Update in Infectious Diseases

Common Outpatient Infections

Upper respiratory tract symptoms take a large toll in terms of both morbidity and mortality. The average adult, for example, has two to three colds per year. The rate of influenza in the 1996-1997 season exceeded the normal rate, and the disease is still among the most common causes of death in elderly persons in the United States.

Zinc Seemed To Help with the Common Cold

Mossad SB, Macknin ML, Medendorp SV, Mason P. Zinc gluconate lozenges for treating the common cold. A randomized, double-blind, placebo-controlled study. Ann Intern Med. 1996; 125:81-8.

Zinc gluconate lozenges for the common cold have received much public attention. Although possibly apocryphal, the following anecdote has supported the contention that the lozenges are effective: The father of a young man receiving chemotherapy for cancer observed that his son, who was also receiving zinc supplements, had no colds during the 2-year treatment course. Several observational studies have suggested a similar conclusion, prompting Mossad and colleagues to test zinc in a clinical trial.

One hundred patients with predefined symptoms of the common cold were randomly assigned to use zinc lozenges (containing 13.3 mg of zinc) or placebo every 2 hours while awake, starting within 24 hours of symptom onset. The 50 patients who received zinc had a shorter duration of symptoms (4.4 days compared with 7.6 days in the placebo group) and fewer days with cough (2.0 compared with 4.5 days), nasal congestion (4.0 compared with 6.0 days), nasal drainage (4.0 compared with 7.0 days), sore throat (1.0 compared with 3.0 days), and headache (2.0 compared with 3.0 days). Zinc recipients experienced more side effects, especially bad taste (80% compared with 30%) and nausea (20% compared with 4%).

This is one of the few studies that has shown a successful intervention for the common cold. The explanation for the outcome is still not clear but could involve an antiviral effect of zinc. Still, problems with the study exist. First, no in vitro evidence suggests that zinc has antiviral activity. Second, the anatomy makes little sense: Most cold symptoms are nasal and the virus is transmitted nasally, but the lozenges are given orally. Third, because more participants in the zinc group had adverse effects, the trial was not truly blinded. However, similar proportions of participants in both groups correctly guessed treatment assignment. Fourth, the adverse effects of the zinc lozenges may be worse than the symptoms of the cold itself. Finally, other controlled trials have shown no benefit with zinc lozenges [1]. The clinical conclusion can be summed up by the following statement from the editors of the Medical Letter [2]: “Zinc might decrease the symptoms and shorten the duration of the common cold, but properly blinded confirmatory studies are needed.”

Immunizing Staff Protected Patients from Influenza

Potter J, Stott DJ, Roberts MA, Elder AG, O'Donnell B, Knight PV, et al. Influenza vaccination of health care workers in long-term-care hospitals reduces the mortality of elderly patients. J Infect Dis. 1997; 175:1-6.

The Centers for Disease Control and Prevention report that rates of influenza immunization are disappointing and that the track record for pneumococcus immunization is terrible [3]. Beyond these findings, a vexing question remains: Who should be immunized? Influenza and pneumonia remain major causes of death in the United States. In 10 of the past 20 years, the annual death toll for influenza and its complications has exceeded 20 000. Most patients at risk are older than 65 years of age, and many are nursing home residents. Potter and colleagues studied whether vaccinating health care workers in long-term care facilities would decrease the incidence of disease among patients in the facilities.

The study involved 653 staff members and 1059 patients in 12 geriatric hospitals in Scotland. The participants were divided into four groups: staff members and patients who were vaccinated, staff members who were vaccinated and patients who were not vaccinated, staff members who were not vaccinated and patients who were vaccinated, and staff members and patients who were not vaccinated. Results (Table 2) showed that mortality rates among the patients were substantially reduced when the staff members were vaccinated but that vaccinating only the patients had little effect on mortality.

Amoxicillin Had Minimal Effect against Sinusitis

van Buchem FL, Knottnerus JA, Schrijnemaekers VJ, Peeters MF. Primary-care-based randomised placebo-controlled trial of antibiotic treatment in acute maxillary sinusitis. Lancet. 1997; 349:683-7.

Sinusitis is commonly seen in primary care practice, and one might assume that choosing an antibiotic might be a fairly straightforward decision. Bacteria are found in about 60% of sinus aspirates taken from adults with symptoms of sinusitis. Streptococcus pneumoniae and Haemophilus influenzae are the most common pathogens [4]. Despite the prevalence of sinusitis, it is remarkable how few data on the efficacy of antibiotic treatment have been collected. This study of treatment was perhaps more informative about what it did not show.

