Replication of IFDO on a chemically defined medium

An agar or liquid medium containing haemoglobin, high density horse lipoprotein, trypsin, Tween 80, phosphate buffer, CaCl(2), glucose, glutamic acid and NaCl supported growth of ileal fluid dependent organism (IFDO). Glucose, glutamic acid and NaCl were not essential but enhanced growth. Trace amounts of lipoprotein were sufficient to support growth, and some human sera and serum fractions rich in low density lipoprotein could be substituted for horse lipoprotein. Addition of lipase enhanced the growth rate, and reduced the requirement for lipoprotein. No nucleic acid precursors were identified as essential for growth. However, nucleosides, especially cytidine, accelerated the growth rate. The growth rate was also increased by DNAase and RNAase. These observations indicate that the organisation of the IFDO particle is more complex than that of a crystal. They are consistent with the hypothesis that IFDO is a replicating agent that utilises specific preformed protein to assemble a proteinaceous particle, and support the postulated relationship of IFDO to transmissible spongiform encephalopathy agents.

Effects of chlorhexidine-coated intrauterine device on the bacteriology of the uterine cavity

Twenty-nine women, aged 31-53 years, scheduled for an abdominal hysterectomy, participated in this study. The patients were divided into three groups. The first received a Multiload MLCu250 intrauterine device (IUD); the second group received a chlorhexidine acetate medicated Multiload MLCu250 IUD; the third group acted as a control group receiving no IUD. Bacteriological cultures of the vagina and ectocervix were taken prior to insertion of the IUD. At hysterectomy, some 18 hours later, specimens were taken from the cervical canal, uterine cavity, and the device itself, for culture of anaerobic and aerobic organisms using a standardized previously validated technique. In three of the nine control patients, organisms were cultured from the uterine cavity. There were no differences between the bacteriological results of the two groups with medicated and non-medicated devices, with regard to the cultures from the cervical canal, uterine cavity or the devices themselves. Nor was there any difference between the control and the IUD group.

Short-term chemoprophylaxis with ceftizoxime vs. five-day aminoglycoside with metronidazole in 'contaminated' lower gastrointestinal surgery

In a prospective, open, randomized controlled trial, 173 patients requiring surgery for potentially "contaminated" lower gastrointestinal surgery were allocated to receive either two doses of ceftizoxime (2 gm) with one dose of metronidazole (1.5 gm) or gentamicin 3 mg/kg/day for five days with one dose of metronidazole (1.5 gm). Eighty-nine patients received ceftizoxime and 84 patients received gentamicin. The groups were comparable with respect to diagnosis, procedure, type of anastomosis, and wound closure. The incidence of withdrawal due to failure to respond to the study drug (11.5 percent) was equivalent in the two groups. There was no difference in the overall incidence of postsurgical infection between the ceftizoxime and metronidazole group (22.2 percent) and the gentamicin and metronidazole group (25.7 percent). The incidence of wound infection (ceftizoxime and metronidazole, 6.9 percent; gentamicin and metronidazole, 10 percent) and deep sepsis (ceftizoxime and metronidazole, 15.3 percent; gentamicin and metronidazole, 15.7 percent) was similar.

Studies of a novel agent possessing resistance to moist heat and disinfectants: parallels with Creutzfeldt-Jakob agent

The resistance to sterilization and disinfection of a novel replicating agent (IFDO) with similarities to Creutzfeldt-Jakob agent (CJA) was investigated. Moist heat at 121 degrees C for 30 min did not kill the agent. Increasing the temperature to 140 degrees C, and the length of the autoclave cycle to 120 min also failed to guarantee sterilization, although some samples were sterilized after these treatments. Dry heat at 160 degrees C for 1 h sterilized 24 of 25 samples. Overnight disinfection with 10% Stericol or 1.2% chlorhexidine left few detectable survivors. Samples pretreated with these disinfectants and then autoclaved at 134 degrees C for 20 min were sterilized. Disinfection with hypochlorite (500 ppm available chlorine) was very effective if organic matter derived from spent culture medium was removed. We have adopted overnight Stericol disinfection, followed by autoclaving at 134 degrees C for 1 h for sterilization of glassware contaminated by IFDO. The agent may provide a valid model for sterilization of items contaminated with CJA and, if so, our data suggest that current disinfection guidelines for CJA by autoclaving at 134 degrees C for 1 h are inadequate.

