Search

Aztreonam, a minimally absorbable monobactam antibiotic that is active primarily against gram-negative bacilli, was given orally to 10 healthy volunteers in three regimens: 60 mg, 300 mg, and 1,500 mg daily for five days. The regimens were administered at least two weeks apart so that the flora could return to normal. Aztreonam had no effect on oropharyngeal flora cultured every other day. Fecal cultures from volunteers receiving the 300-mg regimen yielded no gram-negative bacilli after an average of 4.4 days of treatment. Administration of the highest dose gave similar results, albeit after a shorter interval (3.0 days). None of the regimens appreciably influenced total microscopic (anaerobic) bacterial counts in the feces. However, during administration of the 1,500-mg regimen, fecal counts of enterococci increased from an average of approximately 10(4)/g to greater than 10(7)/g, and fecal counts of yeasts tended to increase by the end of the five-day treatment period. Fecal concentrations of aztreonam increased as the dose increased in eight volunteers. No aztreonam was recovered from the feces of two volunteers, perhaps because of drug inactivation.

In order to prevent infections after irradiation and bone marrow transplantation, rhesus monkeys were subjected to strict reverse isolation. They were given oral nonabsorbable antibiotics in order to eliminate their microflora. The antibiotic combination was selected on the basis of a sensitivity test and was added to the liquid food supply. The same antibiotics were also given orally for five days so that a high bactericidal concentration could be rapidly achieved in the intestine. After the additional antibiotics were discontinued colonizations were found. In contrast to colonizations persisting from the first day of treatment, these were fairly easy to suppress. All colonizations found after Day 7 were considered a break in isolation. It was concluded that a germ-free status had not been achieved and an extreme microflora suppression was all that had been accomplished. Apart from a 5% decrease in the haematocrit, no effect of the decontamination was seen on parameters for liver and kidney functions. The antibiotic concentration in the faeces was well above the amount indicated by the sensitivity test.

Summary Conventional mice were treated orally with various doses of norfloxacin and rosoxacin, respectively, to determine to which extent the “colonization resistance” (CR) of the digestive tract is affected. In addition, the effect was studied of both drugs on the endogenous Enterobacteriaceae and enterococci in the digestive tract (faeces). According to parameters for the CR of the digestive tract, such as the appearance of β-aspartylglycine and the relative caecal weight, norfloxacin did not decrease the CR in a dose range of up to 0.30 mg/mouse/day. A dose of 0.05 mg/mouse/day sufficed to clear the digestive tract of endogenous Enterobacteriaceae within three days. The highest dose of rosoxacin of 6.0 mg/mouse/day did not decrease the CR. A dose of 0.9 mg/mouse/day cleared the digestive tract of endogenous Enterobacteriaceae . These two drugs could therefore be considered for use in selective decontamination of the digestive tract of neutropenic patients.

SummaryTwelve antimicrobial drugs were studied for their effect on the endogenous aerobic potentially pathogenic bacteria(Enterobacteriaceae, Streptococcus faecalis)in the intestines and on the colonization resistance (CR) of the digestive tract. Three subclasses of antimicrobial drugs could be recognized: (1) those which suppress the CR following low oral doses (rifamycin, penicillin V, cloxacillin, fenethicillin); (2) those in which the CR is suppressed only following relatively high oral doses (amoxycillin); and (3) those in which no obvious suppression of the CR was noticed even following substantial oral doses (nalidixic acid, cinoxacin, co-trimoxazole, oral cephalosporins, piv-mecillinam and doxycyclin). Some of the drugs in the third category were found to suppress endogenousEnterobacteriaceae(nalidixic acid, co-trimoxazole, piv-mecillinam and doxycyclin) andS. faecalis(doxycyclin) at dose levels at which they did not decrease CR.

