84

EDITORIALS

8. Naveau S, Poynard T, Zourabichvili V, Poitrine A, Chaput JC: A randomized study of a coaxial fiber versus a naked fiber for endoscopic Nd:YAG laser therapy of esophageal and rectal tumors. Gastrointest Endosc 1989; 35:201-206 9. Jensen DM, Machicado G, Randall G, Tung LA, English-Zych S: Comparison of low-power YAG laser and BICAP tumor probe for palliation of esophageal cancer strictures. Gastroenterology 1988; 94:1263-1270 10. Payne-James JJ, Spiller RC, Misiewicz JJ, Silk DBA: Use of ethanol-induced tumor necrosis to palliate dysphagia in patients with esophagogastric cancer. Gastrointest Endosc 1990; 36:43-46 11. Kozarek RA, Ball TJ, Patterson DJ: Metallic self-expanding stent application in the upper gastrointestinal tract: Caveats and concerns. Gastrointest Endosc 1992; 38:1-6 12. Patrice T, Foultier MT, Yactayo S, et al: Endoscopic photodynamic therapy with hematoporphyrin derivative for primary treatment of gastrointestinal neoplasms in

inoperable patients. Dig Dis Sci 1990; 35:545-552 13. Kozarek RA, Low DE, Ball TJ: Tunable dye laser lithotripsy-In vitro studies and in vivo treatment of choledocholithiasis. Gastrointest Endosc 1988; 34:418-421 14. Faulkner DJ, Kozarek RA: Gallstones: Fragmentation with a tunable dye laser and dissolution with methyl tert-butyl ether in vitro. Radiology 1989; 170(ptl): 185189 15. Cotton PB, Kozarek RA, Schapiro RH, et al: Endoscopic laser lithotripsy of large bile duct stones. Gastroenterology 1990; 99:1128-1133

Eight Years and CountingWhat Can Americans Do? THE SEARCH FOR President Bush's "new world order" and the World Health Organization (WHO) agenda, "Health for All by the Year 2000," are neither theoretical nor rhetorical. In two previous articles, "International Health Beyond the Year 2000'" 1 and "International Health,"2 I reviewed the major challenges and some solutions to global health problems. If the United States adopts global leadership in efforts to improve the health of people everywhere, 1992 could be a pivotal year.3 Health for the "global person" could translate into "the new diplomacy"; better health could be the new lingua franca replacing the arms race and the cold war. The total US budget is about $1.5 trillion for fiscal year 1992; the WHO budget for fiscal year 1992 is only $735 million, less than the budget of the US Public Health Service's Centers for Disease Control in Atlanta, Georgia.3 One suggestion is that US assistance to the WHO can be provided as extrabudgetary funds-a special contribution above and beyond its regular assessment of 25%.3 President Bush and Congress may reduce the armed services budget by $50 to $100 million during the next five years. This peace-time "fallout" of extra funds could be used to improve health both at home and globally. In its new role as leader in efforts to improve the health of all people, the U-nited States would have to put increased pressure on the European Economic Community, Japan, Taiwan, Singapore, and other economically strong nations to contribute more to the WHO's dwindling resources as AIDS, cholera, floods, drought, and earthquakes continue to wage their own wars on

humanity.

Elsewhere in this issue, Alfred Sommer, MD, adroitly outlines global health priorities, including those affecting people in the United States, and emphasizes maximizing quality years of life at a cost society can afford.4 He identifies the organization and use of health resources as the pivot. Dr Sommer emphasizes the need to maintain close communication with other health professionals who have a common agenda; he rightfully calls for eliminating the divisive barriers hindering close relationships among global health professionals. Moreover, the International Health Medical Educational Consortium (IHMEC) has also mandated reorienting medical school academic agendas to reflect society's demands and realities.2 Dr Sommer calls for the same efforts by the schools of public health,4 which for nearly 100 years have given top priority to the health needs of society. IHMEC

task forces are currently addressing issues of clinical sites, curriculum, career track and certification, communication, and liaison in a comprehensive effort to ensure that candidates not only have health profession experience but have also "looked after sick people." Because both dollars and human resources are scarce, we must use plain economic sense and social justice and put increased pressure on our academic health centers to return to their primary mandate of working for public and societal good.5-7 Academic institutions must claim the "three-legged stool" of cost, access, and quality of care.6 Those of us in public health and medical schools must care for this "stool" and ensure that it also sprouts a fourth leg of societal priorities with equity. We must become part of the "new world order" instead of reluctant players. ANVAR VEUI, MD, FRCP, FACP(C) Associate Clinical Professor of Medicine

University of California, Davis, Medical Center Sacramento, California Division of Infectious Diseases Department of Medicine Kaiser Permanente Medical Center South Sacramento, California Assistant Editor Westem Journal of Medicine

