Trends in respiratory medicine: Tuberculosis and other infections

Trends in respiratory medicine: Tuberculosis and other infections

G. Douglas Campbell

Antimicrobial resistance is an escalating dilemma

ABSTRACT: The Editorial Board of The Journal of Respiratory Diseases continues its discussion of the major developments concerning respiratory infections. They begin with a review of the current status of tuberculosis, the effect of directly observed therapy, and advances in diagnostic techniques. The Board also discusses the increased recognition of atypical mycobacterial diseases, particularly in non-HIV-infected patients, and the advances made in treating such diseases. The roundtable continues with a discussion of the impact of antimicrobial resistance, the challenge of distinguishing between acute bronchitis and pneumonia, and the best approach to treatment. (J Respir Dis. 1999;20(10):658-673)


The Journal of Respiratory Diseases (JRD): HIV infection has been blamed to a large extent for the resurgence of tuberculosis, but some experts have said that complacency on the part of the health care community and the government has also contributed to the resurgence. Is tuberculosis now being treated as seriously as it should be?

Dr File: Ten years ago, one of my biggest concerns was what would happen with tuberculosis over the next decade. We were in the middle of that tremendous resurgence that started in 1984 and, by 1989, was escalating. If that trend had persisted, tuberculosis would be an even greater public health problem. Recently, there has been a decrease in the number of new cases in the United States. However, the consequences of the recent increase in the prevalence of tuberculosis will be felt for years to come.

Dr Lillington: Tuberculosis is still the number one respiratory disease process in much of the rest of the world. While the incidence is decreasing in this country, it is increasing in other countries.

Dr Squillace: My impression is that tuberculosis is a disease that generally clusters in big cities, but there is wide geographic variation. Last year, there were only nine new cases of tuberculosis in Virginia; seven cases were reported in 1997. Charlottesville alone has a population of 150,000.

Dr Matthay: That is an extraordinarily low incidence of tuberculosis.

Dr Campbell: Or that may reflect low reporting, but I agree that there is a clustering of disease. At one point during this resurgence, about 50% of US counties had not reported one case of tuberculosis. In northwestern Louisiana, we have the second highest incidence of active tuberculosis in the state, but it is not close to the incidence in New Orleans.

Dr File: As a result of the increase in tuberculosis, we have had to become more compliant with the guidelines for identifying patients and contacts and for following precautions, such as those for negative-pressure rooms and the use of masks. We have also had to become more compliant with guidelines for tuberculin skin testing of health care providers.

Dr Campbell: The resurgence has also led to the use of directly observed therapy (DOT).

Dr File: There was a big debate over whether we should use DOT, but it is probably the most effective intervention. DOT has really turned this problem around (Table).

JRD: Is DOT indicated for all patients with tuberculosis?

Dr Campbell: The vast majority of patients who have active pulmonary tuberculosis should get DOT, regardless of HIV status. When DOT was instituted in Mississippi, multidrug resistance was substantially reduced. When we instituted it in Louisiana about 4 years later, we also saw a drop in multidrug resistance. DOT is more expensive than conventional therapy, but it is effective when given twice weekly.

We closely monitor these patients because many of them are unable or unwilling to finish a whole course of therapy, and some of them are homeless. Only about 50% of patients will take a full course of therapy if they are not directly observed. The public health department will go out into the community to look for these people if they have to.

Dr File: It is fascinating to talk to public health officials from New York, where tuberculosis is still related to HIV infection and injection drug use. They work with the police to find tuberculosis patients and entice them to take their medications. And it works.

Dr Lillington: This was partly a result of changing laws to make it compulsory to incarcerate people if hey resisted. When I grew up with tuberculosis–and I say that because 35% of my schoolmates in grade 6 or 7 had positive tuberculin skin tests–the treatment was to go o the sanatorium for 3 or 4 years if you were lucky enough to live that long. In 1978 and 1979, the CDC’s goal was to eradicate tuberculosis in this country in the next 10 years. Then HIV changed everything. But HIV was not the only factor contributing to the increase in tuberculosis. Another factor was that some sanatoria or hospitals were closed, and mentally ill patients became homeless.

In addition, there was an attempt to mainstream tuberculosis as a disease that all primary care physicians should treat. But not all physicians had the appropriate information and, therefore, some of them gave monotherapy or no therapy at all. There has been enormous improvement in the last few years, and compliance is the key.