A randomized trial to compare amoxycillin/clavulanate with metronidazole plus gentamicin in prophylaxis in elective colorectal surgery

A randomized controlled trial was designed to compare antibiotic prophylaxis with a standard combination of agents, metronidazole and gentamicin, with a single preparation, amoxycillin/clavulanate in 400 patients undergoing elective colorectal surgery. There were 41 patients who were excluded or withdrawn (wrong dose, inappropriate operation, established sepsis or concurrent disease). Abdominal wound sepsis occurred in 14% of the assessable patients in the amoxycillin/clavulanate group and in 15% of the metronidazole plus gentamicin group. Perineal sepsis occurred in 27% of the amoxycillin/clavulanate group with a perineal wound compared with 18% in the metronidazole plus gentamicin group. Intra-abdominal abscess occurred in 8% of those who received amoxycillin/clavulanate compared with 6% of those given metronidazole plus gentamicin. Only two patients in each group developed septicaemia. Postoperative diarrhoea occurred in 11 patients receiving amoxycillin/clavulanate compared with four given metronidazole plus gentamicin. Clostridium difficile was not isolated from the stool cultures in any of these cases. Thirteen of the 164 abdominal or perineal wounds were infected by 15 strains of Bacteroides spp. in the group receiving amoxycillin/clavulanate compared with only three of the 165 wounds in those given metronidazole plus gentamicin. (P less than 0.01). There was no other significant difference in the pattern of isolates between the groups.

The penetration of aztreonam, a monobactam antibiotic, into intra-abdominal abscesses

We have studied the penetration of aztreonam into pus with a low mortality intra-abdominal abscess model in the Wistar rat. The mean peak serum concentration (standard error) of aztreonam was 74.6 +/- 9.88 mg/l at 30 min falling to 2.31 +/- 1.28 mg/l at 8 h. The mean concentration in pus was 17.04 +/- 2.01 mg/l at 2 h and was 14.47 +/- 2.23 mg/l at 8 h. These concentrations are in excess of the MICs of most Gram-negative aerobic bacilli commonly isolated from intra-abdominal abscesses.

A novel replicating agent isolated from the human intestinal tract having characteristics shared with Creutzfeldt-Jakob and related agents

A novel replicating agent (IFDO) was isolated from ileal fluid. Growth occurred in vitro under aerobic and anaerobic conditions, and was faster at 37 degrees C than at room temperature. The doubling time was 15.8 min. Colonies were dark brown in colour and occurred beneath the surface of agar after conventional surface inoculation. Provisional data indicate that the agent may be a normal intestinal commensal. The agent was remarkably resistant to inactivation by steam at 134 degrees C, formaldehyde and glutaraldehyde; it was relatively resistant to ionising radiation, and it was filterable through membranes with a nominal pore diameter of 10 nm. Such properties, with the exception of growth in cell-free medium, are shared by "unconventional agents" such as those of Creutzfeldt-Jakob disease and scrapie. Further comparison of the properties of the intestinal agent and of slow viruses revealed additional shared characteristics, including resistance to proteinase K and trypsin, and inactivation by guanidine thiocyanate, diethyl pyrocarbonate, phenol and sodium hydroxide. The agent differs from that of scrapie in being inactivated by ethidium bromide, zinc nitrate, EDTA, hydroxylamine in the presence Sarkosyl, and, under certain circumstances, by ribonuclease. Broth cultures of the agent contained particles possessing considerable size heterogeneity. The smaller filterable particles were generally more susceptible to inactivation, did not survive autoclaving, and were inactivated by papaya protease and lipase. It is possible that the replicating agent may be formed by crystallisation from constituents of the medium, and not by a biological process. This does not exclude the postulated relationship to slow viruses.

A comparison of the penetration of two quinolones into intra-abdominal abscess

A low-mortality model of an intra-abdominal abscess in the rat has been used to study the penetration of two quinolone agents into pus. Maximum concentrations in pus after intravenous injections were achieved at four hours (ciprofloxacin: 12.7 +/- 3.69 mg/L, fleroxacin: 2.25 +/- 1.82 mg/L), whereas fleroxacin given orally reached the maximum level at two hours (13.39 +/- 3.13 mg/L). Higher concentrations of fleroxacin were recorded in pus than in serum at each time point up to eight hours after administration, but pus levels of ciprofloxacin only exceeded serum levels after 1.5 hours. These antibiotics appear to have a unique property of high penetration into established abscesses and may have an important therapeutic role in the treatment of patients with multiple interloop abscesses.

