Donald Armstrong

Cryptococcosis in a cancer hospital. Clinical and pathological correlates in forty‐six patients

The clinical and pathological findings in 46 patients with cryptococcosis at Memorial Sloan‐Kettering Cancer Center from 1956 to 1972 are reported. The striking predilection for cryptococcal infection in patients with leukemias and lymphomas is again confirmed. Of 41 patients with neoplastic disease, those with chronic lymphatic leukemia (CLL), Hodgkins Disease, chronic myelogenous leukemia (CML), myeloma and lymphosarcoma had the highest incidence of cryptococcosis. In all cases, neoplastic disease was widespread when infection occurred. All of these patients had leukopenia and absolute lymphopenia at the time of infection. Thirty‐nine were on steroids. Thirty‐one patients with neoplastic disease had disseminated infection. Review of pathology revealed a spectrum of inflammatory lesions. Histiocytic‐lymphocytic infiltrates occurred in the central nervous system in 10 patients. In six cases, reaction was granulomatous. There were single instances of suppurative and fibrotic reactions. Mortality from infection was high in patients with neoplastic disease. Twenty‐four of 28 deaths occurred within 60 days as a result of infection. Within one year, 10 more patients died, nine of cryptococcosis. Only three survived more than one year, and all patients died within 600 days. Twenty‐nine patients with neoplastic disease received amphotericin B. Only nine survived more than 60 days.

INFECTIOUS MORBIDITY IN CRITICALLY ILL PATIENTS WITH CANCER

Infection frequently complicates the course of cancer treatment and often adversely affects the outcome. Patients have a greater tendency for acquiring infections caused by opportunistic microorganisms. Agents with low virulence potential may lead to invasive and often life-threatening infections because of altered host immune function. The immune dysfunction may be caused by the underlying malignancy, by antineoplastic chemotherapy, or by invasive procedures during supportive care.

Catheter-related Malassezia furfur fungemia in immunocompromised patients

PURPOSE, PATIENTS, AND METHODS Malassezia furfur has usually been described as a cause of catheter-related sepsis in neonates receiving intravenous lipid emulsion. We report seven cases of catheter-related M. furfur fungemia that occurred in seven immunocompromised patients including four adults and three children who were not neonates. Only two of these patients were receiving concurrent intravenous lipid emulsion. RESULTS All positive blood cultures were obtained from a central venous access device, one of which was a port device. Quantitative M. furfur colony counts ranged from 50 cfu/mL to greater than 1,000 cfu/mL. All seven patients were treated with amphotericin B. Blood drawn through the central lines of three patients yielded additional organisms. One central venous access device required removal due to persistently positive M. furfur blood cultures despite treatment with amphotericin B. CONCLUSION We conclude that catheter-related M. furfur fungemia occurs in immunocompromised patients with central venous access devices whether or not they are receiving intravenous lipids. Prompt, aggressive treatment with amphotericin B (1 mg/kg/d) may spare patients removal of their central venous access device. Further studies are needed to determine the role of endogenous lipids in the development of catheter-related M. furfur fungemia and to determine if there is a seasonal incidence in populations other than neonates, since all of our cases occurred between late March and July.

Listeriosis in Patients at a Comprehensive Cancer Center, 1955–1997

Listeria monocytogenes infection occurred in 94 patients during 1955-1997 at Memorial Sloan-Kettering Cancer Center. The incidence was 0.5 (1955-1966), 0.96 (1970-1979), and 0.14 (1985-1997) cases per 1000 new admissions. Eighty-five patients (90%) were bacteremic, and 34 (36%) had evidence of intracranial infection. In 91 patients with cancer, 70 (77%) received chemotherapy for advanced or relapsed malignancy (n=51; 56%); 64 (68%) received corticosteroids. Breast cancer was the most common solid-organ cancer (n=14; 45%), and 34 (36%) had preexisting advanced liver disease. In 14 (39%) of 37 patients who died of listeriosis, death occurred within 48 h of L. monocytogenes isolation. Four (80%) of 5 patients with extracranial foci of infection died of their infection, compared with 33 (37%) of 89 patients with isolated bacteremia and/or intracranial infection (odds ratio, 2.34; P=.05). Most infections (60%) were due to L. monocytogenes serotype 1/2, and the remainder (40%) were due to serovar 4b. Listeriosis in these patients with cancer occurred most often in individuals receiving antineoplastic therapy for advanced or relapsed malignancy and systemic corticosteroids. The presence of advanced liver disease may have increased the risk of systemic listeriosis in susceptible patients with underlying cancer.