After oral treatment with polymyxin for only 1 day, fecal aerobic gram-negative bacteria were found completely suppressed in C3H/Law mice. Complete suppression of aerobic gram-negative bacteria was accompanied by a reduction of the fecal endotoxin concentration from 100 to 10 micrograms endotoxin per gram of feces as measured with the Limulus amebocyte lysate assay. Oral administration of polymyxin affected hemopoietic stem cell kinetics at different stages. The kinetic behavior of hemopoietic stem cells was determined as the in vivo sensitivity to the S phase specific cytostatic drug hydroxyurea. The hydroxyurea kill of bone marrow spleen colony-forming cells diminished not significantly (p less than 0.10) from 14 to 4% after 2 days of polymyxin treatment. Already after 1 day of treatment the hydroxyurea kill of bone marrow progenitor cells forming granulocyte-macrophage colonies in vitro decreased from 29 to 7% (p less than 0.05). It took 8 days of treatment before the hydroxyurea kill of splenic granulocyte-macrophage colonies was found reduced from 53 to 14% (p less than 0.001). The decreased susceptibility of hemopoietic stem cells to hydroxyurea during polymyxin treatment appears to argue for a role of intestinal aerobic gram-negative bacteria in the regulation of hemopoiesis, probably mediated by endotoxin.

The influence of temocillin on the colonisation resistance of the digestive tract was studied in mice, with doses ranging from 0.062 to 1 mg/day administered orally and 0.25 to 4 mg/day administered subcutaneously. The effect on the colonisation resistance was measured by the concentration of resistant bacteria (either endogenous enterococci or an exogenous strain of Enterobacter cloacae) as well as by the relative caecal weight and the appearance of beta-aspartylglycine in the caecal contents. None of these parameters indicated a decreased colonisation resistance by any of the dosages investigated, meaning that, in the mouse the drug can be used safely without an increased risk of overgrowth by resistant bacteria causing superinfections. Both systemic and oral administration of 0.25mg of temocillin/mouse/day (10 mg/kg/day) appeared to completely suppress the sensitive Gram-negative bacilli from the digestive tract within 2 days. This selective decontamination of the digestive tract in mice was the reason for a subsequent study in man. It emerged that selective decontamination with parenterally administered temocillin was possible in 7 out of 10 volunteers. The remaining 3 however, did not respond to temocillin although their faecal Gram-negative bacilli were susceptible to the drug. In none of the volunteers did the colonisation resistance appear to be affected. The consequences of the use of temocillin for either prevention or therapy of infections are discussed.

Nystatin was administered in ten healthy adult volunteers in increasing doses of 3 X 10(6) I U, 6 X 10(6) I U, 9 X 10(6) I U and 12 X 10(6) I U per day, each dose being given for a five-day period. Faecal samples were collected daily for the determination of their concentration of biologically active nystatin. Nystatin concentrations were determined biologically; the sensitivity of this method was less than or equal to 20 mcg/g of faeces. During the four treatment periods with increasing doses, 38%, 31%, 26% and 20% respectively of the faecal samples contained biologically undetectable amounts of nystatin. This means that nystatin is either inactivated or unevenly distributed through the intestinal contents, or both. The practical consequences of this may be that in a significant portion of the colon there is no inhibitory nystatin concentration against Candida albicans, despite treatment with as much as 12 X 10(6) I U of nystatin per day.

In the present study we investigated the occurrence of delayed-type graft-versus-host disease (DT-GvHD) after allogeneic bone marrow transplantation (BMT) between two H-2 incompatible mouse strains. BMT was performed on mice with a conventional intestinal microflora as well as on mice in which the Enterobacteriaceae were selectively eliminated from the intestinal microflora by oral antibiotic treatment. None of the conventional or the selectively decontaminated (SD) chimaeric mice suffering from DT-GvHD died of bacteraemia. While DT-GvHD was mitigated when C3H/He recipient mice were SD-treated, this was not the case when C57B1/6J recipient mice were SD-treated. SD-treatment of the digestive tract of donor mice only mitigated DT-GvHD when the recipients were also SD-treated. We conclude that Enterobacteriaceae in the digestive tract may only play a minor role, if any, in the occurrence of DT-GvHD. Instead, we postulate that in this study DT-GvHD was determined by differences in the composition of the resident intestinal microflora (IM) of both mouse strains together with the cellular composition of the bone marrow graft. The interaction between antigenic components of the recipient's IM and the developing donor immune system in the recipient as a possible cause for DT-GvHD is discussed.

