Comparison of bedside inoculation of culture media with conventional cerebrospinal fluid culture method in patients with bacterial meningitis.

BACKGROUND: The yield of bacterial cultures from cerebrospinal fluid (CSF) at Kenyatta National Hospital (KNH) is very low. Bedside inoculation of culture media with CSF may improve yields. OBJECTIVE: To compare the culture yield of CSF inoculated onto culture medium at the bedside to that of CSF inoculated onto culture medium in the microbiology laboratory. DESIGN: Cross-sectional comparative study. SETTING: Accident and Emergency Department and medical wards at Kenyatta National Hospital. SUBJECTS: Cerebrospinal fluid from patients at KNH with a clinical diagnosis of acute meningitis. RESULTS: Two hundred and twenty CSF specimens were obtained during a four month period. S. pneumaniae was isolated from 24 CSF samples and H. influenzae from one. Bacterial cultures were positive in 25 (11.4%, 95% CI 7.0-15.6%) samples inoculated at the bedside and 23 (10.5%, 95% CI 6.5-14.5%) samples inoculated at the laboratory. Bacteria were isolated 5 hours earlier in samples inoculated at the bedside (95% CI 4.34-6.86 hrs, p < 0.05). Four per cent of S. pneumaniae isolates were resistant to crystalline penicillin. CONCLUSION: There was no significant difference in culture yield after bedside inoculation of culture media with CSF compared to traditional CSF culture method. Bedside inoculation of culture media with CSF resulted in faster time to positive culture.

Evaluation of a rapid membrane-based assay (HIV-CHEK) for detection of antibodies to HIV in serum samples from Nairobi.

We evaluated a rapid membrane-based assay (HIV-CHEK) for detection of antibodies to HIV using 737 serum samples in Nairobi, Kenya. The rapid assay had a sensitivity of 96.3% and specificity of 99.8% when compared with enzyme-linked immunosorbent assay (ELISA) and Western blot assay. Results were similar using fresh or previously frozen serum samples, although the latter occasionally left debris on the assay device membrane yielding uninterpretable results. This rapid HIV assay may be of particular use in developing countries where laboratory resources are limited.

Infection with HIV as a risk factor for adverse obstetrical outcome.

We carried out a case-control study to investigate the role of sexually transmitted diseases (STDs), including infection with HIV, as risk factors for adverse outcome of pregnancy. Overall, 1507 women were enrolled within 24 h of delivery. Cases (n = 796) were mothers of low-birthweight infants (less than 2500 g) or of stillborns. Low-birthweight infants were divided into preterms (n = 373) and neonates small for gestational age (n = 234). Stillborns were separated into intrauterine fetal deaths (n = 120), and intrapartum fetal deaths (n = 69). Controls were selected from mothers delivering a live baby of greater than or equal to 2500 g (n = 711). The maternal HIV seroprevalence in the control group was 3.1%. Prematurity was associated with maternal HIV antibody [8.6% seropositive; adjusted odds ratio (OR) 2.1; 95% confidence interval (CI) 1.1-4.0], as was being born small for gestational age (7.7% seropositive; adjusted OR 2.3; 95% CI 1.2-4.2). In mothers who delivered a stillborn baby, both intrauterine fetal death (11.7% seropositive; adjusted OR 2.7; 95% CI 1.3-5.5) and intrapartum fetal death (11.6% seropositive; adjusted OR 2.9; 95% CI 1.3-6.5) were independently associated with HIV seropositivity in the mother. Maternal syphilis was confirmed as an important risk factor for intrauterine fetal death (14.3% positive; adjusted OR 4.8; 95% CI 2.4-9.5). No significant association was found between other STDs, including gonococcal and chlamydial infection, and adverse obstetrical outcome. These results suggest an association between maternal HIV infection and adverse obstetrical outcome, defined as low birthweight and stillbirth.

Minimising antibiotic resistance to Staphylococcus aureus in developing countries.