Single dose, oral antibiotic cover for transurethral prostatectomy

A double-blind, randomised, placebo-controlled study was carried out to determine the incidence and significance of bacteriuria in 110 patients undergoing transurethral resection of the prostate (TURP) and to assess the effect of a single pre-operative dose of Ciprofloxacin, a 4-quinolone antibiotic. Fifteen (68%) of the 22 patients in the placebo group with a positive post-operative urine culture subsequently developed a clinically apparent urinary tract infection (UTI) or received antibiotics in view of a positive urine culture. Adequate prostatic concentrations of Ciprofloxacin were achieved in all who received the drug. A significant reduction in the number of positive post-operative urine cultures and urinary tract infections requiring antibiotic therapy was achieved in this group. Six patients (5.5%) developed clinical evidence of septicaemia, 5 of whom were in the placebo group. No organisms resistant to Ciprofloxacin were encountered. Prior to surgery, 19% of all patients were found to have previously unsuspected bacteriuria. Ciprofloxacin tended to reduce the chances of this group developing a UTI or requiring antibiotics. Further, there was a highly significant reduction in post-operative infective complications in those with sterile urine at the time of resection who had received the drug. This study suggests that antibiotic cover for TURP is of clinical benefit. Ciprofloxacin may prove suited to this purpose, although further experience with the drug is still required.

Comparison of short-term antibiotic cover with a third-generation cephalosporin against conventional five-day therapy using metronidazole with an aminoglycoside in emergency and complicated colorectal surgery

In a prospective, randomized control trial, 152 consecutive patients requiring emergency or complicated colorectal surgery were allocated either to two doses of cefotetan or to five-day cover with gentamicin, and a single dose of metronidazole. Twenty-one patients received 6 gm of cefotetan before prolongation of prothrombin time dictated a change in the dose regimen such that all remaining patients (N = 55) received only 4 gm of cefotetan. The groups were well matched for diagnosis and surgical procedure. Rates of postoperative infection did not differ significantly between the groups, with wound infection rates occurring in 17 of 75 patients receiving gentamicin and metronidazole (22.7 percent) compared with ten of 75 receiving cefotetan (13 percent). Although wound infection rates were lower in the cefotetan group, the incidence of intra-abdominal abscess was similar in both groups. Eight patients receiving cefotetan developed intra-abdominal abscesses (11 percent), compared with seven receiving gentamicin and metronidazole (9 percent). Prolongation of prothrombin time in excess of 13 seconds occurred in six patients receiving cefotetan compared with no patients receiving gentamicin and metronidazole. None of these patients developed clinical bleeding, however.

The effects of Clostridium difficile crude toxins and toxin A on ileal and colonic loops in immune and non-immune rabbits

Rabbits were solidly immunised by parenteral injection of purified Clostridium difficile toxin A such that they resisted an intravenous challenge with a normally lethal dose of toxin A. Ileal and colonic loops constructed in non-immune and immune animals received challenge injections of crude culture filtrate or purified toxin A of C. difficile. Protection of ileum was manifest after sufficient initial mucosal damage resulted in release of high levels of antitoxin A into the loop lumen of immune animals. There was less fluid accumulation in ligated ileal loops of immune than of non-immune rabbits. Less protection was observed when loops were challenged with crude culture filtrate containing toxins A and B than when challenged with purified toxin A. In-vitro studies with Ussing chambers yielded no evidence for tissue-localised immunity as judged by electrical responses and histology of toxin-treated tissue from non-immune and immune animals. No differences were found in the degree of epithelial damage, or volume or composition of fluid accumulating in colonic loops of non-immune and immune rabbits challenged with toxin A or crude culture filtrate. However, in colonic loops of immune rabbits there was no overt tissue-localised haemorrhage, whereas in those of non-immune rabbits tissue-localised haemorrhage was marked. In contrast to our findings with ileal loops, fluid accumulating in colonic loops was watery and contained substantially less total protein and (in immune animals) antitoxin A.

Biological mode of action of Clostridium difficile toxin A: a novel enterotoxin

Antibody neutralisation and toxin A elution experiments showed that toxin A uptake from rabbit intestinal lumen was a continuous process. The kinetics of the ileal and colonic responses were significantly different; a much longer incubation (4 h) with toxin was required for colon, compared with 45 min for the ileum, to induce fluid accumulation at 12 h. Fluid secretion was induced only when toxin had gained access to deeper tissues, probably achieved by several toxin uptake-tissue damage cycles. Toxin A induced haemorrhage in both ileal and colonic tissues. In ileum, the villus architecture was severely damaged and this gave rise to protein-rich bloody luminal fluid. In the colon, although colonocytes were removed, the basement membrane remained intact; this resulted in a tissue-localised haemorrhage and a protein-low watery ultrafiltered luminal fluid. Toxin A is thus a novel type of histotoxic enterotoxin.

