The impact of laboratory closing times on delay of adequate therapy in blood stream infections

BACKGROUND

Patients with bloodstream infections need early adequate antimicrobial treatment to reduce mortality. This raises the question of timing and logistics. How important is the time of day when a culture is flagged positive to the processing of blood cultures and optimisation of antimicrobial therapy?

METHODS

We performed a retrospective study assessing the time delay of a positive blood culture result during and after office hours and its impact on adequate antimicrobial therapy. Process duration from the moment of culture positivity to Gram stain completion was compared at different timepoints during the day in a medium-sized hospital with an offsite microbiological laboratory.

RESULTS

Ninety-four patients with positive, noncontaminated blood cultures were included. Sixty-six patients (70%) received adequate empirical therapy; this increased to 76 cases (82%) and to 88 cases (95%) after analysis of Gram stain results and complete determination, respectively (p < 0.05 for all comparisons). Median duration from culture positivity to Gram stain completion (including offsite culture transport) increased from a median of four to 12 hours if time of cultures turned positive after office hours (p < 0.05), irrespective of the adequacy of empirical coverage. This also resulted in a median 12-hour delay for the complete process from time of culture positivity to administration of the antimicrobial drug (p < 0.05).

CONCLUSION

Processing blood cultures after office hours is often deferred, which can lead to a delay in adequate antimicrobial therapy for patients with bloodstream infections.

[Shigellosis, a sexually transmitted disease in homosexual men]

A study of the registered number of Shigella infections in Amsterdam for the period 1997-2001 revealed that in 2001, there was an increase in the number of homosexual patients with oral-anal contacts as source for their infections. All of these infections were caused by subtype Shigella sonnei. The infections were not associated with a specific meeting place for homosexual men. A relationship with an underlying HIV-infection is likely.

[Increased Neisseria gonorrhoeae quinolone resistance in Amsterdam]

In addition to a rise in the number of cases of gonorrhoea, the susceptibility of Neisseria gonorrhoeae to antibiotics is also a cause for concern. After a period of high resistance rates to penicillin and tetracycline between 1985 and 1995, resistance rates have dropped considerably in recent years, probably due to changes in treatment regimens. However, recently we have seen an increasing number of quinolone-resistant N. gonorrhoeae isolates in Amsterdam, the Netherlands, a development that has previously been reported in other parts of the world. Some form of national resistance monitoring for gonococci is therefore urgently required to allow timely detection of changes in N. gonorrhoeae resistance.

Pneumococcal carriage in children in The Netherlands: a molecular epidemiological study

In 1999, Engelen and coworkers investigated colonization in Amsterdam among 259 children attending 16 day-care centers (DCCs) and among 276 children who did not attend day-care centers (NDCCs). A 1.6- to 3.4-fold increased risk for nasopharyngeal colonization was observed in children attending DCCs compared with NDCC children, while no difference in antibiotic resistance was found between groups. The serotype and genotype distributions of 305 nasopharyngeal Streptococcus pneumoniae isolates of the latter study were investigated. The predominant serotypes in both the DCC and the NDCC groups included 19F (19 and 18%, respectively), 6B (14 and 16%, respectively), 6A (13 and 7%, respectively), 23F (9 and 7%, respectively), and 9V (7 and 7%, respectively). The theoretical vaccine coverage of the 7-valent conjugate vaccine was 59% for the DCC children and 56% for the NDCC group. Genetic analysis of the pneumococcal isolates revealed 75% clustering among pneumococci isolated from DCC attendees versus 50% among the NDCC children. The average pneumococcal cluster size in the DCC group was 3.8 and 4.6 isolates for two respective sample dates (range, 2 to 13 isolates per cluster), while the average cluster size for the NDCC group was 3.0 (range, 2 to 6 isolates per cluster). Similar to observations made in other countries, these results indicate a higher risk for horizontal spread of pneumococci in Dutch DCCs than in the general population. This study emphasizes the importance of molecular epidemiological monitoring before, during, and after implementation of pneumococcal conjugate vaccination in national vaccination programs for children.