Article de synthese. Etude des divers aspects de l'ecologie microbienne de l'intestin dans des modeles animaux. Developpement de la flore intestinale apres la naissance. Facteurs ecologiques associes a la colonisation de l'intestin par Clostridium difficile

Animal and human studies have suggested the concept of “colonization resistance” of the gastro-intestinal tract, which can be decreased by administration of antibiotics that inhibit the anaerobic portion of the normal flora of the gut. This effect can be prevented by the production, by resistant members of the flora, of bacterial enzymes that inactivate or destroy the antibiotic in question. Possible changes in the prevailing gut flora and the implications for the incidence of different infecting agents and for antibiotic therapy are discussed.

Summary If antimicrobial drugs are used without paying notice to their effect on the digestive tract microflora, resistance, « overgrowthand super-infection may emerge. Regarding their effect on the alimentary tract flora, antimicrobial drugs can be classified in : 1. drugs which do suppress the CR-determining (anaerobic) flora and 2. in antimicrobial agents which leave the CR responsible flora unchanged. The first category of antibiotics will not cause « overgrowthby resistant microorganisms. They can even be successfully applied for infection prophylaxis in neutropenic patients, provided they are sufficiently high dosed to completely suppress all susceptible potential pathogenic microbes which colonise the digestive tract. This application of CR-indifferent drugs is called Selective Decontamination of the digestive tract (SDD). Antibiotics on the other hand which decrease the CR by suppressing CR-responsible (anaerobic) bacteria in the alimentary canal, should become « second choice antibiotics; particularly in a clinical setting. If these drugs are indicated on the basis of the sensitivity pattern of microorganisms involved in an infection, on pharmacokinetic properties, or on kidney function or allergy for CR-indifferent drugs, combination of therapy with SDD should be considered.

Lethally irradiated conventional mice were killed at daily intervals after irradiation. A high correlation was found between the occurrence of bacterial over-growth in the digestive tract and invasion into the regional lymph nodes the spleen and the blood. By oral contamination of mice with quite high doses of either an exogenous or an endogenous biotype of an Enterobacteriaceae species at day 4 after irradiation, it was attempted to induce such condition of intestinal over-growth. In all animals such an abnormal colonization of the contaminant was indeed achieved in this way and the mice died from an infection caused by the biotype used for the contamination. The interval between contamination and death was different when different biotypes were used for contamination and occurred sooner after irradiation when an endogenous biotype was used for (re)contamination. It is concluded that oral contamination with exogenous as well as with endogenous potentially pathogenic bacteria during a period of severely decreased resistance to infections must be avoided. The colonization resistance of the digestive tract of these animals should be maintained as high as possible. © 1978 H. Veenman & Zonen B.V. Publishers.

Oral and faecal flora were monitored in ten healthy volunteers who received three different oral dosages of either norfloxacin or ciprofloxacin during five consecutive days. Oral flora was not influenced by either quinolone. At the end of the five day-treatment period aerobic Gram-negative bacilli could no longer be cultured from the faecal samples of any of the volunteers. Enterococci were suppressed in some volunteers or remained present in concentrations as high as before treatment. All ciprofloxacin-treated volunteers became colonized with yeasts in their faecal samples, even during treatment with the lowest dose. As measured by total microscopic bacterial clump counts and the appearance of beta-aspartylglycine in faecal samples neither of the two quinolones appeared to decrease the colonization resistance of the digestive tract.

