Virological examination of 385 patients with suspected heart disease and 26 with Bornholmdisease over a period of 6 years suggested that Coxsackie group B virus infections were associated with at least half the cases of acute myocarditis and one third of the cases ofacute non-bacterial pericarditis. Complement-fixation tests revealed only a few cardiac illnesses associated with other infections (influenza and Mycoplasma pneumoniae). No evidence of infection was found in chronic cardiac disease.

Seven hundred and sixty-eight patients were seen and tested at frequent intervals after transfusion of whole blood. Eight patients were judged to have developed icteric or anicteric post-transfusion viral hepatitis, an incidence of 1%. Five were icteric and four of these were hepatitis B antigen (HB Ag) positive; two of these four died. One of the fatal cases and one non-fatal HB Ag positive case had received HB Ag positive blood. Two other antigen-positive patients had received blood or plasma or both which had not been tested for antigen.

Thirty-five patients showed conspicuous or sustained elevations of alanine transaminase without clinical features of hepatitis.

Four were positive for HB Ag but had not received antigen positive blood.

Two who had received antigen positive blood remained antigen negative, but one developed hepatitis B antibody (HB Ab).

Two other patients were also transfused with plasma.

Five had serological evidence of cytomegalovirus (CMV) infection accompanying the enzyme changes.

One patient who had received HB Ag positive blood remained antigen-negative and showed no abnormalities.

Five patients who became HB Ag positive, although they had been given antigen-negative blood, remained clinically and biochemically well.

Cytomegalovirus primary infection or reactivation occurred in another 32 patients; five had isolated, transient enzyme rises, one other was associated with a drug-induced focal liver necrosis and 26 showed no enzyme changes. Epstein–Barr virus infections, one of which was associated with a transient upset of enzyme activity, were detected in five patients. There were no cases of post-perfusion syndrome.

Since the earliest days of disinfectant testing bacteriostatic effect has misled many workers in this field. This problem has long been appreciated by some and a considerable battery of neutralizers has been employed, often with great success. Other anti-bacteriostatic measures, used without control, fail to revive damaged though viable organisms.

This paper sets out to describe some of the problems that are encountered in this confused field and the means whereby they may be overcome.

In both mice and one patient, a similar correlation was found between the oral dose of a ‘nonabsorbable’;aminoglycoside antibiotic selected for antibiotic decontamination of the digestive tract and the resultant faecal concentration. In mice, absorption of the orally-given antibiotics could only be demonstrated in animals treated with extremely high doses of 1440 mg. per kg. body weight per day. Evidence of absorption of gentamycin was found to occur in a patient after doses as low as 70 mg. per kg. per day.

The 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.

During a period of five years 1192 water samples from swimming pools were examined for staphylococci and 338 for coliform organisms only. Eighty-nine different pools were sampled.

Numbers of staphylococci, estimated by the membrane filtration technique did not bear any significant relation to either bathing load or concentration of free chlorine.

Wide variation in the staphylococcal count was observed when different parts of a pool were sampled on the same occasion.

The only practicable standard for pool samples in relation to staphylococci would appear to be that these organisms should be absent from 100 ml. water when the pool has been out of use during at least ten hours before sampling if filtration and chlorination are adequate.

Studies on brucellosis were carried out to elucidate the epidemiology of the disease in Isfahan province, where Brucella melitensis is highly prevalent in animals and man. A positive milk ring test of 32% and 23% was found in unvaccinated goats and sheep respectively. Card and tube agglutination tests showed an infection rate of about 12% in sheep and goats and 42% in cattle. B. melitensis was isolated from 8% of 677 samples of fresh cheese examined.

Of 1526 clinically suspected human cases, 476 showed laboratory evidence of brucellosis. Of these patients, 291 cases were from urban and 185 cases from rural areas. Cases from urban and rural areas were seen principally in the younger age groups. The median age of infection was 19·7 in urban and 15·7 in rural patients respectively. The infection was encountered mainly from April to August. This correlates with animal parturition and the greatest amount of sheep and goat milk production, which is introduced to the local market as fresh cheese. Raw dairy product consumption is the most probable way of Brucella transmission in urban patients. In rural areas, both dairy product consumption and contact with animals are sources of infection.