REFERENCES 1. Velji AM: Intermational health beyond the year 2000. Infect Dis Clin North Am 1991; 5:417-428 2. Velji AM: International health (Editorial). West J Med 1991; 155:308-309 3. Gunby P: 1992 could be pivotal year in efforts to improve health of people everywhere. JAMA 1992; 267:15-19, 23 4. Sommer A: Global health leadership-A continuing US challenge. West J Med 1992 Jul; 157:71-73 5. Bryant JH, Zuberi RW, Thaver IH: Alma-Ata and health for all by the year 2000: The roles of academic institutions. Infect Dis Clin North Am 1991; 5:403-416 6. Barondess JA: The academic health center and the public agenda: Whose threelegged stool? Ann Intern Med 1991; 115:962-967 7. White KL, Connelly JE: The medical school's mission and the population's health. Ann Intern Med 1991; 115:968-972

Meta-analysis Redux-Steroids and Meningitis Revisited THERE ARE MANY CLINICAL SITUATIONS for which firm therapeutic recommendations cannot be made based on existing data. Sometimes these situations can be identified by the

fervor of the discussions surrounding them. Loud, data-free arguments, by necessity based on belief rather than a critical appraisal of relevant literature, frequently occur when new therapies become available or are applied to new uses and before there are enough trials to give clear answers to clinical questions. When more data become available, quieter deliberations are the rule as clinicians try to sort out what to do in the face of sometimes conflicting information from different studies. Elsewhere in this issue, Geiman and Smith use the techniques of meta-analysis in an attempt to sort through the available data and make a recommendation concerning the use of corticosteroids as adjunctive therapy for the treatment of pediatric patients with bacterial meningitis.1 Their excellent analysis offers a useful summary of the information available from clinical trials in this area. Their recommendations for use are somewhat firmer than I think the data allow at this time. In the pediatric studies included in their meta-analysis,' early2 and late3 neurologic sequelae and moderate or more severe bilateral hearing loss2 were significantly diminished by dexamethasone therapy in only one trial each. When

THE WESTERN JOURNAL OF MEDICINE

o

JULY 1992

e

* * JULY 1992 JOURNAL OF MEDICINE THE WESTERN

157 157

9 *

85

1

85~~~~~~~~

1

results of all available studies were combined, however, patients treated with dexamethasone had improved neurologic outcome at hospital discharge and at follow-up examination and less bilateral hearing loss at follow-up examination. The authors conclude that dexamethasone use in conjunction with bactericidal antimicrobial therapy is useful in reducing neurologic sequelae and hearing loss following bacterial meningitis in children, and they recommend its routine use, especially for severe disease caused by Haemophilus influenzae. Existing data are insufficient to allow this strong a recommendation. Although a useful technique for summarizing data, the combining of results from several clinical studies is always open to criticism, as shown by an examination of the details of the studies included in this meta-analysis. Four of the studies2`4 were done under the direction of a single investigator, and two were reported in a single publication.2 While those studies were well designed and carefully reported, the evidence would be stronger if there were trials from a variety of different groups. Different antibiotics were used in the various trials. For example, medications for meningitis caused by H influenzae were chloramphenicol plus sodium sulfadiazine,5 ampicillin plus chloramphenicol,6 cefuroxime,2'4 ceftriaxone,2 or cefotaxime.3 Dexamethasone doses also were not uniform in the six studies included in the pooled data. One trial was published in 1969, and the results may not have a strong relationship to medical practice in 1992. Another trial was performed in Egypt,6 and patients in general had more advanced disease at presentation than the subjects in the other trials. The results may not be comparable to those obtained in other locales with patients with different disease severity. It is possible that dexamethasone appears to be useful in these studies in part because of the high rates of poor outcome in the control groups. Four of the trials used a second- or third-generation cephalosporin as the antibiotic therapy for these patients with bacterial meningitis.2`4 The placebo groups in these studies showed high rates of neurologic abnormality (29/149; 19.5%)2-6 and hearing loss (22/157; 14%) at follow-up examination.2-4'6 High rates of poor outcome in the control groups have the effect of increasing the measured differences in risk between treatment and control groups in a given population sample, even if the outcome in the treatment group in the study sample is no better than in a control group (not treated with dexamethasone) at another institution or from another era. The question to be answered is not "Does dexamethasone show benefit in these trials?" but rather, "How do the rates of poor outcome in the dexamethasone-treated study patients compare with the rates of poor outcome in patients treated in other institutions with other antibiotics but without dexamethasone?" Although dexamethasone use decreases the risk of poor outcome in patients treated with cephalosporins, there may be no improvement when compared with the outcome of patients treated with ampicillin and chloramphenicol during a previous era. Pomeroy and co-workers evaluated neurologic function following bacterial meningitis in 185 patients treated with ampicillin and chloramphenicol between 1973 and 1977.7 The mean duration of follow-up for these patients was 8.9 years.' Of the 185 patients, 8 (4%) had neurologic abnormalities at follow-up and 18 (10%) had persistent hearing loss. Comparison of the rates of poor outcome in the control