Oral treatment of mice with a combination of non-absorbable broad spectrum antibiotics at a dose level high enough to sterilize the digestive tract within five days of treatment, has been found to have a significant negative effect on the relative weight of the thymus. From the age of 12 weeks onwards, the mean relative weight of the thymus of conventional mice appeared to differ significantly from the mean relative thymus weight in germ-free mice of the same strain. In the course of two to three weeks of continuous oral antibiotic treatment the mean relative weight of the thymus decreased from conventional values to values seen in germ-free mice of the same strain and age. It is concluded that the intestinal microflora apparently exerts an influence on the thymus.

The suppressive effect of amikacin, gentamicin, tobramycin and paromomycin on the aerobic endogenous flora and on the colonization resistance of the digestive tract was tested by administering one of the antibiotics orally at five different dose levels. At a certain dose level, all antibiotics suppressed the endogenous Enterobacteriaceae species. Amikacin was particularly effective in this respect. Low doses of amikacin rapidly destroyed the colonization resistance. This resistance only remained unaffected in animals treated with tobramycin in doses that were still adequate to completely suppress the endogenous Enterobacteriaceae species. We concluded that of all the antibiotics tested in this study, only tobramycin may have a future in (clinical) application for the selective decontamination of the digestive tract. Selective decontamination can be considered an effective method for infection prevention in leukopenic patients.

SUMMARYDuring systemic treatment of mice with ampicillin or streptomycin, oral contaminations with exogenous bacterial species resulted in an abnormal colonization pattern. The contaminants persisted much longer and in much higher concentrations in the caecum of systemically treated mice than in control animals. Spread of the contaminant into the mesenteric lymph nodes and the spleen was found much more often in the antibiotic treated group. This, however, was only seen when the contaminant was ‘resistant’ to the antibiotic injected. The experiments suggest that the ‘CR-inducing species’ of the microflora live in close contact with the mucosa and therefore could be identical with the anaerobic tapered rods described by Savage & Dubos (1968).

SUMMARYArguments in favour of biotyping of Enterobacteriaceae excreted in the faeces of isolated patients, as a method of investigating the efficiency of the isolation procedures, are presented as well as a technical outline of the procedure. The study included three kidney transplantation patients, five acute myeloid leukaemia patients and four healthy persons as controls.The results show, apart from new colonizations during isolation, a difference in the mean number of contaminations and colonizations with different Enterobacteriaceae biotypes. It is concluded from these results, that the isolation procedures were not completely effective and that the AML patients studied had a decreased colonization resistance of their digestive tract. This was less evident in the kidney transplant group.

SUMMARYThe effect of oral administration of antibiotics on the intestinal flora of conventional mice and their resistance to colonization by orally introducedEscherichia coli, Klebsiella pneumoniaeandPseudomonas aeruginosawas studied. Colonization resistance (CR) was expressed as the log of the oral bacterial dose followed by a persistent take in 50 % of the contaminated animals. The intestinal flora was virtually eliminated by the antibiotics and this elimination was accompanied by a precipitous fall of CR. CR gradually returned to normal values during the period of repopulation of the intestinal tract by the organisms surviving the treatment. Antibiotic treatment resulted in the disappearance of Enterobacteriaceae, enterococci, staphylococci and yeasts and, under appropriate housing conditions, the animals remained free of these organisms indefinitely. Germ-free mice contaminated with the intestinal flora of an antibiotic-treated animal and their offspring housed in a germ-free isolator showed high values of CR. Their intestinal flora consisted of anaerobic bacteria only. Apparently, these anaerobes are responsible for CR in these and in conventional mice.

The possible use of pipemidic acid for preventing infections in granulocytopenic patients by selective elimination of the gram-negative rods from the digestive tract was evaluated. Fecal samples were collected before, during, and after therapy from 23 adults who were undergoing treatment of urinary tract infections with pipemidic acid at a dosage of 200 mg orally four times daily for 10 days. The concentration of pipemidic acid in the majority of the stool samples tested was 220 to 600 micrograms/g. However, in 19% of the samples collected on day 5, the concentration was below the detection limit (23 micrograms/ml). Approximately 90% of the fecal samples collected during or shortly after therapy were negative for members of the family Enterobacteriaceae. Resistance of the gastrointestinal tract against colonization by microorganisms from the environment (so-called colonization resistance) remained unimpaired, as measured by the absence of beta-aspartylglycine in the stool.