During a 1-year period, 53 general practitioners in England suspected sinusitis in 488 patients. These patients were referred for radiography; 214 had positive results and were randomly assigned to either placebo or amoxicillin, 750 mg three times daily. In this area of England, in contrast to much of the United States, 95% of H. influenzae strains are sensitive to amoxicillin and there is little problem with penicillin-resistant pneumococci. Outcome measures included subjective sickness scores, meatal secretion at 1 week, and radiographs at 2 weeks.

Few differences were seen between the groups. At 2 weeks, 77% of the controls and 83% of patients given amoxicillin reported almost complete or complete resolution of symptoms. Both radiographic improvement and adverse effects were more common among patients receiving amoxicillin than among controls (74% compared with 60% and 28% compared with 9%, respectively).

So what should the clinician do? The answer is far from clear. Antibiotics for sinusitis may not be effective, even though most patients have bacteria in their sinuses. Although more than 20 drugs are used to treat sinusitis, none has proven superior to amoxicillin in clinical trials. It now seems that even amoxicillin shows minimal efficacy. In my opinion, most patients with sinusitis derive minimal benefit from antibiotics. A minority of patients, however, clearly derive much benefit, and the challenge is to determine which clinical or laboratory features identify this subset. In the meantime, amoxicillin remains a good choice for initial treatment of uncomplicated disease.

Pneumococcus Was the Number One Killer in Community-Acquired Pneumonia

Fine MJ, Smith MA, Carson CA, Mutha SS, Sankey SS, Weissfeld LA, et al. Prognosis and outcomes of patients with community-acquired pneumonia. A meta-analysis. JAMA. 1996; 275:134-41.

Many groups have created clinical practice guidelines for managing community-acquired pneumonia. However, the guidelines differ as to which pathogen is most important to treat and how forceful treatment should be. Fine and colleagues sought to clarify the causes and clinical course of community-acquired pneumonia in recent years.

These researchers are part of the Patient Outcomes Research Team, which is comprehensively investigating community-acquired pneumonia. They performed a meta-analysis to determine the prognosis and outcomes of patients with community-acquired pneumonia by systematic review of 122 English-language articles between 1966 and 1995. They identified 127 cohorts that comprised 33 148 patients. The overall mortality rate was 13.7%.

The most provocative observation was the bacteriology of the pneumonia. An infectious agent was found in only about 7000 patients (38.6%). Table 3 lists the agents found to be responsible in those patients; S. pneumoniae accounted for two thirds of cases of pneumonia. These data are provocative in the face of sequential worldwide surveillance studies that show a downward trend in rates of recovering S. pneumoniae as a cause of community-acquired pneumonia. This trend probably reflects deterioration in laboratory detection and the recognition of new agents, such as Chlamydia pneumoniae.

One Fourth of Pneumococcal Isolates in the United States Show Reduced Sensitivity to Penicillins

Doern GV, Brueggemann A, Holley HP Jr, Rauch AM. Antimicrobial resistance of Streptococcus pneumoniae recovered from outpatients in the United States during the winter months of 1994 to 195: results of a 30-center national surveillance study. Antimicrob Agents Chemother. 1996; 40:1208-13.

Pneumococcus is the most common cause not only of death in patients with community-acquired pneumonia but also of acute sinusitis. It is also one of the most common bacterial pathogens in pyogenic meningitis and is second only to H. influenzae in causing acute otitis media and exacerbations of chronic bronchitis. Because two thirds of antibiotics prescribed in the United States are for respiratory tract infections, S. pneumoniae is an important target. The rise of the pneumococcal resistance to penicillin and other antibiotics makes for a worrisome situation.

Doern and colleagues studied the extent of pneumococcal resistance to antibiotics in a national sampling of patients who were not hospitalized. Clinical isolates of S. pneumoniae were collected from 30 U.S. medical centers between November 1994 and April 1995 and were tested for in vitro susceptibility to multiple antibiotics.

A total of 1527 clinical isolates were tested. The frequency of reduced sensitivity to penicillin was 23.6%: intermediate resistance in 14.1% of isolates and high-level resistance in 9.5%. Resistance to other antibiotics was as follows: cefotaxime, 3%; ceftriaxone, 5%; cefuroxime, 12%; macrolides (erythromycin, clarithromycin, and azithromycin), 10%; chloramphenicol, 4.3%; tetracycline, 7.5%; and trimethoprim-sulfamethoxazole, 18%. All strains were sensitive to vancomycin. The most active oral cephalosporins were cefpodoxime, cefuroxime, and cefprozil.