Analysis of latex agglutination test for Clostridium difficile toxin A (D-1) and differentiation between C difficile toxins A and B and latex reactive protein

Virulent toxigenic and avirulent non-toxigenic strains of Clostridium difficile gave a positive result in the latex agglutination test (LAT) for C difficile toxin A (D-1). Similar concentrations of latex agglutinating antigen were produced by these strains in vivo. Positive reactions were also given by C sporogenes, proteolytic C botulinum Types A, B, and A/F, and Bacteroides assaccharolyticus. The latex agglutinating antigen was denatured by boiling for 10 minutes, but not by heating at 56 degrees C for 30 minutes. The reaction was abolished by incubation of test material with crude C difficile antitoxin but not with other clostridial antitoxins or specific antitoxin to C difficile toxin A. The latex agglutinating antigen present in C difficile eluted between 0.39% and 0.47% M sodium chloride, and that produced by the other clostridia, between 0.35% and 0.43% M sodium chloride by fast protein liquid chromatography. The latex agglutinating antigen of C difficile was neither cytotoxic nor mouse lethal and was distinct from toxin A and toxin B. In the analysis of faecal specimens from patients with diarrhoea the latex agglutination test correlated better with the presence of C difficile than with toxin B and detected both toxigenic and non-toxigenic strains. The latex agglutination test should only be used in the laboratory as an alternative to culture for C difficile and not as a method for the detection of C difficile toxins.

The effects of Clostridium difficile crude toxins and purified toxin A on stripped rabbit ileal mucosa in Ussing chambers

Clostridium difficile crude toxins and purified toxin A had similar effects on stripped rabbit ileal mucosa in Ussing chambers. Both toxin preparations caused secretion of sodium and chloride ions by increasing serosa to mucosa (s----m) fluxes. Transmural potential difference and resistance decreased after toxin treatment. Onset of changes in electrical measurements and ion fluxes coincided with onset of histological changes. The response to theophylline was greatly reduced in toxin-treated tissue compared with control tissue.

The effects of Clostridium difficile toxins A and B on membrane integrity and protein synthesis in intestinal cells in vivo and in vitro and in McCoy cells in vitro

Clostridium difficile toxins A and B inhibited protein synthesis in McCoy tissue-culture cells but not in intestinal cells in vitro or in vivo. Toxins A and B had no effect on membrane permeability of either intestinal cells or McCoy cells.

Sporogenesis and toxin A production by Clostridium difficile

The kinetics of spore production by Clostridium difficile were not paralleled by release of C. difficile toxin A in vitro. Toxin A was not found to be associated with either purified whole spores or spore coats. Residual traces of toxin A detected in spore contents were almost certainly derived from contaminating vegetative cell debris. Thus, toxin A is unlikely to be a spore constituent or associated with sporogenesis.

Pharmacokinetics of single dose intravenous ciprofloxacin in patients undergoing gastrointestinal surgery

The pharmacokinetics of single dose intravenous ciprofloxacin in patients undergoing gastrointestinal surgery (100 mg, n = 8; 200 mg n = 18) have been studied. Following 100 mg therapeutic serum levels were maintained for approximately 40 min only and tissue concentrations were frequently less than 0.15 mg/kg. 200 mg maintained therapeutic serum levels for at least 150 min and produced mean concentrations in fat, muscle, peritoneum and gut wall of 1.04, 1.94, 1.59, and 3.39 mg/kg respectively. 200 mg iv ciprofloxacin would appear to provide adequate serum and tissue concentrations for at least 150 min.

Growth of Clostridium difficile and production of toxins A and B in complex and defined media

The ability of several strains of Clostridium difficile to grow and to produce toxins A and B in complex and defined culture media has been studied with special reference to the amino-acid composition of the medium. The production of these toxins varied with the strain used and with the composition of the growth medium. Toxin A production was not inextricably linked to production of toxin B since conditions were found in which only one or other toxin was produced.

Bacteraemia in diabetics

During a period of 22 months one or more episodes of bacteraemia were detected in 168 patients in hospital. Of these, 29% also had diabetes compared with 10% of the total number of patients admitted to hospital during this time (P less than 0.001). The diabetics with bacteraemia were elderly and diabetes had usually been present for many years. Most of them were not receiving insulin at the time bacteraemia was diagnosed. Escherichia coli was the commonest pathogen (33%) in the diabetics, the main source of infection being the urinary tract. This finding may be due to diabetic autonomic neuropathy, which leads to a poorly emptying, chronically infected bladder. Urinary tract infections with bacteraemia in elderly diabetics are often accompanied by vague non-specific symptoms and poor diabetic control. Fever is infrequent. Prompt antibiotic therapy and insulin injections to control the diabetes usually cure these serious infections.