During five consecutive days three different dosages of 200, 400 and 800 mg norfloxacin were administered to ten healthy males. Cultures of feces and oral washings revealed that even the lowest dose was effective in eliminating gram-negative bacilli from the gi-tract within a reasonable period of time. At this dose signs of a decreased CR due to norfloxacin and measured by both overgrowth by resistant ppmo or the appearance of beta-asp-gly in feces were not seen. Treatment with 100 mg norfloxacin twice daily seems therefore suitable for SDD, at least in volunteers. However, further studies in patients should be performed.

SUMMARYFive pigs were treated orally with norfloxacin for 5 consecutive days in two well-separated periods. This was done to determine the lowest dose required to free the pigs of Enterobacteriaceae. In the first period of the study, the animals were treated with 400 mg per day while in the second treatment 800 mg norfloxacin was given. Daily faecal culturing indicated that the faeces became free of Enterobacteriaceae in 3–5 days when treated with 400 mg/day, while all animals were found negative on culturing after 3 days of treatment with 800 mg/day. Investigation of the concentration of norfloxacin in the faeces revealed that a substantial fraction of the dose was either absorbed or inactivated by faecal substances. An in vitro study in faeces confirmed that a substantial part, some 75% of the dose, may have been inactivated by intestinal contents. This finding helps to explain the much lower concentration of 120 mg norfloxacin per kg of faeces in the pig in comparison with the almost tenfold higher concentration reported to develop in man during treatment with an identical dose.

Publisher Summary This chapter discusses the role of microbial flora in the development, maintenance, and modulation of host immune functions. The digestive tract is a very rich source of living and dead antigens. Continuous subsequent colonization with the species for which tolerance exists could result in the sensitization of other parts of the immune apparatus and the development of hypersensitivity reactions. The gastrointestinal tract, and particularly the lower bowel, consists largely of anaerobic bacteria. In the mouse and in man, most of these anaerobic species may be apathogenic. Some of the anaerobes appear even to exert a direct as well as strong indirect protective activity against colonization by most if not all newly arriving potentially pathogenic micro-organisms.

The in-vitro inactivation of seven antimicrobial drugs by human faecal substance has been investigated. Nalidixic acid, colistin, neomycin, tobramycin, temocillin, trimethoprim, and aztreonam were separately mixed in graded concentrations with faecal suspensions prepared from faeces of eight healthy volunteers. Each was mixed separately with a sample of each of the eight suspensions. All seven antimicrobial drugs appeared to be rapidly biologically inactivated by intestinal contents in a dose-dependent fashion, although not all to the same extent. The results may explain why with some unabsorbed or minimally absorbed drugs higher oral doses are required to achieve eradication of Gram-negative bacilli from the digestive tract than others, when selective decontamination (SD) is attempted in clinical practice.

The introduction and widespread use of antibiotics has resulted in profound changes in the number and character of infections that are encountered. This was already evident in the sixties (22, 15). Although the problem of staphylococcal infections still received considerable attention and emphasis from many workers in the early sixties, several groups of workers, including ourselves, became impressed by the increasing prevalence and seriousness of infections with Gram-negative bacilli (2). The great majority of these infections were seen in immuno-compromised patients (2,4). This led us to investigate this type of infections in an animal model, particularly with regard to prophylaxis of such infections.

A case of bilateral pneumonia, mediastinitis and septicaemia caused by Acinetobacter calcoaceticus and Candida albicans is described. The infections occurred after a palliative operation for an esophagotracheal fistula in a thorotrastoma patient. The oropharynx was colonized by the two microorganisms at admission and is presented as the source of these infections. Clinical management and antimicrobial policy, including oropharyngeal decontamination, leading to a good outcome are reported.