Little doubt remains that resistance of S. pneumoniae to antibiotics is increasing and is widely distributed throughout the United States. The newer fluoroquinolones (levofloxacin and sparfloxacin) were not tested in this study, but these drugs do have high activity against most strains of pneumococcus in vitro, including 99% of penicillin-resistant strains. Finally, and perhaps most significantly, more emphasis is likely to be placed on vaccination against S. pneumoniae because the serotypes of resistant strains are well represented in the vaccine.

The future of antibiotic treatment for respiratory infections has become blurred by the increasing in vitro resistance to the time-honored drug of choice and the escalating concern for antibiotic abuse. In the Netherlands, for example, current guidelines call for withholding antibiotics from children with acute otitis media, a condition that accounts for 50% of antibiotic prescriptions for children in the United States. Among adults, the most controversial areas are bronchitis, exacerbations of chronic bronchitis, and sinusitis (discussed above). With regard to treatment of S. pneumoniae (and H. influenzae), a surprising development is the emergence of the fluoroquinolones as the possibly preferred treatment. A report published last year showed that ofloxacin was active for 99% of strains, including penicillin-resistant strains [5]. New agents with similar activity are levofloxacin and sparfloxacin. However, the retail cost for a course of these newer agents runs about $60 to $70, compared with about $10 for amoxicillin or doxycycline. In an era of cost concerns, the decisions therefore remain difficult for conditions in which superior outcome has not been shown. My recommendations are listed in Table 4.

Infections and the Public Health

The ecology of food production in the modern world may be opening new pathways for infectious agents to create new public health crises. For example, it is said that one McDonald's hamburger contains meat from as many as a thousand different cows. The opportunities for large outbreaks of food-borne infections are obvious. In addition, the epidemiology of sexually transmitted diseases is changing, and new screening philosophies may be needed.

Salmonella Was Traced to Ice Cream Mix

Hennessy TW, Hedberg CW, Slutsker L, White KE, Besser-Wiek JM, Moen ME, et al. A national outbreak of Salmonella enteritidis infections from ice cream. N Engl J Med. 1996; 334:1281-6

One of the two major outbreaks reported in the United States in 1996 was the infection of about 224 000 persons with food-borne Salmonella enteritidis. This organism had been known to be a common pathogen in homemade ice cream but had never been reported as part of a pasteurized ice cream-making process.

In late 1994, the Minnesota Department of Health began receiving reports of isolates of Salmonella enteritidis taken from patients with gastrointestinal symptoms in southeastern Minnesota. Of the first 15 patients who were matched with controls, 73% had eaten Schwann's ice cream, compared with 13% of controls. Hennessy and colleagues set out to identify if and how the pathogen entered the ice cream production system.

The overall attack rate among persons who consumed Schwann's ice cream in late 1994 was 6.6%. The case rate was highest among persons who had eaten ice cream from a tank truck that had carried nonpasteurized eggs immediately before carrying the dairy products. The ice cream plant itself was found not to harbor Salmonella enteritidis. This outbreak underscores how one small mistake in the food production pathway can spread disease throughout an entire nation.

Another example of widespread food-borne infection was the outbreak of Cyclospora organisms in mid-1996. This outbreak affected 1450 persons in 20 states and Canada and was curiously associated with adults rather than children. The attack rate was 56% [6, 7].

This outbreak of cyclosporal infection represents overlap between public health and industry, and it underscores the importance of certainty about sources of infection. When the outbreak had affected only about 10 persons who had eaten mixed berries, investigators first thought that the source of infection was strawberries from California. They soon found that the source might be raspberries from Guatemala; in the short time between the formulation of these hypotheses, however, the initial hypothesis became public. As a result, the California strawberry industry suffered a huge depression.

Virus-Associated Diseases

Research on two viruses received much attention in 1996. In one study, a herpesvirus that seems to cause Kaposi sarcoma was grown and used to reinfect healthy cells. Another study found that cytomegalovirus may play an important role in coronary artery disease.

Human Herpesvirus from Kaposi Sarcoma Was Grown

Foreman KE, Friborg J Jr, Kong WP, Woffendin C, Polverini PJ, Nickoloff BJ, et al. Propagation of a human herpesvirus from AIDS-associated Kaposi's sarcoma. N Engl J Med. 1997; 336:163-71.