Oral neomycin and erythromycin compared with single-dose systemic metronidazole and ceftriaxone prophylaxis in elective colorectal surgery

A prospective randomized trial was performed to compare oral neomycin and erythromycin with single-dose intravenous metronidazole and ceftriaxone in elective colorectal surgery. The study was discontinued after 60 patients were entered. The overall rate of infection was 41 percent in the oral neomycin and erythromycin group (n = 29) compared with 9.6 percent in those who received intravenous metronidazole and ceftriaxone (n = 31) (p less than 0.01). Infections in the oral group were principally due to resistant Staphylococcus aureus, Bacteroides fragilis, and Escherichia coli. Preoperative administration of oral neomycin and erythromycin was associated with a significant reduction of Escherichia coli counts (1 X 10(7) to 3 X 10(5) organisms/ml, p less than 0.05) compared with the intravenous group, but there was no significant reduction in the counts of Bacteroides fragilis (2 X 10(8) to 1 X 10(7) organisms/ml) and there was an increase in the counts of Clostridia (2 X 10(4) to 1 X 10(6) organisms/ml). These results indicate that single-dose systemic prophylaxis with appropriate antibiotics is superior to oral neomycin and erythromycin.

Effect of toxin A and B of Clostridium difficile on rabbit ileum and colon

The effect of purified toxin A and partially purified toxin B on rabbit ileum and colon was investigated. Toxin A caused tissue damage which was followed by permeability changes and fluid accumulation in both tissues. Toxin A did not increase the permeability of the colon to the extent observed for ileum; secreted fluid contained less protein of plasma origin. Toxin B had no effect on either tissue. Secretory and tissue damaging properties of crude C difficile toxins were found to be due to toxin A.

Influence of rectal washout on bacterial counts in the rectal stump

Forty-three patients undergoing a restorative colorectal resection were randomized to have their rectal stump washed out with either 0.9 per cent saline, 2.5 per cent povidone-iodine or 0.3 per cent sodium hypochlorite. The bacterial counts before and after washout for Escherichia coli and Bacteroides fragilis were compared using the Mann-Whitney U test. There was no significant difference in the counts for E. coli before and after washout with saline (log 9.7-log 9.7 organisms/ml) and saline had no influence on the counts of B. fragilis (log 7.9-log 6.5 organisms/ml). Hypochlorite was associated with a significant reduction in E. coli counts (log 6.1-log 1.1 organisms/ml, P less than 0.005) and the counts of B. fragilis (log 8.1-log 0.0 organisms/ml, P less than 0.005). The counts of B. fragilis were significantly reduced by povidone-iodine (log 8.1-log 1.1 organisms/ml, P less than 0.005) but there was no significant reduction in counts of E. coli (log 5.6-log 5.3 organisms/ml). The data suggest that sodium hypochlorite is better than povidone-iodine or saline as a rectal washout for colorectal anastomosis.

The influence of single dose intravenous antibiotics on faecal flora and emergence of Clostridium difficile

The influence of a single intravenous dose of antibiotic on faecal flora and the emergence of Clostridium difficile was studied in volunteers. Seventy-eight volunteers (13 groups of 6 receiving 5 penicillins and 8 cephalosporins) were given a single intravenous dose of antibiotic. Results were compared with a control group of 6 volunteers who did not receive an antibiotic. Changes in the faecal flora were monitored over two weeks. Only cephalosporins were associated with emergence of Cl. difficile, penicillins and controls were not. Frequency of emergence of Cl. difficile in the 6 volunteers in each of the cephalosporin groups were as follows: cephaloridine 0, cephazolin, 1 cefuroxime 1, cefoxitin 2, cefotaxime 2, latamoxef 3, ceftriaxone 2, cefotetan 4. Latamoxef (moxalactam) was associated with a significant rise in total aerobic bacterial counts largely due to increased counts of enterococci. Cefotetan eliminated Escherichia coli and Bacteroides fragilis from the faecal flora three days after administration.