OBJECTIVE: To assess the role of rational drug use and laboratory service in preventing the emergence of multiple antibiotic resistant Staphylococcus aureus in developing countries. DATA SOURCE: Literature search on compact disk-read only memory (CD-ROM) Medline and Internet using the key words: Staphylococcus and antibiotic resistance. A few articles were manually reviewed. STUDY SELECTION: Relevant studies or articles on antibiotic resistance with special reference to Eastern Africa, region are included in the review. DATA EXTRACTION: From individual studies or articles. DATA SYNTHESIS: Evidence for the spread of S. aureus multiple antibiotic resistance is synchronized under the headings: Introduction, current situation, antibiotic resistance control strategies, are outlined. CONCLUSION: There is need for concerted efforts between different groups to monitor changes in the epidemiology and antibiotic resistance of S. aureus. Strategies aimed at preventing transmission of resistant strains are remarkably effective when strictly enforced. Necessary attention should be given on the subject so that meaningful control measures preventing the expansion of antimicrobial resistance can be formulated, thereby ensuring the future successful treatment of Staphylococcal infections.

Typhoid fever: a review of its impact and diagnostic problems.

A retrospective analysis was done on the diagnosis of typhoid fever based on clinical symptoms and available laboratory data over the last 16 years from rural areas of four African countries. This analysis concentrated on the reliability of diagnosis without cultures which cannot be performed in most rural hospitals due to lack of the necessary expertise and equipment. The analysis showed the problem to be increasing perhaps because of interaction of salmonella infection with human immunodeficiency virus (HIV), malnutrition and other infections together with neglected sanitary facilities and lack of clean water. The use of certain cardinal clinical symptoms combined with available laboratory tests were shown to enhance the diagnosis of typhoid fever, especially in vulnerable persons. In conclusion the paper suggests that using the approach followed to obtain this data in rural tropical areas one can confidently make a diagnosis of typhoid fever.

Salmonella typhimurium outbreak at Kenyatta National Hospital (1985).

A total of 560 Salmonellae species were isolated from Jan-Dec 1985. Of these, 347 (62%) were from blood cultures, 180 (32%) from stools and 33 (6%) were from cerebrospinal fluid (CSF) and other body aspirates. S. typhimurium were the highest isolated. These were, 291 (52%) from blood cultures, 94 (17%) from stool cultures and 32 (6%) from CSF. S. typhimurium was also multi-drug resistant. More than 50% strains of S. typhimurium were resistant to ampicillin, tetracycline, kanamycin and chloramphenicol. The need for coordination between the laboratory and clinical staff to prevent the spill-over of infection with S. typhimurium and its epidemic spread is discussed.

Association of human rotavirus infection and intestinal rotavirus-specific immunoglobulin A in children with diarrhoea.

The purpose of this study was to determine the role of intestinal specific rotavirus IgA antibody in protection against diarrhoea due to rotavirus infection. Stool from children aged below 5 years with diarrhoea who reported to the Paediatric Observation Ward, Kenyatta National Hospital were examined for micro-organisms and IgA antibody. Specific rotavirus IgA antibody and antigen were determined using enzyme linked immunosorbent assay technique. Out of 153 stool specimens, 22% (34/153) were positive for rotavirus antigen and 15% (23/153) had IgA specific antibody to rotavirus. Children with specific IgA to rotavirus had no rotavirus except in two cases (p < 0.05). There was no difference in levels of specific IgA antibody between normal and malnourished children (p = 0.4). It is probable that intestinal specific IgA to rotavirus protects children against rotavirus diarrhoea.

Aerobic and facultative bacterial isolates from blood cultures of children with clinically diagnosed septicaemia.

A total of 120 sets of blood cultures were performed aerobically from 60 children with clinically diagnosed septicaemia at Kenyatta National Hospital, Nairobi. Out of these, 36 (30%) sets from 19 (31.7%) patients yielded bacterial growth while 84 (70%) sets from 41 (68.3%) were negative. Salmonella typhimurium was the most frequently isolated bacteria (63%), followed by Staphylococcus aureus (15.8%). Salmonella typhimurium isolates were mostly multi-antibiotic resistant, most of them only sensitive to amikacin and cefotaxime, while all were resistant to ampicillin and co-trimoxazole, the most frequently used antibiotic in this hospital.