Dr Campbell: We have seen increasing multidrug resistance, not in foreign-born persons, but in schizophrenic or other mentally ill persons. We see it in patients who stop taking the antituberculosis drugs after 1 or 2 months, come back 6 months or 1 year later, and then take another course for a couple of months.

Dr File: Clearly, one of the major problems associated with the resurgence of tuberculosis in the latter part of the 1980s has been multidrug resistance. Several factors have been responsible for this, including inappropriate therapy, noncompliance among patients, and residence in institutions and homeless shelters. In addition, clinicians should consider multidrug-resistant tuberculosis in patients who have had close contact with someone in whom the infection is suspected or in those who have been to areas where multidrug-resistant tuberculosis is common.

Appropriate therapy depends on in vitro susceptibility test results and diligent DOT. New drugs may become available in the next decade, but it has been both stringent surveillance and therapy that have reduced the prevalence of multidrug-resistant tuberculosis in the United States. Therefore, it is of utmost importance that sufficient funding continue for surveillance and education programs to control this problem.

Dr Lillington: We used to blame the migrant workers for tuberculosis in California. In the mid-1970s, when I was the tuberculosis controller for Sacramento County, we did a couple of studies of skin testing in migrant farm workers. Out of 400 or 500 persons, we found only 1 person with active tuberculosis and about 10 with positive skin tests.

JRD: What has been the impact of the newer diagnostic techniques? Dr Lillington: Our ability to diagnose tuberculosis rapidly and accurately with molecular biologic techniques represents a fantastic improvement. Not only can you make the diagnosis within 24 hours instead of waiting 6 weeks for a culture, but you can also prove that the organism is Mycobacterium tuberculosis and not another organism. The older tests could not do that (Figure 1). Moreover, the test is much more sensitive, and the specific typing of DNA patterns makes it possible to identify who the patient got the infection from. It is remarkable and represents a major advance.

Dr Campbell: A recent article suggested that smear-negative, sputum culture-positive patients can transmit tuberculosis.[1] The incidence of transmission is lower than it is with smear-positive patients, but it does occur Should this affect the way that we treat patients with tuberculosis?

Traditionally, if I had a patient with upper lobe cavitary disease and three consecutive negative sputum smears, I would consider moving that patient to the ward, rather than keeping the patient under isolation in a negative-pressure room. Is this no longer appropriate?

Dr Lillington: The traditional approach is based on a surprising lack of direct information on how safe it is to be in contact with someone who is known to have tuberculosis but has been receiving therapy for 10 days or 2 weeks and now has negative smears. Clearly, the patient is less infectious, but the conditions still may not be safe. I have always had concern about that and have emphasized that these patients should be watched very carefully You should perhaps wear a mask, and try to prevent them from coughing.

Dr Campbell: Some interesting research shows that not all Mycobacteria spread at the same rate; some are more virulent. Also, there is variability in the host response to M tuberculosis exposure and infection. As with HIV infection, some persons are more resistant than others.

I have a question about short-course preventive therapy. In the past, we would give isoniazid for 6 months, and if the patient was allergic to it, we would consider switching to rifampin. But now three- or four-drug regimens are being given to HIV-infected patients for 2 months. We are starting to compress the duration of therapy in these patients. Is this a reasonable approach?

Dr File: The recent CDC recommendations concerning tuberculosis and HIV include a regimen of rifampin and pyrazinamide daily for 2 months for “preventive therapy” but I have not had any clinical experience with this regimen.[2] As with any regimen, compliance must be carefully evaluated.

Atypical mycobacteria

Dr Lillington: We have been talking about typical tuberculosis, but what changes have you seen in diseases that are caused by atypical mycobacteria, such as Mycobacterium avium-intracellulare (MAI)? Dr File: We are seeing much more MAI, not just in HIV-infected patients but also in older patients, such as those with chronic obstructive pulmonary disease (COPD), who are receiving immunosuppressive therapy or corticosteroids. More HIV-negative patients are being referred to me for pulmonary disease caused by Mycobacterium kansasii and MAI.