Diagnosing genital ulcer disease in a clinic for sexually transmitted diseases in Amsterdam, The Netherlands

The most common etiologic agents of genital ulcer disease (GUD) are herpes simplex virus type 1 (HSV-1), HSV-2, Treponema pallidum, and Haemophilus ducreyi. In an outpatient clinic for sexually transmitted diseases in Amsterdam, The Netherlands, specimens from 372 patients with GUD were collected from February to November 1996. Sera were collected at the time of the symptoms and, for most patients, also during follow-up visits. Swabs in viral transport medium were used for HSV culture and for detection of DNA. The most prevalent pathogen found was HSV-2, which was detected by culture in 35% of the patients and by PCR in 48% of the patients. Also, HSV-1 infection was more often detected by PCR (7.8%) than by culture (5.6%). Evidence for an active infection with T. pallidum was found in 1.9% of the patients, using serological tests. A multiplex PCR for simultaneous T. pallidum and H. ducreyi DNA detection was positive for T. pallidum in 3.3% of the samples and for H. ducreyi in only 0.9% (3 out of 368) of the samples. The sensitivity of the PCR was superior to that of culture for HSV detection and to that of serology for T. pallidum detection. Specific H. ducreyi immunoglobulin G antibodies were detected in sera of 5.2% of the patients, with no concordance between serology and PCR. In 37% of the cases, none of the tested microorganisms was detected. Performance of PCR in addition to conventional techniques significantly improved the diagnosis of GUD.

[Shigellosis in Amsterdam, 1991-1998: incidence and results of contact tracing]

OBJECTIVE

To determine the incidence of reported shigellosis in Amsterdam (1991-1998) and the factors related to the secondary attack rate of shigellosis.

DESIGN

Retrospective.

METHODS

From an automated data bank data were collected on all registered shigellosis patients in Amsterdam in the period 1991-1998, and their contacts found by contact tracing.

RESULTS

567 Patients with shigellosis had been reported and a Shigella infection has been diagnosed in 96 of the 983 contacts. The annual incidence decreased statistically significantly from 13 to 8/100.000. 329 males and 334 females were infected. The highest number of cases was found in children from 0-4 years and among adults from 20-34 years. Of the 663 patients 106 (16%) required hospitalization; for children under 7 this proportion was 30% (32/108). Among the contacts (excluding the 14 contacts with the same foreign source and the same first sickness day as the index patient) the secondary attack rate was 8%. Young age, of both the patient and the contact was independently related to a higher secondary attack rate.

CONCLUSION

With contact tracing for shigellosis a large number of secondary infections were found. Given the high infection rate and large number of hospitalisations among children, contact tracing should focus at this young age group.

[Tinea capitis in primary school age children in southeastern Amsterdam: primarily due to Trichophyton tonsurans]

OBJECTIVE

To establish prevalence and aetiology of tinea capitis in children attending primary school in Amsterdam South-East.

DESIGN

Prevalence survey.

METHODS

The Municipal Health Service Amsterdam, the Netherlands, selected classes of four primary schools in a survey with children clinically suspected of having tinea capitis as well as classes without clinical cases. Per school at least one class with a notified case was included. Written information was provided to all parents; the parents of three children refused participation. After clinical examination by a team of physicians and nursing staff scalp cultures were taken using the brush method from 315 children (aged 4-12 years) in 16 classes.

RESULTS

The clinical diagnosis 'tinea capitis' was established in 25 children, significantly more in boys than in girls (relative risk (RR) = 2.92: 95% confidence interval (95% CI) 1.20-7.11). Of all children 7% had positive cultures, more boys than girls (RR = 3.13: 95% CI: 1.18-8.28). Symptomatic cases (confirmed by culture) amounted to 3.2% of all children. The carrier rate was 3.8%. Trichophyton tonsurans was the most frequently isolated dermatophyte both in symptomatic children and in carriers.

CONCLUSION

In 7% of all children (3.2% with symptoms of tinea capitis, 3.8% without) dermatophytes were cultured from scalp samples, the main aetiologic agent being T. tonsurans.

[Direct antibiotic sensitivity testing in the treatment of urinary tract infections in a nursing home]

Urinary tract infections occur frequently in nursing homes. Treatment is usually empiric without microbiological investigation. Because there is widespread resistance to the most frequently used antibiotics, we tested a relatively simple method (direct sensitivity testing for 7 antibiotics) to get information about the antibiotic resistance. The results, with 49 urine samples in the evaluation, showed a concordance of our direct sensitivity test with conventional culture (the golden standard) of 89.8% (35/342 evaluations were not in concordance). The concordance for the result 'sensitive' was 98.8% (3/240 results showed 'false-sensitive'), and the the result 'resistant' 68.6% (32/342 results showed 'false-resistant'). Urinary tract infections in the nursing home can be treated more quickly and more adequately with antibiotics with the information obtained by direct sensitivity testing. This method can be recommended as an alternative to empiric treatment.