Antibiotic therapy for treatment in relapse of intestinal Crohn's disease. A prospective randomized study

We have undertaken a prospective randomized trial of one month's antimicrobial therapy for patients with symptomatic relapse of Crohn's disease. Criteria for entry included two major symptoms: fever, abdominal pain, diarrhea, weight loss, abdominal mass or complications (excluding perianal disease); and two hematologic abnormalities: hemoglobin, ESR, albumin, C reactive protein, iron, or total iron binding capacity. Patients were monitored for the aforementioned clinical and hematologic (hemoglobin, albumin, CRP) parameters over six weeks and for changes in fecal flora. Randomization was to four groups: metronidazole alone (M), cotrimoxazole alone (C), metronidazole and cotrimoxazole (C plus M), or double placebo (P). Seventy-two patients entered the study (18 = M, 16 = C, 21 = C plus M, 17 = P). After two weeks, improvement was reported as follows: M = 67 percent, C = 17 percent, C plus M = 71 percent, P = 35 percent. In the metronidazole group, two patients required surgery and one had troublesome side effects. In the cotrimoxazole group, two had side effects. In the combined group (C plus M), four had troublesome side effects and two of the placebo group (P) required operation. By four weeks, there was no difference in response among the groups: (M = 44 percent, C = 62 percent, C plus M = 57 percent, P = 41 percent). Antimicrobials had no effect on fecal flora or hematologic parameters. These results indicate that antimicrobials have little therapeutic potential for relapse of intestinal Crohn's disease.

The effect of a single intravenous dose of cefotaxime on the faecal flora

Cefotaxime had an immediate effect on the faecal flora, mainly involving a loss of gram-positive organisms, but also reducing the number of Escherichia coli, Bacteroides spp. and Clostridium spp. There was little effect on the total bacterial counts because of compensatory increases in other species. These changes appear to initiate a period of instability in the composition of the faecal flora lasting for at least 14 days after the administration of cefotaxime. Other third-generation cephalosporins showed a similar prolonged disturbance to the stability of the faecal flora, but there were minor differences from cefotaxime in the bacterial species initially affected.

A high incidence of bleeding is observed in a trial to determine whether addition of metronidazole is needed with latamoxef for prophylaxis in colorectal surgery

A prospective randomized trial has compared the use of latamoxef sodium (two doses) with latamoxef and metronidazole for elective colorectal surgery. The incidence of wound infection in patients receiving latamoxef alone was 34 per cent compared with 32 per cent in patients receiving latamoxef and metronidazole. Only eight of the 36 wound infections in this study could be classified as major. Only two patients developed an intra-abdominal abscess postoperatively and there was only one episode of septicaemia. Postoperative haemorrhage was recorded in 17 patients (15 per cent). Twelve episodes of bleeding occurred in the first 97 patients who entered the trial and prolongation of the prothrombin time was recorded in eight of 16 patients. In view of these findings 10 mg vitamin K was given with each dose of latamoxef to the last group of patients. However, bleeding occurred in five of 13 patients receiving vitamin K and entry to the study was therefore discontinued.

Treatment of pseudomembranous colitis and antibiotic-associated diarrhoea

Pseudomembranous colitis is caused by release of toxins from Clostridium difficile when it colonizes the large intestine. This clostridium is susceptible to concentrations of vancomycin which are readily attained in the colon after oral administration. When vancomycin is given orally to infected patients in a dose of 125 mg every 6 h, a rapid clinical cure can be expected. Some patients may relapse after the vancomycin is stopped, but a further course of treatment will control symptoms.

Production and release of toxins A and B by Clostridium difficile

The production and release of toxins A and B by Clostridium difficile during in-vitro culture was investigated. Cell-associated toxin A was detected by immunoelectrophoresis of bacterial extracts released by ultrasonication and by fluorescent antibody labelling of whole cells. Extracellular toxin A was detected by immunoelectrophoresis and by enzyme-linked immunosorbent assay; extracellular toxin B was detected by cytotoxin assay. Both toxins A and B were produced and released during the decline phase of the bacterial growth cycle. The possible significance of these results in relation to the pathogenesis of pseudomembranous colitis is discussed.

Incidence of pathogenic bacteria from mesenteric lymph nodes and ileal serosa during Crohn's disease surgery

Samples of ileal serosa and mesenteric lymph nodes have been harvested before antibiotic administration during 46 non-contaminated operations for Crohn's disease and compared with 43 operations for conditions other than Crohn's. Potentially pathogenic bacteria were isolated from the serosa in 12 (27 per cent) Crohn's patients, compared with 6 (15 per cent) controls (P = 0.04). Intestinal bacteria were recovered from mesenteric nodes in 15 (33 per cent) Crohn's patients compared with 2 (5 per cent) controls (P = 0.006). These findings suggest that bacteria leak from the small bowel lumen in a high proportion of Crohn's disease patients. This may explain the pathogenesis of abscess and fistula in this disorder as well as the high rate of sepsis following elective surgery even in the absence of macroscopic contamination.