Dr Campbell: I have seen some patients whose MAI infection was diagnosed because they had a solitary pulmonary nodule. One patient presented with weight loss and hemoptysis; the solitary pulmonary nodule turned out to be a granuloma that grew MAI. Other patients had cough, low-grade fever, and weight loss; tuberculosis was suspected, but it turned out to be MAI infection.

I have seen several cases of what they call the Lady Windermere syndrome–a 50- or 60-year-old woman with a smoking history who presents with a nonproductive cough. These women do not have COPD; they have MAI infection. Dr Matthay: At Yale, the radiologists sometimes alert clinicians to the possibility of MAI. For example, they can look at the chest CT scan of a patient with infiltrates or nodules and, without even knowing the details of the case, say, “I see a tree and bud appearance in addition to mild bronchiectasis. Does this patient happen to have MAI disease?”

One patient was referred to me because she had multiple recurrent episodes of acute bronchitis. She had a history of multiple pneumonias starting in childhood and had bronchiectasis on her chest CT scan. She had MAI isolated from her sputum. By the way most of my patients with bronchiectasis have been women; they tend to be thin, asthenic women who may or may not have been cigarette smokers.

Incidentally, I am not suggesting that MAI is the only organism we find in this setting. Many of these patients have Pseudomonas and other organisms.

Dr Lillington: It is essentially a new disease. In the past, bronchiectasis was a unilateral lobar disease that developed in childhood following whooping cough or recurrent pneumonias. On bronchogram or CT, it now appears as a cylindrical bronchiectasis rather than the varicose dilatations and cystic changes seen in the past.

Dr Squillace: And the problem is not allergic bronchopulmonary aspergillosis (ABPA)?

Dr Matthay: That’s a good question. ABPA occurs in certain parts of the United States but is rare in the Northeast. It responds to corticosteroids. Do you see much of it in Virginia?

Dr Squillace: No. We look for it when we see patients with the presentation we have been discussing, but the diagnosis of ABPA is fairly uncommon.

Dr Lillington: It is not particularly common in California either. One possible reason for the greater prevalence of ABPA in Great Britain is that the climate and the soot that used to be in the air may have predisposed asthmatic patients to it.

JRD: What progress has been made in the treatment of MAI disease?

Dr File: Ten years ago, it was very difficult, because it required four-or five-drug therapy, which patients could not tolerate. With the advent of the new macrolides and fluoroquinolones, treatment has been much more successful. I often use two-drug therapy, although three-drug therapy is recommended (for example, clarithromycin, a fluoroquinolone, and ethambutol).

It is my impression that women with the Lady Windermere presentation do not respond as well to treatment as do the male patients or those who have this problem secondary to COPD or other chronic lung disease.

Antimicrobial resistance

JRD: Antimicrobial resistance is clearly a significant development in infectious disease. To what extent has it affected empiric therapy for

respiratory infections?

Dr Campbell: When we developed the American Thoracic Society (ATS) guidelines for community-acquired pneumonia (CAP) in 1992, drug-resistant Streptococcus pneumoniae was not thought to be a major problem in the United States. [3] In the last 3 years, however, awareness of this problem has increased. The National Committee for Clinical Laboratory Standards (NCCLS) established the minimum inhibitory concentration (MIC) break points for drug-resistant pneumococci at 0.1 to 1 [micro]g/mL for intermediate resistance and at 2 [micro]g/mL and above for high resistance, but those break points were based on meningitis data.

Dr File: The NCCLS set these break points at one level to be consistently followed for all infections, which we now know is not clinically appropriate. Because meningitis is a closed-space infection, the break point needs to be lower than it does for pneumonia or even bacteremia without meningitis. The NCCLS may change the break points to make them disease-specific; so, for example, the break point for a cerebrospinal fluid isolate would be different from that for a blood or respiratory isolate.

Dr Campbell: There is a considerable amount of intermediate resistance and low-level high resistance (2 [micro]g/mL) but not much higherlevel resistance (above 4 [micro]g/mL). There is evidence that if the MIC is 2 [micro]g/mL or less, you can treat pneumonia with high-dose penicillin, or ceftriaxone or cefotaxime, and the patient will do quite well. [4] But if the patient has meningitis, it is a different story.

Dr File: Two reports presented at the Infectious Disease Society of America (IDSA) meeting last year are of concern. One from the CDC surveyed invasive S pneumoniae isolates from about 21 centers. Other multicenter studies have analyzed thousands of pneumococcal isolates across the country.