Detection of Chlamydia trachomatis and Neisseria gonorrhoeae by ligase chain reaction-based assays with clinical specimens from various sites: implications for diagnostic testing and screening

Ligase chain reaction (LCR)-based tests for the diagnosis of Chlamydia trachomatis and Neisseria gonorrhoeae infections in men and women attending a sexually transmitted disease clinic were evaluated. LCR testing of urethral swab and urine specimens from men and cervical swab and urine specimens from women was compared with culture of male urethral swabs and female cervical and urethral swabs, respectively. An expanded "gold standard" was defined as a positive culture or at least one specimen confirmed to be positive by LCR testing. The prevalence of C. trachomatis infection as detected by cell culture was 7.0% among 614 men and 5.0% among 602 women. By LCR, these values increased to 11.4 and 9.9% with urethral swabs and urine, respectively, for men and 9.6 and 9.1% with cervical swabs and urine, respectively, for women. Relative to the expanded gold standard, the sensitivity of cell culture with male urethral swabs or female cervical swabs was 57.3 and 45.5%, respectively, compared with corresponding values of 93.3 and 87.9% for LCR. The sensitivity of LCR with urine specimens was 77.3 and 78.8% for men and women, respectively. The prevalence of N. gonorrhoeae infection as detected by culture was 5.9% among 220 men and 2.9% among 383 women. The corresponding values were 8.2 and 5.5%, respectively, by LCR testing of swabs. Prevalence values by LCR testing of urine were 7.3% for men and 2.9% for women. The sensitivity of culture was 72.2% for men and 50.0% for women. The sensitivities of LCR were 100% with male urethral swabs, 95.4% with female cervical swabs, 88.9% with male urine, and 50.0% with female urine. These results indicate that the LCR-based assays represent a major improvement in C. trachomatis and N. gonorrhoeae diagnostics. The sensitivity of testing of urethral or cervical swabs by LCR was markedly greater than that by culture. The sensitivity of testing female or male urine specimens was equal to or greater than that of culturing cervical or urethral specimens. LCR testing of urine specimens may prove useful for screening for C. trachomatis.

[An outbreak of methicillin-resistant Staph aureus in a nursing home]

The introduction of methicillin-resistant Staphylococcus aureus (MRSA) in an Amsterdam nursing home (312 beds) by a patient discharged from a local hospital, the spread of this strain to other residents and staff of the nursing home and the actions taken to prevent this spread are described. During the period when this MRSA strain was circulating residents and staff members were checked periodically for presence of MRSA. In spite of all preventive measures as implemented in Dutch hospitals (including isolation of infected residents) additional residents (in a total of 8) and staff members acquired the MRSA strain. Relaxation of the precautions (no more isolation) and continued instruction of the members of the staff was attended by a decrease of further contamination. Four of 8 residents died while being colonised or infected. The course of this epidemic illustrates that limited precautions and proper training of staff in hygiene can restrict spread of MRSA in nursing homes.

A comparison of yield from cervix versus vagina for culturing Candida albicans and Trichomonas vaginalis

OBJECTIVE

To determine the agreement of culture results of Candida albicans and Trichomonas vaginalis from the cervix versus posterior fornix in women with vaginal symptoms.

DESIGN

Same patient comparison of culture results from two sample sites.

SETTING

Twenty one general practices in Amsterdam and the east of the Netherlands.

SUBJECTS

Six hundred and eighty two women aged 15 to 55 years with vaginal symptoms, seen between 1 October 1987 and 31 May 1990.

MAIN OUTCOME MEASURES

For each site (cervix and posterior fornix) the proportion of detected C albicans and T vaginalis. The sensitivity of the cervical swab related to the vaginal one. The percentage of concordance for both microorganisms.