Mutagenicity in gastric juice

Mutagenicity has been measured in the gastric juice of 228 patients using the Ames bacteriological test system; while mutagenicity in control and duodenal ulcer patients did not differ from saline controls, mutagenicity was significantly increased compared with controls in patients suffering gastric ulcer (p less than 0.002), carcinoma (p less than 0.002), and in patients after gastric resection (p less than 0.01). A transient rise in mutagenicity was seen following H2 antagonist ingestion (p less than 0.002). Increased levels of mutagenicity were found to correlate closely with gastric juice pH and bacterial count. Histidine concentrations in gastric juice did not explain the mutagenicity results.

Prospective randomized trial of single dose cefuroxime against mezlocillin in elective gastric surgery

A prospective randomized control trial was made in 78 patients undergoing elective operations on the stomach where the viscus was opened at operation, and the efficacy of single dose intravenous prophylaxis with cefuroxime 1.5 g or mezlocillin 2 g was compared. The overall rate of sepsis was 10.2 per cent. Infection in the cefuroxime group was significantly lower (2.5 per cent) than in the mezlocillin group (18 per cent) (P less than 0.05). Four of the seven patients with infections in the mezlocillin group were due to antibiotic resistant staphylococci. The only infections in the cefuroxime group were due to an antibiotic resistant strain of Pseudomonas aeruginosa. One case of pseudomembranous colitis occurred in the cefuroxime group. In view of the high rate of resistant organisms in patients receiving mezlocillin we believe that cefuroxime remains the antibiotic of choice for patients undergoing elective gastric operations.

Intragastric N-nitrosation is unlikely to be responsible for gastric carcinoma developing after operations for duodenal ulcer

Three groups of patients studied after operations which had cured their duodenal ulcer were compared with a control group (no operation, n = 8). The surgical procedures included: proximal gastric vagotomy (n = 7), truncal vagotomy and pyloroplasty (n = 7), truncal vagotomy and antrectomy (n = 8). Samples of gastric juice were aspirated half hourly or hourly over 24 hours for measurement of pH, counts of all identified bacteria, nitrite and total N-nitrosocompounds. Although the pH over 24 hours was significantly higher after proximal gastric vagotomy (p less than 0.05) and truncal vagotomy and antrectomy (p less than 0.001) than controls, there was no difference between truncal vagotomy and pyloroplasty and controls. Counts of nitrate reducing bacteria over 24 hours were also significantly higher after truncal vagotomy and antrectomy than controls (p less than 0.1) but no differences were observed between the other groups. Only after truncal vagotomy and antrectomy was nitrite over 24 hours significantly increased compared with controls (p less than 0.01). Despite these higher values after truncal vagotomy and antrectomy, there was no significant difference in total N-nitrosocompounds between any of the four groups. Whereas bacterial counts and nitrite increased with pH, no correlation was found between total N-nitrosocompounds and pH. These results provide no evidence that exposure to total N-nitrosocompounds is increased after operations for duodenal ulcer.

A prospective randomized trial to compare mezlocillin and metronidazole with cefuroxime and metronidazole as prophylaxis in elective colorectal operations

A prospective randomized trial has compared a broad spectrum ureidopenicillin with a broad spectrum cephalosporin for prophylaxis against the aerobic organisms encountered during elective colonic surgery. Even though only two doses of antibiotics were administered the incidence of severe sepsis was low. Severe wound infection occurred in three of the patients receiving mezlocillin and metronidazole (6 per cent) compared with six in the group receiving cefuroxime and metronidazole (13 per cent). Minor wound sepsis was recorded in 24 per cent of patients receiving mezlocillin and metronidazole compared with only 11 per cent after cefuroxime and metronidazole. There were two episodes of septicaemia, one in each group, and three abscesses, all of which occurred in patients receiving metronidazole and mezlocillin. The total number of surgically related infections was, however, significantly less with cefuroxime and metronidazole (N = 13) compared with mezlocillin and metronidazole (N = 23; P less than 0.03). Escherichia coli was the principal organism responsible for surgically-related postoperative sepsis: (22 isolates: 14 mezlocillin and eight cefuroxime) all of which sensitive to the agents used. Pseudomonas aeruginosa was recovered from 10 patients (three mezlocillin and seven cefuroxime), all of the isolates were resistant to both antibiotics and were associated with severe morbidity. There were 11 isolates of Staphylococcus spp. (nine mezlocillin and two cefuroxime: P less than 0.03). Postoperative diarrhoea occurred in six patients, all were in the group receiving cefuroxime and metronidazole. (Clostridium difficile was recovered from the stool in three of which one was associated with Cl. difficile cytotoxin.)