SUMMARYThe effect of oral administration of neomycin cephalothin or kanamycin cephalothin on the aerobic intestinal bacterial flora, was studied in dogs maintained under isolation conditions in a conventional animal room. The dogs were successfully freed of aerobic bacteria with both combinations within two to seven days after the start of antibiotic treatment, and were maintained bacteria free for up to 21 days. Decontamination was attained more rapidly in dogs that were bathed in hexachlorophene surgical soap before and during the first and third days of antibiotic treatment. There was no evidence of toxicity from either of the antibiotic combinations. These results indicate that, as with mice and monkeys, decontamination of dogs with oral antibiotics is feasible. The technique is of potential value in preventing endogenous bacterial infections in canine experimental studies involving use of immunosuppressive agents.

In this paper we described the results of bacteriological monitoring of oropharynx and stool samples from granulocytopenic patients with leukaemia who received oral infection prophylaxis with two different regimens for selective decontamination of the digestive tract. Patients were prospectively randomized either into a group receiving non-absorbable antimicrobial drugs for selective decontamination (polymyxin and neomycin: group A) or into a group receiving polymyxin and co-trimoxazole (group B). The oropharynx was, or became, free of gram-negative bacilli within one week of treatment in 94% and 90%, respectively, of the patients in group A and group B. The stool samples were, or became, negative after the same treatment interval in 91% and 80%, respectively, of the two patient groups. Antibiotic therapy during selective decontamination treatment significantly increased the incidence of positive cultures from the oropharynx and stools. The sensitivity of the gram-negative bacilli isolated during selective decontamination treatment to the drugs administered did not influence the average response to treatment. Both resistant and sensitive gram-negative bacteria appeared to disappear from the patients' samples, mostly within a week, without the need to adjust the selective decontamination treatment. Yeasts behaved in almost the same way as gram-negative bacilli. All patients received oral amphotericin B; some patients occasionally yielded oropharyngeal or faecal cultures which were positive for yeasts.

This paper describes a novel method for detecting immunoglobulin A (IgA) coated potentially pathogenic microorganisms (PPMs) in the human intestine. Essentially, the technique consists of 2 phases: one in which IgA coated bacteria are detected by immunofluorescence and a second in which these bacteria are subcultured in situ and subsequently identified. In this way transient bacteria are differentiated from resident bacteria. These results show that the resident bacteria are coated with IgA. Resident microorganisms are always highly concentrated in the digestive tract. These results strengthen the hypothesis that only the high antigen concentrations achieved by a large number of resident bacteria are capable of IgA induction.

The Gut Associated Lymphoid Tissue [GALT] is considered to play an important role in infection prevention and in excluding the absorption of antigens [immune exclusion]. Secretory immunoglobulin A [S-IgA] on the mucous membranes may be involved in this protective GALT activity by interfering with bacterial adherence. Attachment to mucosal surfaces via fimbriae may be a prerequisite for many bacteria to colonize and to penetrate the digestive tract. Adherence of fimbriaeted bacteria is apparently inhibited by S-IgA, because the latter blocks interaction fimbriae and mucosal surface. The anti-adhesive or coating phenomenon of S-IgA may therefore be a more important measure of the functional characteristics of S-IgA than the simple quantitation of S-IgA in serum and secretions. In the present study we discuss an in vitro test system, the fluorescence immuno-assay [FIA] to determine the coating’s incidence as a functional parameter of GALT activity [GALT-FIA].

Potentially pathogenic bacteria and yeasts meet a number of resistance factors. These colonization resistance (CR) determining factors control the growth density of potentially pathogenic microorganisms in the various parts of the alimentary canal. The CR of the digestive tract is to be seen partially of host origin but to a greater extent due to direct and indirect activity of the (anaerobic) indigenous microflora. This implies that the CR mechanism is vulnerable to antibiotics which affect the indigenous flora during treatment. A decreased CR enhances colonization by microorganisms that are resistant to the antibiotic(s) used for therapy. Increased population densities of potentially pathogenic bacteria (“overgrowth”) may correlate with invasion of the mucosal tissues often followed by translocation to remote lymphatic organs. Overgrowth also may correlate with development or acquisition of resistance to the antibiotic applied. By screening in experimental animals antibiotics have been selected which do not affect the CR. Such antimicrobial drugs have successfully been used to selectively eliminate potentially pathogenic bacteria and yeasts in neutropenic patients.