Last year's Update in Infectious Diseases [8] discussed the isolation of human herpesvirus 8 and the association of this virus with Kaposi sarcoma. This virus has now been found in most forms of Kaposi sarcoma and has been implicated in body-cavity lymphomas.

About 1% of all blood donors have serologic evidence of human herpesvirus 8. In sexually transmitted disease clinics, the prevalence of the virus is about 10%; among HIV-infected patients, prevalence varies: 4% in women, 3% in hemophiliacs, 30% in homosexual men, and almost 100% in patients with Kaposi sarcoma. Semen and saliva have been found to harbor the virus and thus are the proposed vehicles of transmission.

Foreman and colleagues' study ends the speculation about the association between human herpesvirus 8 and Kaposi sarcoma. The researchers isolated virus from patients with Kaposi sarcoma and found that the DNA sequences were identical to previously described sequences in human herpesvirus 8. Foreman and colleagues then cultured these specimens with human epithelial cells. The virus was cytotoxic to cells and was recovered from infected cells. Again, the recovered DNA sequences were identical to those of human herpesvirus 8.

These findings do not really present any surprises. Since the era of Koch, however, scientists have been more comfortable about identifying a “cause” of disease when an infectious agent could be isolated, produce disease, and then be isolated from cells that were diseased.

Cytomegalovirus Was Associated with Coronary Stenosis

Zhou YF, Leon MB, Waclawiw MA, Popma JJ, Yu ZX, Finkel T, et al. Association between prior cytomegalovirus infection and the risk of restenosis after coronary atherectomy. N Engl J Med. 1996; 335:624-30.

In 1996, infectious diseases entered the field of cardiology as studies showed an association between cytomegalovirus infection and the risk for restenosis after coronary artery atherectomy and angioplasty. Cytomegalovirus DNA has been found in lesions with restenosis, but the question has remained: Is a patient who harbors cytomegalovirus at greater risk for restenosis after coronary atherectomy?

Zhou and colleagues studied 75 patients who had directional coronary atherectomy for symptomatic coronary artery disease. Blood levels of anticytomegalovirus IgG antibodies were obtained before the procedures, and coronary angiography was performed 6 months later to assess restenosis.

After atherectomy, the mean minimal luminal diameter of the target vessel was greater in the 49 patients who were seropositive for cytomegalovirus than in the 26 patients who were seronegative. After 6 months, however, the seropositive patients' arteries had more extensive restenosis (43% stenosis compared with 8%; P = 0.002). In a multivariable logistic regression model, seropositivity for cytomegalovirus and the cytomegalovirus antibody titer were independently predictive of restenosis (odds ratios, 12.9 and 8.1, respectively). No evidence showed acute infection. Previous cytomegalovirus infection is a strong independent risk factor for restenosis after coronary atherectomy. If confirmed, these findings may help identify patients at risk for restenosis.

Yet another agent is associated with coronary artery disease: Chlamydia pneumoniae[9]. Serologic studies have shown a strong association between this organism and coronary artery stenosis. The problem is that serologic testing for C. pneumoniae is not particularly good. In several studies, however, C. pneumoniae was subsequently found in atherosclerotic plaques by polymerase chain reaction or electron microscopy in 86% of patients (who had coronary disease) compared with 0% of controls (who did not have coronary artery disease).

Some skeptics believe that C. pneumoniae infection simply represents a secondary event that reflects inflammation. Others claim that this organism has a central etiologic role, at least in some patients. The issue may be resolved with therapeutic trials using antibiotics.

HIV Disease

At its peak in the 14th century, bubonic plague killed 25 million persons. Influenza A killed 20 million persons in the early 20th century. It now seems that HIV has infected at least 28 million persons and by the end of the century will become the greatest lethal epidemic in the history of the civilized world. Table 5 lists the estimates for mechanisms of transmission in the United States.

Thai Politics Decreased the Rate of HIV Infection

Nelson KE, Celentano DD, Eiumtrakol S, Hoover DR, Beyrer C, Suprasert S, et al. Changes in sexual behavior and a decline in HIV infection among young men in Thailand. N Engl J Med. 1996; 335:297-303.

The good news is that in most of the developed world, incidence curves for HIV infection have leveled off. One area that still shows a rapid increase, however, is Southeast Asia. On the basis of current projections, Southeast Asia will have more new cases of HIV infection in 1997 than the rest of the world combined. The seroprevalence of HIV infection in injection drug users in Thailand increased from 1% to 30% in 1 year. In commercial sex workers, the seroprevalence rate in 1993 was 29%.