In one study, 4% of all isolates had penicillin MICs of 4 [micro]g/mL or greater; in another, it was 7%. This is very disturbing.

Dr Campbell: Patients with the highest risk of resistance are those who have had recent (in the last 3 months) antibiotic exposure and those who tend to take antibiotics often, such as HIV-infected patients and persons older than 70 years. Also, there has been a substantial increase in antibiotic use in children younger than 15. The drug-resistant S pneumoniae steering committee suggested that antibiotic use has increased 50% over the last 10 years. In particular, resistance is being observed in children younger than 6 years and in persons from middle- and upper-income groups, who can afford antibiotics.

Of importance, the resistance is not due to [beta]-lactamase production; rather it is a change in the way the antibiotic binds, so certain [beta]-lactams work better than others. Resistance also tends to occur in certain pneumococcal serogroups–about 85% of resistance occurs in six serogroups. The new adjuvant pneumococcal vaccine may disrupt some of this resistance.

Another area of concern is the development of vancomycin-resistant pneumococcus. We have seen vancomycin-resistant enterococci, and we now are starting to see Staphylococcus aureus with intermediate resistance to vancomycin.

Generally, penicillin-resistant pneumococcus is a marker for multidrug resistance. More than half of the isolates that are resistant to penicillin are probably resistant to another antibiotic, and a substantial number are resistant to three or more antibiotics, often the macrolides. This is a concern for patients with COPD exacerbations.

Dr File: Although surveillance studies have indicated that about one third of pneumococcal isolates are resistant, we are not seeing clinical failures with the macrolides. We still consider the macrolides to be appropriate therapy, at least for early empiric use.

JRD: Do you expect new antimicrobials to be available in the next few years that will be effective against these resistant organisms?

Dr File: There are several new agents in established drug classes (new fluoroquinolones, for example) as well as new classes of drugs that are undergoing trials or awaiting release–all of which will have greater activity against drug-resistant S pneumoniae. The new classes include streptogramins, azolidinones, ketolides, everninomicin, glycopetide agents, and a minocycline derivative (such as glycylcycline).

JRD: What measures should be taken to help prevent the development of resistance?

Dr File: Certainly one of the biggest problems concerning respiratory diseases over this last decade has been the emergence of resistance of respiratory pathogens. About three quarters of all oral antimicrobial agents prescribed are for respiratory infections, and most of that is probably unwarranted.

We need to advocate more judicious use of antibiotics because inappropriate use is what is promoting resistance more than anything else. This means using antibiotics only when necessary. For example, we should not give antibiotics for “the common cold” or for acute bronchitis if the patient does not have chronic lung disease. About 95% of episodes of acute bronchitis are caused by viruses.

However, more than 50% of patients who present with a cold will get antibiotics from their physician, as will two thirds of those with a diagnosis of acute bronchitis. One reason for this is patient pressure. The CDC and other organizations, such as the American Academy of Family Practice, American Academy of Pediatrics, and American Society of Microbiology, have developed brochures to educate patients about the problems associated with overuse of antibiotics. Some primary care providers tell me that putting these brochures in their examining rooms has helped reduce patient requests for antibiotics. Another important point is that surveys of patient satisfaction show that patients are much more satisfied if the physician gives just a 30-second explanation for why he or she is not prescribing antibiotics, as opposed to giving antibiotics without explanation.

Another factor that would help guide physicians in giving appropriate antibiotic therapy is to improve the distinction between acute bronchitis and CAP and between viral rhinosinusitis and secondary bacterial rhinosinusitis. Inappropriate antibiotic use could also be reduced by improving the differentiation between acute otitis media, which is usually bacterial, from otitis media with effusion, which is not [Editor’s note: see “The debate over otitis media”].

JRD: Can you give some examples of how to make those distinctions?


Dr File: Many patients get viral rhinosinusitis two or three times a year, but bacterial rhinosinusitis is much less common. Viral rhinosinusitis typically occurs with the common cold. The patient may feel some pressure and may have purulent drainage but usually does not have fever or severe sinus tenderness. Fever and significant point tenderness are more likely to occur with bacterial infection.