A physiological appraisal of polyethylene glycol and a balanced electrolyte solution as bowel preparation

Seventy-seven patients undergoing bowel preparation with either sennosides and rectal washout (S + RWO; n = 26) or polyethylene glycol and a balanced electrolyte solution (PEG + E) given orally (n = 25) or by nasogastric infusion (n = 26) have been studied. PEG + E was not associated with significant physiological change in electrolytes or blood gases. Patients tolerated PEG + E well and it provided a moderate to good bowel preparation in over 80 per cent of cases. PEG + E was not associated with a change in colonic microflora and was not associated with explosive mixtures of colonic gas due to hydrogen. It is therefore a safe, quick and efficient bowel preparation and provides a safer alternative to mannitol. The solution (PEG + E) man be used orally or by nasogastric infusion and does not require an accompanying intravenous infusion.

Clinical and microbiological evaluation of piperacillin in elective biliary surgery

Fifty patients undergoing electric biliary surgery were given piperacillin 2 g i.m. 2 h preoperatively followed by 2 g i.v. at the beginning of the operation. Venous blood and bile from the gall bladder and common bile duct were assayed for piperacillin. Blood levels exceeded 100 mg 1(-1) in all instances unless the protocol was not followed. Both gallbladder and common bile duct bile levels exceeded 50 mg 1(-1), and the only exceptions were due to poor penetration into three obstructed gallbladders. Common bile duct levels exceeded 50 mg 1(-1) even in the presence of obstructive jaundice. Only one patient developed a minor postoperative wound infection.

Evaluation of mezlocillin in elective gastrointestinal surgery

We report the interim results of two trials to evaluate the place of mezlocillin in elective intestinal operations. Sixty-four operations for gastro-oesophageal disease have been performed where patients were randomly allocated to mezlocillin or cefuroxime. Wound sepsis occurred in 19% of the mezlocillin patients compared with 3% of those receiving cefuroxime. Seventy-three operations have been performed for colorectal cancer in whom three doses of antibiotic were used for prophylaxis. Patients were randomly allocated to mezlocillin and metronidazole or cefuroxime and metronidazole. The rates of abdominal wound sepsis in the groups were 15% and 12% respectively. Post-operative Clostridium difficile colitis occurred in four of the cefuroxime patients, compared with none of those receiving mezlocillin. Mezlocillin appears to be a safe and effective antibiotic when used in combination with metronidazole in colorectal resections for cancer.

Metronidazole need not be combined with an aminoglycoside when used for prophylaxis in elective colorectal surgery

A prospective randomized controlled study has compared prophylactic intravenous metronidazole alone with the combination of metronidazole and gentamicin in 100 patients having elective colorectal surgery. There was no significant difference in mortality, infection, dehiscence or postoperative stay between the antibiotic groups. Postoperative sepsis was almost entirely due to aerobic organisms but was associated with minimal morbidity. A significantly higher rate of anastomotic dehiscence and postoperative infection was seen where the quality of bowel preparation was judged to be poor.

Incidence and microbiology of abdominal and pelvic abscess in Crohn's disease

The incidence of abdominal and pelvic abscess is reported from a consecutive series of 111 patients undergoing 124 resections for Crohn's disease. Preoperative abscesses were found as a complication of Crohn's disease in 13 patients (10%) and 8 were clinically unsuspected. The majority of preoperative abscesses were confined to one site (localized to bowel, psoas sheath, pelvis, or in the abdominal wall). All preoperative abscesses occurred in patients requiring an emergency or urgent operation. Patients with a preoperative abscess had significantly lower serum albumin levels and significantly increased serum alkaline phosphatase values than the patients without an abscess. Postoperative abscesses occurred in 17 patients (14%) and six were multiple. Five of the postoperative abscesses occurred in patients who had had a preoperative abscess; these recurrent abscesses all presented 6-14 wk after an uncomplicated initial operation. The principal bacterial isolates were Escherichia coli (54%), Bacteroides fragilis (44%), enterococci (41%), and viridans streptococci (31%). The incidence of abscess was unrelated to the use of preoperative steroid therapy.