RESULTS

In 248 (34%) women C albicans was diagnosed and in 38 (6%) T vaginalis. In 99% of the proven C albicans cases, the yeast was found in the vagina. In 94% C albicans was isolated from the cervix. Sensitivity of the cervical swab was 94%. In 98% of the patients a concordant observation was made regarding detection of yeast. In 97% of the proven T vaginalis cases the protozoon was found in the vagina. In 91% T vaginalis was detected from the cervical swab. Sensitivity of the cervical swab was 92%. The culture results were concordant in 99%.

CONCLUSION

The yield from the vaginal source was slightly better than that from the cervix for culture of both microorganisms. For screening purposes, specimen-collection for culture of N gonorrhoeae, C albicans and T vaginalis can be combined in one swab taken from the cervix.

[Benefit of the serological screening program for syphilis in pregnant women in Amsterdam in the period 1985-1989]

The results of the serological syphilis screening programme for pregnant women in the region of Amsterdam in the period 1985-1989 are described. In addition, a cost-benefit analysis was done to determine useful effects of the programme. Every GP and obstetrician routinely takes blood samples from pregnant women to determine the blood group, rhesus factor and syphilis serology. In the period 1985-1989, a total of 37,520 serum samples were screened for syphilis in the Amsterdam region by the Central Laboratory of the Blood Transfusion Service (until mid-1986) and by the Regional Public Health Laboratory of the Municipal Health Service (from mid-1986). In 55 women (0.15%) both the Treponema pallidum haemagglutination assay (TPHA) and the venereal disease research laboratory (VDRL) tests were positive. Further evaluation of these 55 women ultimately brought to light 10 women with positive syphilis serology who had not been treated and consequently might infect their child with Treponema pallidum. Considering the various transmission percentages of the bacterium from mother to foetus, which depend on the stage of the maternal infection, the screening programme has prevented two cases of congenital syphilis. This brings the cost-benefit ratio of the syphilis screening programme for the region of Amsterdam to 1:6.2. In view of these findings there is no reason to reconsider the current screening practice in Amsterdam.

Epidemiology of the Bacteroides fragilis group in the colonic flora in 10 patients with colonic cancer

We report the relative frequencies of members of the Bacteroides fragilis group in the faeces, in colon lavage fluid obtained pre-operatively, and in colonic tissue specimens obtained at operation from 10 patients with colonic cancer. B. vulgatus was the most and B. fragilis and B. ovatus were the least frequently isolated Bacteroides spp. in the faeces of the 10 subjects. B. uniformis and B. thetaiotaomicron ranked second and third in the faeces. The relative frequencies of all species except B. fragilis were lower in the lavage fluid and in cultures of mucosa. The relative frequency of B. fragilis increased from 4% in faeces to 39% in the final lavage fluid and to 42% in the colonic mucosa culture. Our results suggest that B. fragilis has a more intimate association with the gut mucosa than other members of the B. fragilis group, which might be one explanation for the high incidence of this species in gut-associated intra-abdominal infections.

Serous otitis media (S.O.M.). A bacteriological study of the ear canal and the middle ear

A bacteriological study of the middle-ear effusions and the ear canals in children with chronic serous otitis media (S.O.M.) was performed. Sixty-eight children (127 ears) were investigated. From this study it appeared that cleansing of the ear canal with 0.5% chlorhexidine in 70% ethanol for 30 s is partially effective; micro-organisms (diptheroids, Staphylococcus epidermidis) could still be isolated in 29%. Cleansing of the ear canal decreases the incidence of middle-ear fluid contamination by non-pathogenic ear canal organisms (diptheroids, Staphylococcus epidermidis, Aerococcus), but after cleansing, 'non-pathogenic' micro-organisms could still be isolated in 33% of the effusions (diptheroids, Staphylococcus epidermidis). From 12% of the middle-ear effusions pathogenic micro-organisms (Hemophilus influenzae, Staphylococcus pneumoniae) were isolated; cleansing of the ear canal did not influence this percentage. Anaerobics were not isolated from the middle-ear effusions.

Comparative activities of five antimicrobial agents in experimental Proteus pyelonephritis in mice

Five antimicrobial agents, ciprofloxacin, ticarcillin, piperacillin, aztreonam and gentamicin, were compared both in vitro (MIC's, time-kill studies) and in vivo, in the treatment of experimental Proteus mirabilis pyelonephritis in mice. In the treatment of the pyelonephritis, ciprofloxacin was clearly superior to the other agents, both with respect to the percentage of sterile kidneys after treatment as with respect to the mean numbers of bacteria per kidney. The results can only be partially explained by the in vitro activities of the different antibiotics.