Other clues that may suggest bacterial infection are dental pain, dullness to transillumination, and a change in color of the nasal discharge. Contrary to popular belief, green nasal discharge does not necessarily mean bacterial infection; green discharge can occur with viral infection too. But a change from green to yellow can be a sign of bacterial infection. In addition, a duration of symptoms of more than 7 days is more suggestive of bacterial rhinosinusitis.

CAP vs acute bronchitis

Dr File: Differentiating CAP from acute bronchitis on the basis of the history and clinical findings is difficult, unless the patient has classic egophony localized to one area. The obvious way to differentiate acute bronchitis from CAP is to get a chest radiograph. Many clinicians follow the ATS [3] or IDSA [5] guidelines and treat CAP without obtaining a chest radiograph. An important study by Fine’s group [6] reviewed the literature and showed that the correlation between clinical manifestations and chest radiographic findings is not very strong.

Dr Matthay: If an adult presents with a recent onset of purulent sputum production, and the chest radiograph does not indicate pneumonia, should antibiotics be used?

Dr File: When acute bronchitis accompanies the common cold, the cough often persists for about 14 days. I would wait 14 days before considering antibiotics, at least until we have better rapid diagnostic tests to define the 1% to 4% of patients who may have Chlamydia pneumoniae, Mycoplasma pneumoniae, or even Bordetella pertussis infection. That does not mean you cannot help the patient, because a [beta]-agonist will provide symptomatic relief.

Dr Campbell: The Canadian guidelines make a similar recommendation, which is to wait 7 to 10 days. [7] If the cough persists, you can give antibiotics to cover Chlamydia and Mycoplasma infections. Since the pertussis vaccine is protective for only 8 to 10 years, we are also seeing B pertussis infection in adults, although it is not the classic whooping cough presentation.

Dr File: About 75% of patients with CAP are not treated in the hospital, and most of them do not have a chest radiograph taken. They are given the diagnosis of CAP when in fact many of them do not have CAP. We did a study in our emergency department (ED) last year that showed that fewer than half of the patients who left the ED with a diagnosis of CAP had a chest radiograph taken, and fewer than half of those had positive chest radiographic findings.

Dr Matthay: If the chest radiograph is positive, would you start with a macrolide?

Dr File: A macrolide or doxycycline, unless there is concern about the possibility of drug-resistant pneumococcus.

Dr Campbell: Considering the number of patients who present with cough, are chest radiographs indicated for all of them? If a 70-year-old person presents to the ED with hypoxemia, a chest radiograph is reasonable. But what about a 21-year-old who presents with cough, fever, and crackles? When we developed the ATS guidelines in 1992, we did not want to specify that chest radiographs were always indicated for legal/medical reasons.

Dr File: Yes, that is a problem. When developing these guidelines, you have to be careful about how they will be followed from the litigation standpoint. But it is important to stress that often the only way to distinguish acute bronchitis (for which antibiotics are generally not warranted) from mild outpatient CAP is by the chest radiograph.

Dr Matthay: The concern is that you will miss a cancer or some chronic lung disease if you do not obtain a chest radiograph. But with an acute situation, perhaps if the patient is sick enough to lead you to consider using antibiotics, a chest radiograph is probably indicated. When antibiotics are warranted, do you think it is better to start with one of the traditional, narrowerspectrum antibiotics rather than the new fluoroquinolones?

Dr Lillington: Patients seem to get better (or not) equally quickly either way.

Dr File: So far, studies suggest that the “traditional” agents (doxycycline or a macrolide) are as effective as the new fluoroquinolones for treating uncomplicated outpatient CAP. One of the consequences of using a limited number of antibiotics is that resistance can be induced. Some people are evaluating the practice of “cycling” in the hospital: one month, you can use a broad-spectrum antibiotic or carbipenem; the next month, you use a [beta]-lactamase inhibitor; the next month, you might use a fourth-generation cephalosporin, such as cefepime. But you would not use the nondesignated antibiotics unless you had directed therapy against the specific pathogen that was isolated (Figure 2).

Dr Campbell: One of the reasons that we have not encountered as much fluoroquinolone resistance as we have penicillin resistance is that these antibiotics are not approved for use in children. It concerns me that some of the fluoroquinolones are being considered for use in children.

Dr Matthay: In other words, giving the fluoroquinolones to children will eventually result in the emergence of resistance?