A comparison was made of the incidence of foot-and-mouth disease virus 'carrier' cattle in an unvaccinated enzootic area and an area where routine 6-monthly vaccination with an inactivated vaccine had been carried out for 3–4 years. The incidence of carriers in the vaccinated area was 0·49% as compared to 3·34% in the non-vaccinated area. The results indicate that, provided the immune status of the vaccinated herd is maintained at a level sufficient to prevent outbreaks of clinical disease and the re-introduction of virus is prevented through livestock movement controls, it should be possible to eradicate the disease from an enzootic area through vaccination.

The subtype variants found in Kenya in the past ten years have been studied. The type O and type SAT 2 subtypes have a distinct geographical distribution which appears to be associated with livestock movement patterns. The type A viruses have a greater tendency to antigenic variation and their geographical distribution is less distinct. In type C only minor differences exist between the three viruses studied.

The distribution and serotype of strains of Escherichia coli from a poultry packing station and an abattoir are described. The results indicated that animal faecal strains contaminated the environment and the animal carcasses.

Using 150 O antisera, a high proportion of the E. coli strains were non-typable. This suggests that the serotype distribution of E. coli in animals is different from that in man.

Strains with single antigenic differences were isolated, and the possibility of genetic transfer of these antigenic structures is suggested.

The incidence of infections caused by Pseudomonas aeruginosa did not increase significantly among general surgical and medical patients between 1967 and 1972, and the majority of such infections were trivial. Serious infections were virtually confined to the intensive care unit and the renal transplant unit, and were usually associated with major trauma, surgical mishap or immunosuppression. The majority of these patients had received prior antibiotic therapy. Persistent isolation of Ps. aeruginosa from surgical wounds was often associated with severe intra-abdominal sepsis, and antibacterial therapy was commonly ineffective in these cases. Apparently susceptible patients did not necessarily acquire infection, though the organism was present in their environment. It is suggested that this may reflect a variation of virulence among environmental strains of Ps. aeruginosa, and that further study of this aspect may contribute to improved control of infection.

Enterotoxin production by strains of Staphylococcus aureus isolated from routine samples of foods and from human beings was investigated. Twenty-one to 26% of 112 strains isolated from raw meat, sausages and poultry and 32–36% of 183 strains isolated from cooked foods, e.g. meat, chicken and frozen seafoods, produced enterotoxins A, B, C, D or E. Staph. aureus isolated from raw meat and chicken less frequently produced enterotoxins A, B, C or E and more frequently enterotoxin D, than those from cooked meat and seafoods. Of the 113 strains isolated from cheese and raw milk 6–11% produced enterotoxin and most of these produced enterotoxin D. Only a few strains isolated from foods produced enterotoxin E. Results of enterotoxin tests on Staph. aureus from human beings resembled those on strains from cooked foods.

A dengue epidemic due to type 2 virus involving some 3,400 cases began in Fiji early in 1971, had a peak during May, June and July, and cases have continued to occur with a low incidence during 1972 and 1973. Many of the notified cases showed classical dengue fever symptoms and there were no confirmed cases of haemorrhagic fever. A serological survey indicated that there had been at least 20,000 subclinical infections. It is probable that the virus was introduced to Fiji either through the port of Lautoka or Nadi international airport in February 1971. The normal travel patterns of residents must have spread the virus to all the more accessible localities but, with the exception of Rotuma, it caused infections only in areas where Aedes aegypti was available as a vector. There was no evidence that pre-existing dengue type 1 serum antibody gave any protection during this epidemic.

Bacteriological contamination of a specific-pathogen-free (SPF) animal production building was more extensive after one year of operation than at the time of commissioning. Throat and nose swabs taken from the staff yielded a number of undesirable organisms, in particular, on one occasion, a Group A streptococcus. It is recommended that a closer bacteriological scrutiny of the staff of SPF production buildings ought perhaps to be made in the light of these findings.

Previous work showed that on the basis of O serotyping alone of Escherichia coli, the majority of babies acquired the same O serotype as was found in the stools of their respective mothers. Further characterization of the E. coli by H serotyping, determination of their antibiotic resistance and ability to ferment six carbohydrates showed that in the majority of cases the previous results were confirmed. In a minority of cases this further testing showed that the strains were not identical. In some instances a number of strains isolated from the same pair showed different combinations of the markers used.