The enzymatic inactivation of aztreonam by the faecal flora was investigated in 25 healthy human volunteers. Ten volunteers received aztreonam orally and 15 volunteers did not receive antibiotic treatment. Residual aztreonam and its inactivated open ring form were simultaneously detected by an HPLC-assay. Independent of aztreonam treatment, beta-lactamase activity was detected in the faecal flora of 20 volunteers. The inactivation of aztreonam by the faecal supernatants was inhibited by clavulanic acid.

SUMMARYA transportable surgery cross-flow unit has been tested under ‘operational conditions’. By the use of artificial aerosols and a volunteer surgical team, or dummies, it was found that, at an air velocity of 0·45 m./sec., a detectable transfer to above the table occurred only when quite highly concentrated aerosols (of more than 103·6 bacteria/m.3 of air or more) existed underneath the table. The short disappearance time under these conditions and the quite stable flow pattern above the table found when a surgical team was working, standing along both sides of the table, make it unlikely that an aerosol of detectable concentration can develop during surgery, at this site. The chance that particles, liberated from the heads of the surgical team, settle on the table, was found to be strongly reduced when a cross-flow tunnel operated at an air velocity of 0·45 m./sec. The transfer from outside the unit to the inside was prevented by closing the upper part of the open front side.

Gram-negative infections are generally regarded as a major cause of severe infections in granulocytopenic patients. To prevent such infections, selective decontamination (SD) of the digestive tract has been introduced. It is assumed that if SD is applied on all patients in an oncologic station it may not only provide protection to patients thus treated, but it may have also epidemiologic consequences. SD may considerably reduce the number of circulating (antibiotic-resistant) strains in the ward and therewith significantly further reduce the chance of severe infection during periods of granulocytopenia. Because the effect which different kinds of antibiotics may have on the accumulation and spread of resistant bacteria cannot be studied in patients, it was investigated experimentally. Groups of 15 conventional mice were lethally irradiated and each group was treated with one antibiotic. On the day of irradiation one mouse per cage was orally contaminated with a multi-resistant strain either of E. coli or Ps. aeruginosa. The spread of these strains among cage mates as well as their effect on mortality was investigated. It appeared that ampicillin promoted spread and enhanced mortality in comparison with a control group, whereas antibiotics currently used for SD did prevent spread of the multi-resistant strain and reduced mortality.

SUMMARYA mobile laminar cross-flow unit for surgery has been evaluated by the use of an experimental bacterial aerosol of Escherichia coli in different concentrations, generated at several different sites. A separate ventilation system, mounted underneath the table, produced an almost downward directed curtain of sterile air along both sides of the table. When the velocity of the air stream, discharged by the cross-flow unit, was adjusted at 0·50 m./sec. at 2 m. from the filter face (at the head end of the operation table), the surface of the table could be maintained free of bacterial settling, even after aerosolization of heavily concentrated suspensions of 108 bacteria/ml. at different sites outside the enclosure and underneath the table as well.

In previous studies we have shown that some antibiotics may influence the growth of experimental tumors. Concerning rifampicin an enhancing effect on the tumor development was combined with the impairment of some parameters of antitumor immunity (6). A marked suppression of the antitumor immunity was also observed in mezlocillin treated mice (5). However, this antibiotic caused a surprising antitumor effect when given during early tumor growth. Because a direct cytotoxic activity of mezlocillin could not be demonstrated, indirect mechanisms have been suggested for this mezlocillin effect. The present paper deals with the influence of mezlocillin and other selected antibiotics on the endogenous intestinal flora of BALB/c mice and on the local growth of sarcoma L-1 tumor.