Prospective epidemiologic analysis of triazole-resistant nosocomial Candida glabrata isolated from patients at a comprehensive cancer center

OBJECTIVE The emergence of Candida glabrata infections among patients with compromised immunity has become a serious concern, especially at centers caring for individuals with cancer. METHODS During a prospective evaluation of Candida species associated with either clinically significant colonization or infection, 26.9% of C. glabrata isolates showed in vitro resistance to fluconazole (MIC of > or = 64 microg/ml). RESULTS Antifungal susceptibility profiles and genetic fingerprinting analysis performed by randomly amplified polymorphic DNA (RAPD) techniques confirmed low-probability of phenotypic and genotypic relatedness among nosocomial C. glabrata isolates. CONCLUSIONS Presence of polyclonal strains of C. glabrata in patients at our hospital was probably related to selection of resistant yeasts from environmental pool rather than monoclonal expansion or clustering of multi-drug resistant C. glabrata in high-risk patients.

Prolonged Candidemia in Patients with Cancer

NOTE. AmB, amphotericin B; CVC, central venous catheter; Flu, fluconazole; Itr, itraconazole; Ket, ketoconazole. a Day that the first blood culture positive for Candida was performed. evaluation of CSF samples are inaccurate. Laboratory evaluation of the initial CSF sample revealed no eosinophils (not 3% eosinophils, as reported), and the diagnosis was made clinically, despite this finding. The second lumbar puncture showed 6% eosinophils (not 16% eosinophils, as reported) and was performed both in expectation of eosinophilia developing and in an attempt to help relieve the patient’s headache. Because a large number of people who arrive in Auckland are either migrants from the Pacific Islands or travelers returning from holidays spent in the Pacific Islands, where the disease is endemic, we are very familiar with eosinophilic meningitis due to A. cantonensis, and we see several cases per year. Indeed, 1 of the series of cases mentioned by Lo Re and Gluckman was evaluated and reported by our neurological colleagues at Auckland Hospital [2]. We do not, in general, find diagnosis of eosinophilic meningitis due to A. cantonensis to be “a major challenge,” because the condition has a characteristic clinical symptomatology with an appropriate epidemiological history (even without the presence of CSF or peripheral blood eosinophilia, as initially was the case for the woman described by Lo Re and Gluckman [1]) and is familiar to those of us in the fields of infectious diseases and neurology in this part of the world. Having said all that, we acknowledge that the ultimate proof of diagnosis was achieved by the Bangkok group mentioned in the article by Lo Re and Gluckman [1].

Central Nervous System Infections in the Compromised Host

Meningitis and other types of central nervous system (CNS) infection in the immunocompromised patient are usually caused by different organisms than in the general population.1–3 For instance, Streptococcus pneumoniae or Neisseria meningitidis may produce meningitis in an immunocompromised patient, but it is more common to find Listeria monocytogenes or Pseudomonas aeruginosa as the cause. The latter organisms may cause meningitis in the general population, but much less commonly as compared with the Pneumococcus or the Meningococcus.4–6 Similarly, the most common type of brain abscess seen in the general population is caused by mixed aerobic and anaerobic bacterial flora7–8 by a process that generally extends from the nasopharynx. In contrast, Aspergillus fumigatus 9–10 or Nocardia asteroides 11, 12 more frequently cause brain abscess in the immunocompromised patient,1,2 and the apparent route of access is hematogenous.

HEALTH EDUCATION: AN APPRAISAL

Perhaps the time has come, as is apt to be the case periodically in most fields, for a fresh orientation as to what in general we mean by health education. There are still those in public health who see little or no value or scope in health education. There are also those, in increasing numbers, who appear to think that health education and public health are practically synonymous terms—coextensive in range and content. Between these two extreme views, what is health educations real status? Parenthetically, let me make it clear at the start that I am attempting to appraise the relative importance of health education in the public health field as compared with other elements. I am not attempting an analysis from a psychologic or other angle, of what constitutes effective health education or what technics will or will not arouse real interest and action. This is another story, though

NON-BACTERIAL INFECTIONS ASSOCIATED WITH NEOPLASTIC DISEASE

There are a large number of non-bacterial microorganisms waiting to infect the host with neoplastic disease. The basic disease may make the individual more susceptible, but more often therapeutic measures are responsible. Although a specific neoplastic disease may predominantly alter one immunological function,anti-neoplastic chemotherapy regimens usually alter to some degree, the others. In the appropriate setting, the clinical picture due to a specific microorganism is frequently recognizable, or at least limited to a selected group of organisms. Diagnostic measures should be prompt and specific,-and treatment instituted early, for these infections are frequently life-threatening in the immunosuppressed host. Candida species have been the most common non-bacterial organisms found at post-mortem at Memorial Sloan-Kettering Cancer Center, followed byAspergillus and Mucor species. Pneumocystis carinii infections were seen frequently during the time period observed in the study. Life-threatening infections with other non-bacterial agents were less frequent, but occurred, and frequently more than one microorganism was associated with the death of the patient.