In 1991, Arman Penyarchen was chosen by the military to serve as prime minister of Thailand. He declared AIDS to be a major disaster in Thailand and increased the AIDS budget from $2.6 million to $44 million. He started a mass media campaign called the “100% condom campaign” in 1991, and he took control of the commercial sex industry.

Nelson and colleagues describe the results of these political decisions. They interviewed five cohorts of 21-year-old men from northern Thailand who were conscripted into the army by lottery in 1991, 1993, and 1995. The 4311 men were also tested for HIV antibody.

In the 1991 and 1993 cohorts, the prevalence of HIV infection was 10% and 12.5%, respectively. In the 1995 cohort, prevalence decreased to 6.7%. Seroprevalence was only 0.7% among men who had had no sexual contacts with sex workers before 1992. The proportion of men who reported having sexual relations with sex workers decreased from 81% to 64%. From 1991 to 1995, the reported use of condoms during the most recent sexual contacts with sex workers increased from 61% to 93%. In 1995, 15% of men had a history of a sexually transmitted disease, compared with 42% in 1991.

The authors concluded that

“The unusual circumstance of having the pragmatic nonpartisan technocrats and non-government officials and academic activists in positions of high government cabinet power with no opposition shows the potential when government and non-governmental forces merge.”

North Carolina Reduced the Rate of Perinatal HIV Transmission

Fiscus SA, Adimora AA, Schoenbach VJ, Lim W, McKinney R, Rupar D, et al. Perinatal HIV infection and the effect of zidovudine therapy on transmission in rural and urban counties. JAMA. 1996; 275:1483-8.

Unfortunately, the broad political climate in the United States has not been as conducive to HIV prevention as that in Thailand. Some areas, however, have made significant progress in preventing new cases of HIV disease in specific groups.

The results of AIDS Clinical Trials Group trial 076, published in 1994, showed that zidovudine prevented vertical transmission [10]. Soon after, the U.S. Public Health Service published guidelines calling for the use of zidovudine in pregnant women [11].

Fiscus and colleagues sought to determine the impact of zidovudine on vertical transmission of HIV in North Carolina. They used a statewide survey to determine the number of newborns whose mothers were HIV positive and to determine whether these infants became infected at birth. The survey was conducted in 1993 and 1994.

In that single year, transmission rates decreased from 21% to 8.5%. This impressive result showed how quickly a scientific discovery can be put into practice and avert disease. Since this report was published, a study that used a zidovudine dose 12 to 15 times the dose usually used in humans prompted some concern about carcinogenesis among offspring. That concern has been somewhat quelled by a National Cancer Institute panel that reviewed the data and found that, overall, the benefit far out-weighed the risk. However, the panel also recommended that women should be informed about a potential risk.

The use of zidovudine to prevent perinatal transmission of HIV is antiquated. Most physicians who treat HIV-infected pregnant women believe that the mother should be the primary recipient of therapy. In addition, zidovudine monotherapy has been supplanted as primary therapy for HIV-infected persons by multiple antiretroviral drugs.

Levels of HIV RNA Became the Primary Prognostic Marker

Mellors JW, Rinaldo CR Jr, Gupta P, White RM, Todd JA, Kingsley LA. Prognosis in HIV-1 infection predicted by the quantity of virus in plasma. Science. 1996; 272:1167-70.

The natural history of clinical HIV progression is well known. Once infected, the average patient will live about 10 years before sustaining an AIDS-defining event. Thereafter, the patient's lifespan averages about 1 year if no therapy is directed against HIV. Until recently, prognosis was determined by measuring CD4+ lymphocyte counts, which inversely correlate with disease severity.

Mellors and colleagues' study was designed to assess the ability of a baseline HIV RNA measurement to predict the risk for AIDS and death. Levels of HIV RNA were measured in plasma samples obtained from 180 men at entry into the Multi-center AIDS Cohort Study. Follow-up information was available for all patients.

The risk for developing AIDS and dying was directly related to plasma viral load at study entry. Levels of HIV RNA were a better predictor than were CD4+ lymphocyte counts (P < 0.001). Quartiles of HIV RNA provided excellent discrimination for estimated median survival: Levels less than 5000 copies/mL suggested survival longer than 10 years; levels of 5000 to 13 000 copies/mL, 9.5 years; levels of 13 000 to 36 000 copies/mL, 7.4 years; and levels more than 36 000 copies/mL, 5.1 years.

