I. Leonard Bernstein, MD. Cincinnati, Ohio. Chnical Professor of Medicine and Environmental Health Sciences,. Department of Medicine, Division of Immunology ...
electrical field hyperresponsiveness in isolated human bronchial tissues.7 After ozone inhalation in dogs, neutrophils are activated to produce thromboxane A2,
which appears to be essential for hyperresponsiveness in this species.8 Polymorphonuclear leukocytes may play a role in neurogenic inflammation, in which it has been shown that increased neutrophil adhesion in the postcapillary venule is associated with loss of neutral endopeptidase activity.9 Finally, increased numbers of neutrophils and extracellular deposition of elastase were associated with loss of epithelium in several fatal cases of asthma.10 The special kinetic conditions imposed by controlled allergen challenges in the laboratory may underesti¬ mate the contribution of polymorphonuclear leuko¬ cytes to allergic inflammation. In both animal and hu¬ man challenge experiments, it has been reported that within the first few hours after neutrophilsandappear are less prominent 24 h later.11'12 There challenge are as yet no systematic human studies of possible ad¬ verse events mediated by polymorphonuclear leuko¬ cytes during the early phases of inflammation. Under in vitro conditions, detachment of epithelium from the basement membrane by neutrophil-derived products may occur within 4 h.6 Recruitment of neutrophils in reactions might also be expected in delayed asthmatic to mast cell release of leukotriene-B4. Once response polymorphonuclear leukocytes are attracted to the in¬ flammatory site,release they could perpetuate the inflamma¬ of elastase and leukotriene-B4, tory cycle by both of which could signal synthesis and exocytosis of the chemokine chemoattractant, IL-8, from epithelial cells and other neutrophils. Very little scientific data are available about the precise role of specific inflammatory cells in the re¬ modeling process that occurs in the bronchial base¬ ment membrane and lamina propria. Could neu¬ trophils contribute to the irreversible obstructive component commonly observed in chronic asthma? In addition to collagen deposition by myofibroblasts in the lamina reticularis, the normal constitutive elastin com¬ ponent in this anatomic matrix could be destroyed by elastase. These combined effects would neutrophil decrease bronchial wall elasticity and eventually lead to irreversible bronchial obstruction. As interest and investigation of asthmatic inflam¬ mation expands, it is now abundantly clear that multi¬ effector cells and their proinflammatory products ple are involved and that this inflammatory milieu may perpetuate itself through intercalated and often re¬ dundant mediator/cytokine pathways. Coupled with the fact that other non-IgE-mediated events (eg,
complexes, complement components, anaphylatoxins) usually elicit a brisk neutrophilic response, the presence of polymorphonuclear leukocytes in both allergic and nonallergic asthmatics, as reported by immune
Frangova et al, should stimulate further investigation of this cell in both early and late inflammatory events of asthma. I. Leonard Bernstein, MD Cincinnati, Ohio Chnical Professor of Medicine and Environmental Health Sciences, Department of Medicine, Division of Immunology, University of Cincinnati Medical Center.
References 1 Kaliner MA, Blennerhassett J, Austen KF. Bronchial asthma. In: Meischer PA, Muller-Eberhard HJ, eds. Textbook of ImmunoNew York: Grune & Stratton; 1976; 77:527-37 pathology. 2 Gleich GJ. The eosinophil and bronchial asthma: Current un¬
derstanding. J Allergy Clin Immunol 1990; 85:422-36
3 Azzawi M, Bradley B, Jeffery PK, et al. Identification of activated T-lymphocytes and eosinophils in bronchial biopsies in stable atopic asthmatics. Am Rev Respir Dis 1990; 142:1407-13 4 Bhalla DK, Young C. Effects of acute exposure to O3 on rats: se¬ quence of epithelial and inflammatory changes in the distal air¬
ways. Inlial Toxicol 1992; 4:17-31 HS, Devlin RB, Graham DE, et al. Ozone-induced in¬ flammation in the lower airways of human subjects. Am Rev Respir Dis 1989; 139:407-15 Venaille TJ, Mendis AHW, Philips MJ, et al. Role of neutrophils in mediating human epithelial cell detachment from native basement membrane. J Allergy Clin Immunol 1995; 95:597-606 Hallahan AR, Armour CL, Black JL. Products of neutrophils and eosinophils increase the responsiveness of human isolated bron¬ chial tissue. Eur Respir J 1990; 3:554-58 Daniel EE, O'Byrne P. Autonomic nerves and airway smooth muscle. Am Rev Respir Dis 1991; 143:S3-S5 Umeno E, Nadel JA, McDonald DM. Neurogenic inflammation ofthe rat trachea: fate of neutrophils that adhere to venules. J Appl Physiol 1990; 69:2131-6 Fujisawa T, Kephart GM, Gray BH, et al. The neutrophil and chronic allergic inflammation. Am Rev Respir Dis 1990; 141: 689-97 Coyle AJ, Tsuyuki S, Bertrand C, et al. Mice lacking the IFY-7-
5 Koren 6
7 8 9 10 11
receptor have an impaired ability to resolve a lung eosinophilic inflammatory response associated with a prolonged capacity of T cells to exhibit a Th2 cytokine profile. J Immunology 1996; 156: 2680-85 12 Fabbri LM, Maestrelli P, Saetta M, et al. Bronchial hyperreac¬ tivity: mechanisms and physiologic evaluation. Am Rev Respir Dis 143:S37-8 1991;
and Toxicology: A Society Double Standard? Or of Mice and Toads (With Apologies to Mr. Steinbeck) ...