Dr Campbell: Yes. Also, we have been using fluoroquinolones often and are seeing specific side effects, such as tendinitis and tendon rupture.

Dr Matthay: To be fair to the fluoroquinolones, when should we use them?

Dr File: The fluoroquinolones are appropriate as first-line therapy for CAP if there are risk factors for drug-resistant pneumococcal infections–for example, if the patient has taken antibiotics in the last 2 or 3 months or lives with or has contact with someone who attends a day-care center. I have seen many patients with drug-resistant pneumococcal infection who frequently take care of their grandchildren who attend day care. Lesser risk factors are older patient age, underlying conditions, and HIV infection. Fluoroquinolones are also appropriate if therapy with a macrolide or doxycycline has failed or if a drugresistant isolate has been cultured.

Dr Campbell: Do you think that all parents and grandparents should receive the pneumococcal vaccine?

And should health care providers receive the vaccine?

Dr File: All persons older than 65 years or who have other risk factors, such as cardiopulmonary disease or another chronic condition, should receive the vaccine. In addition, the new conjugate vaccine engenders a much better immuno-logic response and is much more effective. It actually prevents colonization, which the polysaccharide vaccine does not do. It would be significant if we could eradicate pneumococcal colonization in children, and it would eventually have an effect on adults.

At a vaccine meeting I recently attended, Pierce Gardner presented interesting data about vaccine-preventable deaths. Last year in the United States, there were 40,000 to 60,000 potentially vaccine-preventable deaths in adults. Clearly, we should generate more effort toward making sure that adults are appropriately immunized.

Dr Squillace: If a patient has been vaccinated before age 65, should he or she be revaccinated after age 65? Are there any other proven effects of revaccinating?

Dr File: Yes, revaccination is indicated in this situation. The recommendations are based on evidence that protection with the present polysaccharide vaccine wanes with time and that revaccination increases antibody levels. How this translates into protection, however, is still not known.

Dr Campbell: Would you give the adjuvant vaccine to a 70-year-old COPD patient who has multiple exacerbations? The reason I ask is that while the vaccine prevents invasive disease, maybe it would also “rev up” the immune system so that lung function would deteriorate even more rapidly. I am thinking of research on patients with cystic fibrosis who are colonized with Pseudomonas.

Dr File: I think the pathophysiology Pseudomonas in cystic fibrosis and bronchiectasis is a bit different. Until the data are known, I would still recommend the pneumococcal vaccine.

Dr Campbell: And everyone should get the influenza vaccine.

Dr File: No question about it. Actually we are much better at giving the influenza vaccine than we are the pneumococcal vaccine.

JRD: Have there been any significant changes in geographic trends in pneumonia? For example, in the past, the Pittsburgh area has had a higher incidence of legionnaires’ disease.

Dr File: There are geographic differences regarding the endemic fungi (Coccidioides immitis, Histoplasma capsulatum, Blastomyces dermatitidis) and other organisms, such as the newly described sin nombre virus (the virus associated with hantavirus pulmonary syndrome). Regarding Legionella, there does appear to be a geographic variance of incidence, since more cases are reported from the north central and northeastern region of the United States. However, some of this variation may be due to the extent of diagnostic testing. There also may be local environmental factors that are important but still not well defined.

JRD: To what extent has managed care affected the decision to hospitalize patients with pneumonia or to discharge them early, or maybe to consider using intravenous home antibiotics?

Dr File: One of the positive effects is that it has taught us that we do not need to hospitalize all these patients. We do not even need to use conventional therapy in most cases. Now there is validated information that helps us predict which patients are sick enough to require hospitalization.[8]

JRD: Some years ago, a Massachusetts team speculated about a possible link between C pneumoniae and coronary artery disease. The evidence is becoming stronger, according to several recent reports.[9-11] Do you foresee any implications for therapy?

Dr Campbell: The move to reduce hospitalization and length of stay has been an impetus for us to look at how long a patient has to receive intravenous antibiotics. Many physicians are comfortable giving intravenous therapy to most patients with CAP for 2 to 4 days; when their white blood cell count and fever improve, you can switch them over to oral therapy and send them home that day.