Application of chromosomal restriction endonuclease digest analysis for use as typing method for Clostridium difficile

The usefulness of restriction endonuclease analysis of chromosomal DNA as a typing method for Clostridium difficile was tested. Over four months all faecal samples were routinely cultured for C difficile. DNA of all isolated strains was isolated and tested with the restriction endonuclease Hind III. The patterns obtained after electrophoresis in agarose gels seemed to be strain specific. Antibiotic susceptibility profiles agreed with the results of the restriction endonuclease analysis, though they were much less discriminating. Analysis of the results indicated that restriction endonuclease analysis is a suitable typing method for C difficile, which may be very valuable in epidemiological studies where a highly discriminating typing method is needed.

Urinary pathogenicity of Proteus mirabilis strains isolated from faeces or urine

Urinary and faecal isolates of Proteus mirabilis were studied with respect to a number of bacterial properties as possible virulence factors in the pathogenesis of urinary tract infections: experimental virulence in a mouse model, haemolysin production, haemagglutinating properties, hydrophobicity of the bacterial surface, sensitivity to the bactericidal effect of human serum, serotype and cell invasiveness. Urinary isolates were slightly more virulent than faecal isolates in the mouse model. No other significant differences were found between both groups. So urinary strains seem to be selected from the faecal reservoir mainly on the basis of their prevalence in the faeces and not on the basis of the possession of particular virulence factors.

Uropathogenic properties of Proteus mirabilis and Proteus vulgaris

A group of faecal isolates of Proteus vulgaris and P. mirabilis was studied for the presence of possible virulence factors such as growth rates in urine and broth, haemolysin production, hydrophobicity, sensitivity to the bactericidal activity of human serum and cell invasiveness. Differences were found in haemolysin production, cell invasiveness and experimental virulence in a mouse model. These differences might explain why P. mirabilis is much more common in human urinary-tract infections than P. vulgaris.

The role of different K antigens of Escherichia coli in phagocytosis by polymorphonuclear leukocytes

The importance of K antigens from six strains of Escherichia coli for the interaction with polymorphonuclear leukocytes (PMNL) was studied. The major factor influencing this interaction was the ability of strains to activate complement by the classical route during opsonisation, this process being reduced for most K-positive strains. Interference of K antigens with the functioning of common pili as adhesions of eukaryotic cells was not observed nor a toxic effect of K antigens on PMNL.

Vero cell invasiveness of Proteus mirabilis

Vero cell invasiveness was studied for a group of Proteus mirabilis strains isolated from the urinary tract and feces and for a limited group of urinary isolates of Escherichia coli. Experimental conditions affecting this invasiveness were studied. All of the P. mirabilis strains tested were capable of cell invasion, whereas none of the E. coli strains was. Correlation between the hemolytic activity of the P. mirabilis strains and their invasive ability suggested that the bacterial hemolysin may be involved in the invasion process. Other experimental evidence supporting this hypothesis is discussed. The differences in the invasive capacities of P. mirabilis and of E. coli may be important for the apparent differences in the pathogenesis of urinary tract infection by both species.

Investigation of the haemolytic activity of Proteus mirabilis strains

Young broth cultures of all P. mirabilis strains tested exhibited haemolytic activity. This activity seemed to be strongly cell-associated as only a very small fraction of this activity was found in the cell-free supernatant. The haemolysin was only produced by actively growing cells. Inhibition studies with trypsin and chloramphenicol suggested that the haemolysin is of protein nature. Lecithin and serum of several species had an inhibitory effect on the haemolysin. Besides erythrocytes of various species also VERO cells were affected by the haemolysin. A correlation was found between the haemolytic activity of a strain and its virulence in an experimental mouse model.

Urinary virulence of Proteus mirabilis in two experimental mouse models

Two experimental mouse models were tested for their suitability in measuring urinary virulence of Proteus mirabilis. In the first model, the kidney-infecting dose and lethal dose were measured. In the second model, the kinetics of the numbers of bacteria in the kidneys and other organs of the mouse were monitored for 13 h after injection.