As
^**
practicing physicians, we are accustomed to fa-
cilely manipulating the concepts and principles of risk assessment on a daily basis. Steeped in the tradi¬ tion of the scientific method, we are most comfortable with testing and differential diagnosis establishing a course of action with at least a 95% level of confidence. It is easy for us to overlook that outside our scientific CHEST / 110 / 5 / NOVEMBER, 1996
Downloaded From: http://journal.publications.chestnet.org/ on 01/23/2013
1139
in levels discipline others haveoursignificant differences will show A at look of certainty. legal colleagues is the concepts of "more likely than not," ie, (>50%), widely will not A structural
understood and used. engineer failure of a dam. structural of a a 5% chance accept These differences in comfort levels extend to society as a whole, who has shown a tremendous inability to un¬ derstand and apply the basic principles of risk assess¬ ment. This is especially true when complex pharma¬ reduced to cologic and toxicologic principles are often a 15-second sound bite for the evening news. Because the regulatory bodies charged with the assessment of the safety of pharmaceutical products are also linked to the values of society through our governmental and that when the science political system, is it surprising of toxicology runs headlong into the values of society there is an ample opportunity for selective interpreta¬ tion and application of the science? From a medical and tobacco have significant, wellstandpoint, alcohol documented adverse health consequences. Yet, society allows, and in some cases encourages, wide exposure of these toxic agents to the members of our society. The scientific toxicologic data have existed condemning those agents for years. In contrast, product liability is¬ sues related to synthetic graft materials and silicone breast implants have removed many of these products from society without scientifically valid data to support the action. Toxicology appears to be a discipline that is applied current so¬ by society inconsistently depending on themedical dis¬ cietal context. This is not unique among that value or the ciplines. However, application in no society of toxicology way changes places on the science the physiologic effects to our patients. It only changes the view society has toward the consequences. From the perspective of the patient, it matters little if the adverse effect is produced by a chemical, regulated drug, food additive, or an unregulated "natural" health food or vitamin. It is useful to remember that the most deadly poisons in the world are largely "naturally oc¬
curring" toxins.
These paradoxes are often reflected in the actions of the US Food and Drug Administration (FDA) which must balance the complexity of the science of toxicol¬ ogy and the pharmacologic evidence of an agent with the needs and desires of increasingly vocal segments of society. All of this occurs with a blending of the risk tolerances of the researcher, physician, lawyer, and engineer in the setting of complex laws and regulations the values of society. New drugs and med¬ reflectingmust demonstrate to a high degree of proba¬ ications lack of toxicity before they safety orThis bility, efficacy and market. is a time-consuming set are allowed on the of steps and frequently costs hundreds of millions of 1140 Downloaded From: http://journal.publications.chestnet.org/ on 01/23/2013
dollars. It has evolved as a mechanism to protect soci¬ ety from the toxicity and expense of potential agents. This often prodigious scientific exercise defines an analytic process which is at the heart of the modern practice of medicine and the biologic sciences. It is supported by a society that remembers the tragedy of the introduction of thalidomide in Europe. This rig¬ orous process, however, is increasingly being called into question by segments of the public, manufactur¬ ing concerns, and the government who appear willing to accept increased risk in exchange for more rapid drug approval, reduced costs of drug development, or reduced governmental regulation. It is curious that the purview of the FDA is largely confined to the some¬ what arbitrary categories of food, drugs, cosmetics, medical devices, and animal feeds. Yet "natural rem¬ edies," "herbal" aids, vitamins, and supplements do not fall under review until adverse effects are reported. We are reminded of it when we hear on the radio both a herbal shampoo's claim of hair regrowth in men with little or no efficacy or toxicity data on file and the similar claim by the manufacturers of minoxidil lotion after spending millions of dollars to demonstrate the and toxicity profile of their product. efficacy The eosinophilia-myalgia syndrome was attributed to a contaminant present in L-trytophan which was sold as a food supplement that afflicted as many as 1500 More recently, food people from 1989 to 1990. Ma containing huang, a Chinese herb supplements with significant amounts of ephedrine, has been mar¬ keted with claims to increase energy and assist in loss. Over 100 adverse effects, including deaths, weightbeen had February 1996, reported to the FDA astheof FDA after one a prompting public warning by manufacturer declined to recall the product. It is interesting to note that pharmaceutical ephedrine, unlike "herbal" ephedrine, is directly regulated by the FDA with requirements for those manufacturers to show safety and efficacy for the labeled indications. Human illness has also been reported possibly associ¬ ated with ingestion of Kombucha tea. The article by Brubacher and colleagues in this is¬ sue of CHEST (see page 1282) describes another ex¬ ample of this consequence of the two standards. Ma¬ terial sold as an aphrodisiac called "Rockhard," "Love Stone," or "Chan Su" in New York was found to con¬ tain hallucinogens, including bufotenine and cardio-
active steroids such as bufalin, cinobufagin, and cinobufetalin. These products are also found in toad venom. Their report describes the clinical toxicity, di¬ agnoses, and reports of treatment with digoxin-specific Fab fragments of this intoxication. This excellent report warns toxicologists and critical care and emer¬ gency medicine physicians of the need to stay alert to new community toxic exposure outbreaks. It also Editorials
points to the possible role for poison control centers and individual clinicians in discovering and dissemi¬ nating information regarding epidemiologic outbreaks of new toxic exposures in the community. The FDA responded appropriately by banning the importations
and sale of this material in the United States. Yet the double toxicology standard in our society still remains. We continue to allow the sale of agents listed as herbs or additives without demanding the same determina¬ tion of efficacy and risks. As physicians, we must remember our roots as "teacher" as well as "healer." It remains our responsi¬ bility to educate our elected leaders, regulatory bodies, and the public at large to the effects of all agents on the human organism (and society) whether the agent is a drug, chemical, food, cosmetic, health food, vitamin, or additive. We must maintain vigilance for new diseases and adverse effects produced by exposures and de¬ velop a warning network to investigate these effects. And after the paradox is examined and debated, we will still be asked to treat the patient, sometimes with data and sometimes with art.
