|
|
| |||||||||||||||||||||||||||||||||||||
About half the adult patients with listerial infection present with meningitis,1,2 presumably blood-borne. The majority (about 70 per cent) have an underlying immunosuppressive illness, primarily hematologic cancer, or are receiving immunosuppressive drugs, usually corticosteroids. An additional small proportion have alcoholism. Although L. monocytogenes accounts for only 1 to 2 per cent of all cases of bacterial meningitis in adults, it is the leading cause of bacterial meningitis in patients with cancer. There are no features of listerial meningitis that set it sharply apart from other pyogenic meningitides, but certain characteristics are suggestive1: movement disorders such as ataxias and tremors seem to be prominent; the course is sometimes surprisingly subacute; the cerebrospinal fluid may not show a predominance of neutrophils or a low glucose concentration; and Gram's stain of the cerebrospinal fluid is usually negative.
Parenchymal involvement of the brain is relatively rare with other organisms that are common causes of meningitis. By contrast, as many as 20 per cent of patients with listeriosis may have a cerebritis.2 The lesions range from macroscopic abscesses to a diffuse encephalitis. A fascinating variant, the "rhombencephalitis syndrome," which involves the pons and medulla, is the human counterpart of circling disease in sheep.
About 25 per cent of patients with listeriosis have a "primary bacteremic form" without evident foci of origin or metastasis.1 This form, too, occurs mainly in immunosuppressed patients but may also strike pregnant women. Other patients have localized infections such as endocarditis, endophthalmitis, hepatitis, and oculoglandular or cutaneous infection.
Perinatal listeriosis may take two forms. The "early type," which results from intrauterine infection, produces the devastating illness known as granulomatosis infantiseptica. The "late type," presumably acquired during or just after birth, presents as meningitis in the second or third week of life. The provocative suggestion that chronic infection of the genital tract may be an important cause of habitual abortion in human beings has not found much support in recent studies.
Two things stand out in this brief account of the clinical epidemiology and manifestations of listeriosis. One is its striking propensity to affect adults who are immunosuppressed or pregnant. The second is the marked tropism, presumably during bacteremia, for organs that are normally among the best shielded in the body — namely, the central nervous system and the placenta. I will return to these points shortly.
L. monocytogenes is widespread in nature. It is found in streams, sewage, silage, and soil, and it causes sporadic disease in many species of mammals and birds. However, the usual reservoir and mode of transmission to human beings have been difficult to determine. Most evidence, including the frequent occurrence of gastrointestinal symptoms at the onset of bacteremic illness,1 points to the intestine as the usual portal of entry. Although the majority of cases are sporadic, rare outbreaks have been attributed to the ingestion of contaminated coleslaw,3 raw milk, and as reported by Fleming et al. in this issue, pasteurized milk.4 Immunosuppressed patients are at special risk in these outbreaks, as they are when sporadic cases occur.4 The excretion of L. monocytogenes in cow's milk is well recognized. Cows with listeric mastitis may produce normal-appearing milk containing large numbers of bacteria. Some of the organisms may survive pasteurization and may then grow better than competing species at refrigerator temperatures — a phenomenon called "cold enrichment."4 Milk may be a particularly effective vehicle because it protects the bacteria from gastric acid.
Human beings may carry L. monocytogenes asymptomatically in the intestinal tract. Carrier rates are in the vicinity of 5 per cent in the normal population and higher in groups at special risk.2 However, person-to-person spread is probably rare except to the fetus in pregnancy.
L. monocytogenes has remarkably few weapons in its storehouse. It produces no known toxins, and its hemolysins are unlikely to be major virulence factors. A component of the cell envelope causes nonspecific activation of B cells but also immunosuppression5; neither effect is of clear-cut importance in the pathogenesis of infection. Looking elsewhere to find an explanation for the disease-producing potential of L. monocytogenes, one finds that, like the brucellae, salmonellae, and mycobacteria, it is a facultative intracellular parasite. Early in infection, L. monocytogenes, especially "smooth" strains, are taken up by macrophages or monocytes within which they survive and even multiply. Resistance to infection depends primarily on cell-mediated immunity, although humoral factors may have a subsidiary role.6 Once specific immunity is acquired, T lymphocytes are able to activate macrophages to kill the organisms that they once housed benignly. As a bonus, macrophages activated in this way show an enhanced ability to kill other species of intracellular parasites. Corticosteroids depress the levels of T lymphocytes and impair the activity of macrophages.1,2 Various cancers may have similar effects. Pregnancy is well recognized as causing depression of cell-mediated immunity.7 Thus, the propensity of listeriosis to infect patients who are receiving corticosteroids, have an underlying cancer, or are pregnant seems understandable. However, we have still not addressed the organ tropism of the bacterium.