The current understanding of the course of infection is as follows: A patient becomes infected and suffers a sharp burst of viremia, followed by an aggressive immune response; after a tumultuous period of about 6 months, the viral load settles into a “set point,” which may be high or low. High loads portend a poor prognosis, and low loads show a slower rate of progression.

Combination Therapy Became Standard

Hammer SM, Katzenstein DA, Hughes MD, Gundacker H, Schooley RT, Haubrich RH, et al. A trial comparing nucleoside monotherapy with combination therapy in HIV-infected adults with CD4 cell counts from 200 to 500 per cubic millimeter. N Engl J Med. 1996; 335:1081-90.

It is increasingly clear that zidovudine monotherapy provides short-lived clinical benefit. The drug depresses viral load for only about 6 to 12 months. Hammer and colleagues compared various antiviral combinations in asymptomatic persons with CD4+ lymphocyte counts of 200 to 500 cells/mm³.

In a multicenter clinical trial, 2467 patients (mean CD4+ lymphocyte count, 352 cells/mm³) were randomly assigned to receive zidovudine (600 mg/d), zidovudine plus didanosine (400 mg/d), zidovudine plus zalcitabine (2.25 mg/d), or didanosine alone. The patients were followed for 2.75 years.

Progression to a primary end point (loss of ≥50 CD4+ lymphocytes, AIDS, or death) occurred more often with zidovudine alone (32% of patients receiving this regimen) than with all other regimens (range, 18% to 22%) (P < 0.001). The overall rate of premature discontinuation of treatment was 53%. Major reasons for discontinuation included drug toxicity, patient request for other treatment, and the number of pills required.

This study helped clarify the importance of combination therapy and the role of viral burden in predicting clinical outcome. The regimens used, however, have been antiquated by the advent of protease inhibitors.

Guidelines for HIV Were Published

Carpenter CC, Fischl MA, Hammer SM, Hirsch MS, Jacobsen DM, Katzenstein DA, et al. Antiretroviral therapy for HIV infection in 1996. Recommendations of an international panel. International AIDS Society-USA. JAMA. 1996; 276:146-54.

A group of 13 expert clinicians formulated answers to the four standard questions about antiviral therapy for any patient:

1. When to start therapy? Treatment should begin when the viral load is greater than 30 000 (possibly 50 000) copies/mL or when the CD4+ lymphocyte count is low, perhaps 350 cells/mm³.

2. What therapy to start with? Initial therapy should be a combination of nucleoside analogues with or without a protease inhibitor.

3. When to stop therapy or change agents? Drug therapy should stop or change when the viral burden decreases to within 0.3 logs of the baseline level.

4. What to change therapy to? If therapy must be changed because of treatment failure, at least two new drugs are necessary.

Protease Inhibitors Were Introduced

The preceding recommendations were based on the data reviewed through May 1996. By the summer of 1996, they were somewhat outdated. This testifies to the rapid development of new information and changes in management strategies for patients with HIV infection. The new test for monitoring therapy was quantitative measurement of plasma HIV RNA levels, the new drugs were protease inhibitors (saquinavir, ritonavir, indinavir, and nelfinavir) and non-nucleoside reverse transcriptase inhibitors (nevirapine and delvirdine), and the new goal of treatment was to suppress the virus to undetectable levels. The rationale of this goal is that complete viral suppression prevents disease progression and eliminates the mutations responsible for resistance, the presumed cause of drug failure. Thus, by early 1997, most authorities had adopted more aggressive treatment strategies that used multiple drugs for “triple therapy” (routine use of two nucleoside analogues plus a protease inhibitor or nevirapine). The advantage of these new regimens was shown in early clinical trials, which reported that a majority of previously untreated patients achieved undetectable virus levels and sustained them for at least 1 year. The disadvantages of these recommendations are that the drug regimens are complex (often involving ≥20 pills per day) and expensive ($10 000 to $12 000 per year) and adverse drug reactions and clinically significant drug interactions are common.

New guidelines based on these observations are now being developed by the International AIDS Society-USA and the U.S. Public Health Service. Guidelines from the Johns Hopkins HIV Care Program are summarized in Table 6.

Table 7 lists the 10 most important advances against HIV-related diseases in 1996. This year promises more advances, and we seem to be entering a new era in which HIV infection will be considered a chronic but manageable disease.

Dr. Roberts (Series Editor): Madrona Medical Group, 4370 Cordata Parkway, Bellingham, WA 98226-8075.