The question about who should be admitted with pneumonia has been addressed by Fine and colleagues.[8] They analyzed a variety of factors, categorizing patients into different risk groups, and proposed a way of stratifying patients. There was some concern about that study because it was possible for a 20-year-old person in septic shock to not have a high enough score to be admitted, and there was also some concern that the scoring system could be abused by a managed care setting. But, generally, most of the changes in the past 10 years have helped us treat patients effectively.

Dr File: Although there is clearly an epidemiologic association between C pneumoniae and atherosclerosis, we are not yet certain of the specific role C pneumoniae plays in the pathogenesis. The hypothesis that C pneumoniae may cause arterial disease is plausible but unproven.

Some preliminary trials suggested a possible benefit of antimicrobial therapy in patients who have had coronary events, but we need better scientific evidence and controlled studies before recommending therapy. At present, there is not enough evidence to suggest an implication for antimicrobial therapy.

Roundtable participants

* G. Douglas Campbell, Jr, MD, Professor of Medicine, Division of Pulmonary/Critical Care Medicine, Louisiana State University, Shreveport.

* Thomas M. File, Jr, MD, Professor of Internal Medicine, Northeastern Ohio Universities College of Medicine, Rootstown; Chief, Infectious Disease Service, Summa Health System, Akron.

* Glen A. Lillington, MD, Clinical Professor of Medicine, Stanford University, Stanford; Professor of Medicine Emeritus, Division of Pulmonary Medicine/Critical Care, University of California, Davis.

* Richard A. Matthay, MD, Professor of Medicine; Associate Director, Pulmonary and Critical Care Section, Yale University, New Haven, Conn.

* Susan P. Squillace, MD, Associate Professor of Family Medicine; Assistant Professor of Internal Medicine, University of Virginia, Charlottesville.

Editorial Board member William W. Busse, MD (Professor of Medicine and Chief, Section of Allergy and Clinical Immunology, University of Wisconsin, Madison), was unable to attend.


(1.) Behr MA, Warren SA, Salamon H, et al, Transmission of Mycobacterium tuberculosis from patients smear-negative for acid-fast bacilli. Lancet. 1999;353;444-449.

(2.) Centers for Disease Control and Prevention. Prevention and treatment of tuberculosis among patients infected with human immunodeficiency virus: principles of therapy and revised recommendations. MMWR. 1998;47(RR-20):1-58.

(3.) Niederman MS, Bass JB, Campbell GD Jr, et al. Guidelines for the initial empiric therapy of community-acquired pneumonia: proceedings of an American Thoracic Society Consensus Conference. Am Rev Respir Dis. 1993;148:1426.

(4.) Pallares R, Linares U, Vadillo M, et al. Resistance to penicillin and cephalosporin and mortality from severe pneumococcal pneumonia in Barcelona, Spain. N Engl J Med. 1995;333:474-480.

(5.) Bartlett JG, Breiman RF, Mandell LA, File TM Jr. Community-acquired pneumonia in adults: guidelines for management. Clin Infect Dis. 1998;26:811-838.

(6.) Metlay JP, Kapoor WN, Fine MJ. Does this patient have community-acquired pneumonia? Diagnosing pneumonia by history and physical examination. JAMA. 1997;278:1440-1445.

(7.) Balter MS, Hyland RH, Low D, et al. Recommendations on the management of chronic bronchitis: a practical guide for Canadian physicians. Can Med Assoc J. 1994;151(suppl):S88-S23.

(8.) Fine MS, Auble TE, Yealy DM, et al. A prediction rule to identify low-risk patients with community-acquired pneumonia. N Engl J Med. 1997;336:243-250.

(9.) Maass M, Bartels C, Engel PM, et al. Endovascular presence of viable Chlamydia pneumoniae is a common phenomenon in coronary artery disease. J Am coll Cardiol. 1998;31:827-832.

(10.) Davidson M, Kuo CC, Middaugh JP, et al. Confirmed previous infection with Chlamydia pneumoniae (TWAR) and its presence in early coronary atherosclerosis. Circulation. 1998;98:628-633.

(11.) Strachan DP, Carrington D, Mendall MA, et al, Relation of Chlamydia pneumoniae serology to mortality end incidence of Ischemic heart disease over 13 years in the Caerphilly prospective heart disease study. BMJ. 1999;318:1035-1040.

(12.) Helsted C, Lepow ML, Balassanlan N, et al. Otitis media. Clinical observations, microbiology, and evaluation of therapy. Am J Dis Child. 1968;115:542-551.