Selective decontamination of the digestive tract by antimicrobial drugs with limited spectra of antibacterial activity has been reported to reduce the risk of superinfection and cross-infection in immunocompromised hosts by leaving the flora responsible for colonization resistance (largely anaerobic) intact. Aztreonam, a monobactam, and temocillin, a beta-lactam antibiotic, are possible candidates for use in selective decontamination of the digestive tract because of their specific activity against gram-negative bacteria. When these drugs were administered orally to mice, low dosages (0.016-1.13 mg per day) did not significantly affect colonization resistance. In volunteers, the indigenous flora either was not affected or was only minimally affected by orally administered aztreonam (20 mg and 100 mg every 8 hr). Intramuscular temocillin also was effective. However, in a small number of subjects inactivation (most likely enzymatic) of these antibiotics occurred in the bowel. Both drugs can be considered serious candidates for use in the selective decontamination of the digestive tract in granulocytopenic patients.

An outline is given of the various host-related and flora-related parts of the colonization resistance of the digestive tract. The host-related part of the colonization resistance has been found to be somewhat decreased by sublethal irradiation and leukemia (or chemotherapy), while treatment with antibiotics active against gram-positive flora may severely decrease the colonization resistance (depending on the antibiotic concentration established within the digestive tract during antibiotic therapy). The flora-related part of the colonization resistance, which controls colonization by potentially pathogenic microorganisms, differs greatly from one individual to the next. This observation appears to be important for the host-related part of the colonization resistance. Finally, it is concluded that in the clinical situation preference should be given to antibiotics which do not affect the part of the flora constituting colonization resistance for two reasons: 1) to limit the spread of (multi-) resistant potentially pathogenic microorganisms and 2) for infection prophylaxis in immunocompromised patients. In the latter situation, the potentially pathogenic microorganisms in the flora are selectively eliminated from the digestive tract, provided the antimicrobial drugs used for selective decontamination are active against the endogenous potentially pathogenic microorganisms and given in sufficient (oral) doses.

A growing body of information is being focused on the infection hazard of special care areas in hospitals. The sources of infection appear to be both endogenous and exogenous in origin. The sources of nosocomial exogenous infection are no longer a mystery. The bacteria, yeasts and molds associated with these infections can be traced to other patients in the unit, to intravenous and urinary catheters and to wash sinks and utility areas. It is generally agreed that prime factors in their transmission are the personnel who work in these units. Consequently, for practical and financial reasons, more emphasis is in general given to handwashing than to strict isolation and architectural design of special care units.

A method to prevent endogenous infections in patients and laboratory animals with an impaired immune capacity is the total decontamination of the gastrointestinal tract. This is performed by oral administration of nonabsorbable antibiotics. One of the disadvantages of total decontamination is the fact that especially after termination of this treatment the decontaminated individual is easily colonized with organisms from the environment. This is the result of the elimination of the anaerobic part of the microflora, which is responsible for the so-called colonization resistance (CR). This CR is the resistance against exogenous microorganisms to colonize the gastrointestinal tract. In the absence of an anaerobic flora, the colonizing microorganisms can reach abnormally high faecal concentrations, thus increasing the risk for infection. In mice, the implantation of an anaerobic, mouse-derived flora after termination of total decontamination resulted in the restoration of a good CR, as could be shown by orally challenging the animals with a strain of Escherichia coli. Therefore, an anaerobic microflora, free of potentially pathogenic microorganisms, which was isolated from a healthy human volunteer was administered to monkeys and patients after a decontamination period in an attempt to restore CR. In the monkeys, this human donor flora (HDF) did not reduce the faecal concentration of microorganisms that had colonized the gastrointestinal tract before the donor microflora had been established, in contrast to the findings in some of the patients. Qualitative analysis of the microflora of patients which were contaminated with the human donor flora showed that the CR is not of the same quality as is found in healthy individuals. This can be the result of the impaired immune capacity of the patients at the time of HDF implantation. However, the results obtained show that implantation of the HDF in monkeys and patients after a decontamination period allows reconventionalization without an undue risk of microbial infection.