Rácz and co-workers,8 studying experimental listerial enteritis in the guinea pig, drew attention to the "epithelial phase of infection." They found that bacteria were promptly taken up by the absorptive epithelial cells at the tips of the ileal villi, after which they seemed to work their way through to the villous stroma to be picked up by macrophages. Similar changes occur with L. monocytogenes in the epithelium of the urinary bladder and eye and are also seen with shigella and salmonella in the intestine8 and with gonococci in the fallopian tube. It is not clear whether, during their brief parasitism of the epithelial cell, L. monocytogenes organisms lie free in the cytoplasm or enfolded in host cell membranes, though the latter is suggested. Epithelial cells share other features with classic phagocytes. Like them, they show signs of "activation" during the uptake process although the bacteria seem to be unharmed.8 Moreover, like monocytes and macrophages, the epithelial cells of the eye and gingiva have been shown to produce interleukin-1.
If uptake by epithelial cells in the intestine and eye furnishes a route of entry for L. monocytogenes into the body, could similar uptake by structural cells in the placenta and central nervous system explain its organ tropism during bacteremia? The trophoblastic cells of the placenta have been thought to pinocytose maternal IgG for transport to the fetus. Perhaps they and some counterpart in the central nervous system carry out a comparable function for L. monocytogenes.
Finally, let us return to the issue of listeriosis and milk. A number of infections may be transmitted by cow's milk, but the pathogenesis is not always the same.9 First of all, organisms may be shed within the milk itself; this is known to occur with L. monocytogenes, brucellae, and Mycobacterium bovis and is thought to occur with salmonellae. Secondly, contamination may occur from lesions on the teats, usually acquired from the milker's hands; this may occur with streptococci, staphylococci, and corynebacteria. Thirdly, contamination may occur after the milk has left the cow; this is usually the case when salmonellosis is milk-borne. It is noteworthy that milk from cows and human beings contains monocytes, especially in the colostrum,10 and that all the bacterial species shed within the milk are facultative intracellular parasites. An intraleukocytic position could protect bacteria from a number of antiinfective substances in milk, including immunoglobulins, lysozyme, peroxidase, and lactoferrin.
Supposing the foregoing hypotheses to be true, let us attempt to close the circle in the outbreak reported by Fleming et al.4 in this week's issue of the Journal. L. monocytogenes organisms excreted in cow's milk escaped pasteurization (in part because they were shielded by their intraleukocytic position4), grew well at refrigerator temperatures, and were ingested. The organisms underwent endocytosis by the epithelial cells of the intestinal villi and progressed to the stroma, where they were phagocytosed by monocytes. Healthy people who drank the milk overcame the infection by virtue of cell-mediated immunity. Immunosuppressed adults (42 patients in this study) became sick, often (31 per cent) with meningitis. Perinatal infections also occurred; five of seven instances suggested intrauterine infection, pointing to the immunosuppressive effect of pregnancy and the susceptibility of the placenta to infection. The usual tropism of the infection for the central nervous system and the placenta was evident, perhaps because unique structural cells at these sites transport the bacteria as do the intestinal epithelial cells. Although details of treatment and outcome were not given, one might have anticipated some instances of relapse because of the survival of organisms within mononuclear cells, which are poorly penetrated by many antibacterial agents.
Many questions are left unanswered by this hypothetical series of events. For example, why don't salmonellae show tropism for the central nervous system or placenta? How do these hypotheses relate to the (largely unexplained) tropism of other bacteria for the central nervous system and other sites? Are uptake of bacteria by epithelial cells and uptake by mononuclear phagocytes really analogous? Why should epithelial cells ingest some species of bacteria and not others? Do epithelial cells behave differently according to whether the host is immune or not? And is chronic excretion in cow's milk related to intracellular parasitism or to some other feature of the chronic carrier state? Given the advances in our understanding of pathogenic processes, L. monocytogenes emerges as even more intriguing in 1985 than when it was first identified as Bacterium monocytogenes in 1926.
Michael Barza, M.D.
Tufts University School of Medicine
Boston, MA 02111
References
| |||||||||||||||||||||||||||||||||||||
This article has been cited by other articles:
HOME | SUBSCRIBE | SEARCH | CURRENT ISSUE | PAST ISSUES | COLLECTIONS | PRIVACY | TERMS OF USE | HELP | beta.nejm.org Comments and questions? Please contact us. The New England Journal of Medicine is owned, published, and copyrighted © 2009 Massachusetts Medical Society. All rights reserved. |
Previous