(13.) Howie VM, Ploussard JH. Efficacy of fixed combination antibiotics versus separate components in otitis media Effectiveness of erythromycin estolate, triple sulfonamide, ampicillin, erythromycin estolate-triple sulfonamide, and placebo in 280 patients with acute otitis media under two and one-half years of age. Clin Pediatr (Phila), 1972;11:205-214.

(14.) van Buchem FL, Dunk UH, van’t Hof MA. Therapy of acute otitis media: myringtomy, antibiotics, or neither? A double-blind study in children. Lancet. 1981;2:883-887.

(15.) Mygind N, Meistrup-Larsen Kl, Thomsen J, et al. Penicillin in acute otitis media: a double-blind placebo-controlled trial. Clin Otolaryngol. 1981;6:5-13.

(16.) Kaleida PH, Casselbrant ML, Rockette HE, et al. Amoxicillin or myringotomy or both for acute otitis media: results of a randomized clinical trial. pediatrics. 1991;87:466-474.

(17.) Burke P, Bain J, Robinson D, et al. Acute red ear in children: controlled trial of non-antibiotic treatment in general practice. BMJ. 1991;303:558-562.

(18.) Appelman CL, Claessen JQ, Touw-Otten FW, et al. Co-amoxiclav in recurrent acute otitis media: placebo controlled study. BMJ. 1991;303:1450-1452.

(19.) Sensakovic JW. Tuberculosis: guidelines for effective prophylaxis and therapy. Consultant. 1997;37:173-180.

(20.) Boyars MC, Mercado AC. Community-acquired pneumonia: update on antibiotic therapy. Consultant. 1998;38:1659-1676.

Major developments and future trends

Major developments in the past 10 years

Awareness of antimicrobial resistance has increased

The resurgence of tuberculosis has abated somewhat but is still a problem in many areas

Directly observed therapy has proved to be an effective intervention

Molecular biologic techniques have greatly improved the diagnosis of tuberculosis

Atypical mycobacterial disease has become increasingly recognized in non-HIV-infected persons, particularly those with COPD

The efficacy of therapies for atypical mycobacterial disease has improved New antimicrobials have been developed new fluoroquinolones and new macrolides, and new classes, such as streptogramins, oxazolidinones, everninomicins, ketalides)

What to expect in the next 5 to 10 years

More judicious use of antibiotics

New rapid diagnostic tests

Immune enhancement therapy for serious respiratory disease

New antimicrobials

COPD, chronic obstructive pulmonary disease,

The debate over otitis media

Dr File: Improving the differentiation between acute otitis media and otitis media with effusion could help reduce the unnecessary use of antibiotics in children.

Dr Squillace: Serous otitis with effusion produces an immobile eardrum. This is commonly used as the hallmark of an ear infection, although a blocked eustachian tube with clear fluid that is not infected also produces an immobile eardrum. An important question is: Do we really need to treat otitis media? Seven studies have compared antibiotic treatment with other therapies, such as antipyretics and decongestants.[12-18] Although the outcome after the first few days is slightly better with antibiotics, the longterm outcomes in terms of resolution of disease, recurrence, and complications were no different.

If the child is older than 2 months and has not had recurring episodes, we try to explain to parents that antibiotics may not be necessary. It is not easy, but we give them a list of things that they can do while waiting the couple of days it takes for otitis media to resolve on its own.

Dr File: Garegivers sometimes rationalize their use of antibiotics by saying they are trying to prevent serious sequelae. So they might give antibiotics to many children with otitis just to prevent the rare case of meningitis. But this approach does not work; all the studies have shown that treating viral infections with antibiotics early does not alter the subsequent course. It does not prevent the progression to bacterial infection.

Dr Matthay: When do you use antibiotics for otitis?

Dr Squillace: No hard data support the widespread use of antibiotics. Theoretically, antibiotics might be indicated for children at higher risk for complications–a child with recurring infections or who already has hearing loss. But this has not been proved. Moreover, the duration of therapy has not been defined. The 10-day recommendation is based on extrapolation from data about pharyngitis. We recommend 5 rather than 10 days of treatment.

COPYRIGHT 1999 Cliggott Publishing Co.

COPYRIGHT 2004 Gale Group