Albertson, MD, PhD, FCCP Timothy R.E. Steven Tharratt, MD, FCCP Sacramento, California Division of Pulmonary and Critical Care Medicine, and the Poison Control Center, University of California Davis Regional Medical Center.
Variations in DNR Rates The Onus Is on Physicians lVTot every patient is a good candidate for cardiopul-
resuscitation (CPR). In -'"^'monary the of survival is
some
groups of
small, while the probability associated with survival is high.1"4 To the morbidity CPR provider, this balance favors withholding CPR and documenting the decision with an order not to attempt resuscitation (DNR order). Ideally, the phy¬ sician makes this decision in dialogue with the patient. The doctor brings compassion, expertise, and an ofthe probable outcomes of CPR to the understanding who shares a unique perspective on his or her patient of life and quality approach to death. A spate of recent research suggests that this dialogue should be getting easier and more common. Patients now come to the hospital with their own opinions about CPR,5 opinions that are not idiosyncratic, personal preferences but are affected by data on mortality and morbidity following resuscitation.6 In particular, pa¬ tients are more likely to decline CPR when given
patients,
quantitative information about outcomes.'8 It there¬
fore comes as a surprise to see several recent reports of wide variation in DNR rates among apparently similar populations.5910 How are we to interpret such differences in practice? Jayes and colleagues in this issue of CHEST (see page 1332) provide one approach to this question. They analyzed the APACHE III database and used logistic regression techniques to calculate an expected DNR rate for their 17,000 patients. Not surprisingly, given their previous report of wide variation in fre¬ quency of DNR orders among this population,9 they found that a statistically significant number of hospitals deviated from their expected rates. The authors do not explain this variance except to note that it did not seem to correlate with a loose, subjective assessment of or¬ ganizational structure in a subset of their ICUs. There are other clues in the recent medical litera¬ ture. Among patients in the SUPPORT study able to complete an interview,11 91% had formed an opinion regarding their resuscitation status (64% wanted CPR, 28% did not). However, only 29% had discussed these preferences with their physicians. Patients do appear eager to have these discussions. In a recent study of ICU withdrawal decisions,12 61% of patients/surro¬ gates approached about withholding or withdrawal of life support agreed to this plan immediately, and another 27% agreed within 48 h. The majority of pa¬ tients were ahead of their doctors in their readiness to discuss resuscitation status, but they were reluctant to initiate discussions about end-of-life care and fre¬ quently deferred to their physicians to raise the topic. For their part, physicians are too slow to address resuscitation status with their patients. Less than half of physicians in the SUPPORT study knew when their patients wanted to forgo CPR.5 In a prospective study of DNR discussions, when physicians did talk to patients about resuscitation, not once did they present the quantitative outcomes that patients seek to make an informed decision.13 What are the barriers to the discussion of resuscita¬ tion? The first is a desire for certainly when making end-of-life decisions. There is no lack of outcomes data in CPR: in-hospital arrest survival rates are reported from 25 to 75%, and rates of survival to hospital from less than 5 to 25%.14"16 However, these discharge are probabilities and physicians have difficulty with probabilistic reasoning when the alternative is death.17 The American Heart Association18 suggested with¬ holding CPR from patients when there were no survi¬ vors among similar patient groups in well-designed studies, and a major review of CPR suggested that such zero-benefit groups could be identified.19 They cannot. There will be some survivors in any patient population realistically considered for CPR. Such standards there¬ fore are not only poorly conceived but they may also CHEST /110 / 5 / NOVEMBER, 1996
Downloaded From: http://journal.publications.chestnet.org/ on 01/23/2013
1141