groups of the four cephalosporin-dexamethasone studies2`4 with the rates reported by Pomeroy and colleagues7 suggests that ampicillin plus chloramphenicol therapy is associated with a better neurologic outcome at follow-up examination (risk difference [RD] = -15%; 95% confidence interval [CI] -22% to -8%) and a lower risk of persistent hearing loss (RD = -4%; 95% CI -11% to 3%) than is therapy with cephalosporin antibiotics alone. Comparing the rates of poor outcome in the treatment groups of the four cephalosporin-dexamethasone studies2-4 with the rates reported by Pomeroy and associates suggests that the use of a second- or third-generation cephalosporin plus dexamethasone results in a lower rate of late hearing loss (RD= -6%; 95% CI -11% to -0.5%) than does therapy with ampicillin and chloramphenicol alone. There is no difference, however, in the rates of neurologic abnormality between the two series, and the risk of late neurologic sequelae actually is lower in the ampicillin-chloramphenicol-treated group (RD = 3%; 95% CI -2% to 8%). These somewhat complex comparisons between treatment, control, and other groups of patients with bacterial meningitis are open to as much criticism as the original metaanalysis that suggested them. The point can be made, however, that the increase in inflammation associated with the use of second- and third-generation cephalosporin antibiotics alone leads to worse neurologic sequelae after treatment than the neurologic sequelae that follow the use of ampicillin or chloramphenicol alone. To make the neurologic outcome after meningitis therapy with cephalosporins as good as that after therapy with ampicillin and chloramphenicol, dexamethasone must be added to diminish the added inflammation (and associated neurologic damage) that may come from the use of rapidly bactericidal antibiotics like cefotaxime and ceftriaxone. In fact, this vigorous inflammatory response from bacterial lysis has led to the suggestion that dexamethasone should be administered before antibiotics.3 Since submission of the manuscript of Geiman and Smith,' there have been two other publications of interest on this subject.8'9 McCracken and colleagues reported a retrospective review of 97 children in Dallas, Texas, with pneumococcal meningitis from 1984 to 1990; they showed no statistically significant differences in the neurologic or audiologic sequelae of patients treated with dexamethasone compared with those not so treated. The Infectious Diseases Society of America in January 1992 published recommendations concerning glucocorticosteroid use in bacterial meningitis,9 stating that there exists "moderate evidence to support a recommendation for use" of dexamethasone as adjunctive therapy for "infants and children with early Haemophilus influenzae meningitis." The authors also stated that because of inadequate data, they could not make recommendations for corticosteroid use in patients with meningitis caused by Streptococcus pneu-

moniae.9 There seem to be enough data to justify a provisional recommendation that dexamethasone be given to patients with bacterial meningitis caused by H influenzae, especially if a second- or third-generation cephalosporin is used for antibiotic therapy. The combination of bactericidal antibiotic (cefotaxime or ceftriaxone) and anti-inflammatory therapy probably helps reduce the risk of persistent hearing loss compared with other therapeutic options. The dose of dexamethasone used in most studies is 0.6 mg per kg per day given in

86

EDITORIALS

four divided doses for a total of four days. Although there are theoretical reasons3 to suggest that giving dexamethasone before antibiotics might be beneficial, there are no clinical data proving that the use of corticosteroids before antibiotics is better than their use after antibiotics. PETER L. HAVENS, MD, MS Associate Professor of Pediatrics and Biostatistics/Clinical Epidemiology Division of Critical Care and Infectious Diseases Medical College of Wisconsin and Children's Hospital of Wisconsin Milwaukee

REFERENCES 1. Geiman BJ, Smith AL: Dexamethasone and bacterial meningitis-A meta-analysis of randomized controlled trials. West J Med 1992 Jul; 157:27-31 2. Lebel MH, Freji BJ, Syrogiannopoulos GA, et al: Dexamethasone therapy for

bacterial meningitis: Results of two double-blind, placebo-controlled trials. N EngI J Med 1988; 319:964-971 3. Odio CM, Faingezicht I, Paris M, et al: The beneficial effects of early dexamethasone administration in infants and children with bacterial meningitis. N Engl J Med 1991; 324:1525-1531 4. Lebel MH, Hoyt MJ, Wagner DC, Rollins NK, Finitzo T, McCracken GH Jr: Magnetic resonance imaging and dexamethasone therapy for bacterial meningitis. Am J Dis Child 1989; 143:301-306 5. Belsey MA, Hoffpauir CW, Smith MH: Dexamethasone in the treatment of acute bacterial meningitis: The effect of study design on the interpretation of results. Pediatrics 1969; 44:503-513 6. Girgis NI, Farid Z, Mikhail IA, Farrag I, Sultan Y, Kilpatrick ME: Dexamethasone treatment for bacterial meningitis in children and adults. Pediatr Infect Dis J 1989; 8:848-851 7. Pomeroy SL, Holmes SJ, Dodge PR, Feigin RD: Seizures and other neurologic sequelae of bacterial meningitis in children. N EngI J Med 1990; 323:1651-1657 8. Kennedy WA, Hoyt MJ, McCracken GH Jr: The role of corticosteroid therapy in children with pneumococcal meningitis. Am J Dis Child 1991; 145:1374-1378 9. McGowan JE, Chesney PJ, Crossley KB, LaForce FM: Guidelines for the use of systemic glucocorticosteroids in the management of selected infections. J Infect Dis 1992; 165:1-13