unlikely to require hospitalization if they demonstrated 20% improvement in peak expiratory flow rate (PEFR) after initial therapy; however, more than 60% of the discharged patients did not show this level of improvement. Furthermore, change in PEFR, after initial therapy, did not accurately identify patients who relapsed after discharge. The other studies cited by Jagoda et al were not blinded, introducing bias during patient disposition. However, in the studies used to provide data for analysis, use of the retrospectively derived spirometry values associated with admission or relapse rates would result in high misclassification rates. As an example (and among the references cited by Jagoda et al), use of the 1-second forced expiratory volume (FEV1) values recommended for discharge by Nowak 5 would have resulted in admission of about one third of the successfully discharged patients. With the use of mandatory hospital admission based on values associated with high relapse rates by Lulla and Newcomer, 6 more than half of the patients discharged without relapse would have been admitted unnecessarily. As noted by Jagoda et al, cardiorespiratory arrest has been associated with the use of
forced expiratory maneuvers. These cases were associated with exacerbation of cough after spirometric testing. If spirometry-induced cough exacerbation is a risk for sudden cardiopulmonary arrest, as many as 20% of acutely asthmatic patients presenting to the ED may be at risk.7 Although Jagoda et al state, "In cases that are clinically clearly life-threatening, spirometry should be deferred until after treatment and improvement," cardiorespiratory arrest following PEFRtesting was noted in a patient during his second hospital day of asthma therapy; it followed a PEFRof 350 L/minute. 8 Most emergency physicians do not follow published guidelines recommending routine spirometric testing.9 This practice is consistent with the recommendation of the authors of blinded prospective studies, which have failed to demonstrate spirometry to be an aid in decisionmaking with regard to discharge from the ED.1-2 Furthermore, once a decision for hospitalization is made, arterial blood gas assay is a better objective measurement than spirometry for determination of the need for intensive care. 1° Inasmuch as spirometric testing has been associated with sudden death, recommendations for its near-routine use in the acutely symptomatic known asthmatic patient should be abandoned.
R Ben Zemenick,DO 1. WorthingtonJR, Ahuja J: The value of pulmonaryfunction tests in the management of acute asthma. Can Med Assoc J 1989;140:153-156. 2. Keren E, TibsNrani R, Canny G, et aI: Predictingthe need for hospitalization in children with acute asthma. Chest 1990;98:1355-1369. 3. Martin 7O, Elenbaas RM, PingletonSH, et al: Failure of peak expiratory flow rate to predict hospital admission in acute asthma. Ann Emerg Med 1982;11:466-470. 4. WessonJH, Sox HC, Neff RK, et al: Clinicalprediction rules: Applications and methodologicalstandards. N Engl J Med I985; 313:793-799. 5. Nowak RM, PensIerMI, Sarkar DD, et aI: Comparison of peak expiratoryflow and FEV1 admission criteriafor acute bronchial asthma. Arm Emerg Med 1982;11:64-69. 6. Lulla S, Newcomer RW: Emergency management of asthma in children. J Pediatr 1980;97:346-350. 7. Idris AH, McDermott MF, RaucciJC, etaI: Emergency department treatment of severe asthma: Metered-dose inhaler plus holding chamber is equivalent in effectiveness to nebulizer. Chest 1993;I03:665-672. 8. Lemarchand P, Labrune S, Herer B, etal: Cardiorespiratory arrest following peak expiratory flow measurement daring attack of asthma. Chest 199I;100:1168-1169. 9. gmerman CL, Cydulka RK, Skobdoff E: Survey of asthma practice among emergencyphysicians. Chest 1996;3:708-711. 10. McFadden ER: Management of patients with acute asthma: What do we know? What do we need to know? Ann Allergy 1994;72:395-389.
The authors of the article under comment chose not to reply.
ANNALS OF EMERGENCY MEDICINE
30:3 SEPTEMBER 1997