Pre- and post-vaccination serum samples from 278 volunteers, who were given live influenza vaccines, were tested by haemagglutination inhibition (HI) and single radial immunodiffusion tests (SRDT) for antibody to influenza A/Hong Kong/1/68 (H3N2) virus envelope antigens. Those with high antibody titres detected in both tests were less frequently infected, and 85% of the 159 infected showed rises by HI and 70% by SRDT. Similarly, 69 pairs were tested for antibody to Hong Kong (N2) neuraminidase by neuraminidase inhibition test (NI) and by SRD tests. Those with high titres in both tests resisted infection and those who were infected showed a rise in antibody detected both by NI and SRD tests. In general, SRDT was less sensitive than HI and NI in detecting antibody and antibody rises, but in some volunteers it did detect antibody rises which were not detected by conventional tests. Because of its simplicity and speed it appeared to be of use in evaluating such vaccines.

Infant feeds, to be safe, must be free from potentially pathogenic organisms but not necessarily sterile. In-bottle terminal heating is the preferred means of producing such feeds and the advantages and disadvantages of high and low pressure heating methods, including the effect upon the food value of the feeds, are discussed. The safety of a low pressure method in use in Princess Alexandra Hospital, Harlow is described. The choice of method of provision of safe feeds; terminal heating in a central milk kitchen or obtaining a commercial supply should be decided on economic grounds. Hospitals using such commercial supplies, however, should make provision for training mothers in the hypochlorite method of disinfection of bottles and teats and also ensure that their trainee nurses and midwives still receive adequate instruction in the hygiene of feed preparation.

The four main serotypes of Bordetella pertussis (1, 2, 3; 1, 2; 1, 3; 1) undergo spontaneous variation involving loss or gain of antigen 2 or antigen 3. By serial subculture from single colonies on charcoal-blood-agar medium, we have detected loss-mutations from type 1, 2, 3 to 1, 2 or 1, 3, and from type 1, 2 to type 1. Likewise we have found gain-mutations from type 1 to 1, 2 or 1, 3, and from 1, 2 to 1, 2, 3.

These mutations apparently occur with a high frequency in some strains. Other strains have a lower mutation-rate and are more stable antigenically. We have not detected, by this method, either gain- or loss-mutations from the type 1, 3 strains that we have tested.

These findings offer an explanation for the changes in serotype that occur during the course of a pertussis infection in the child and in the marmoset. They also constitute a warning on the possible antigenic instability of laboratory strains, especially relevant in the production, absorption and testing of diagnostic antisera and in the preparation of pertussis vaccine.

In a maternity hospital in which the umbilicus and trunk of healthy newborn infants were treated with 0·33% hexachlorophane dusting powder, the hexachlorophane content of blood was measured in mothers before delivery, in infants' umbilical samples at birth, and at 8 days of age in capillary blood samples. One mother and her baby had rather high blood concentrations of hexachlorophane, probably derived from a toilet preparation used before admission to hospital. Hexachlorophane was absent or barely detectable in the other mothers' blood and in the infants' umbilical blood. The hexachlorophane concentrations in the blood of 8-day-old infants ranged from nil to 0·166 μg./ml. (mean 0·066 μg./ml.). These were much less than the concentrations reported to be toxic in animals.

In a previous trial now reported here, a dusting powder containing chlorhexidine instead of hexachlorophane was found to delay the separation of the umbilical cord.

A trial of influenza A vaccines in general practice is described. Five hundred and seven subjects were vaccinated with either inactivated monovalent A/Hong Kong vaccine, A/England vaccine or influenza B vaccine as control. Local reactions were noted in 24% and general reactions in 12% of patients. Antibody titres in serum were measured by haemagglutination inhibition (HI) and complement fixation (CF) tests in 465 subjects. The influenza vaccines produced substantial increases in both homologous and heterologous antibodies as measured by the HI test and a comparatively poor response as measured by the CF test. Although clinical influenza was confirmed in only a few cases, there was serological evidence of significant subclinical infection in the control group.