PIP: Between March 1987-January 1988, physicians enrolled 60 pediatric patients with a fever who were admitted to the Kenyatta National Hospital in Nairobi, Kenya for various clinical conditions in a study to determine the types, frequency, and antibiotic sensitivity patterns of aerobic and facultative bacterial isolates. Most of the patients were 13 months-4 years old (45%). 31.7% of the patients had positive blood cultures. Staphylococcus aureus was the 2nd most common bacteria (15.8%) among these patients. Laboratory personnel isolated Salmonella typhimurium in most patients (63%). In fact, during the same period, the Diagnostic Microbiology Laboratory at the hospital identified Salmonella species in 48% of all isolated bacteria and 35% of these were S. typhimurium. S. typhimurium tended to be present in children with gastroenteritis (41.8%) or a fever of unknown origin (33.3%). S. typhimurium was very sensitive to amikacin and cefotaxime, but resistant to ampicillin and sulfamethoxazole-trimethoprim. Health workers in Kenya have frequently administered ampicillin and sulfamethoxazole-trimethoprim, but not amikacin and cefotaxime. 67% of the strains of S. typhimurium were resistant to gentamicin and 33% to chloramphenicol. These results along with those of other reports from this hospital indicated a dramatic rise in Gram negative bacteria resistance to antibiotics. Therefore physicians should no longer consider gentamicin as a 1st line antibiotic in treating suspected septicemia patients.

Adaptation of a method of measuring zone diameters of bacterial growth inhibition by antibiotics to suit developing countries.

A study that devised a modified method of reporting antibiotic sensitivity results was undertaken. Enterobacteriaceae and Gram positive cocci were tested for drug sensitivity by a disc diffusion method. Zones of bacterial growth inhibition were measured, dividing the isolates into four groups: the highly sensitive, the moderately sensitive, the slightly sensitive and the resistant ones. The slightly sensitive isolates were taken as indicators of antibiotic resistance acquisition. By that system, when more than 50% of the isolates fell into the slightly and resistant groups, that meant that the antibiotic concerned would be discontinued for some time until the bacteria reverted to being moderately sensitive. The study also provided a method of making antibiotic discs from local blotting papers, and a sample of a form on which antibiotic sensitivity results could be recorded was presented. The method is considered to be easy and very appropriate for developing countries in detecting gradual and abrupt acquisition of antibiotic resistance by bacteria.

Bacteriology of infections in a rural tropical area of Kenya: isolates and antibiotic susceptibility.

No microbial data are available for rural tropical areas. Yet most people in Africa live in rural areas where the burdens of infectious diseases remain the primary cause for morbidity and mortality. A pilot study was done by culturing midstream urine from patient with dysuria (100 with Gram-negative fermentative rods), blood from patients with FUO (55/307 positive) and swabs from patients with infected wounds (29 with Staphylococcus aureus). Bacteria were identified and susceptibility tests were performed according to protocols. Klebsiella was the most common isolate from urine (48%). The overall resistance in urine isolates for ampicillin was 76% and for cotrimoxazole, 60%. The main isolates from blood were Klebsiella Enterobacter sp (21/55) and Salmonella (14/55) in patients from agricultural areas, and Staphylococcus aureus (12/55) in patients from desert areas. The Staphylococcus aureus from wounds were all penicillin resistant and 9/29 were cloxacillin resistant. Influence of the environment, underlying pathology and previous use of antibiotics give unexpected predominance of Klebsiella and high resistance in all isolates. Large scale surveillance studies are needed so that prescription of antibiotics can be based on locally obtained data.

Bacteraemia in patients presenting with fever.

In three studies, in Ghana and Kenya, blood from 639 patients admitted with fever was cultured. Standard treatments were antimalarials (54-100%) and antibiotics (39-90%). According to the criteria in use, however, only 10-31% had malaria alone; of those who received antibiotics, 66% were diagnosed with malaria, gastrointestinal infections, post-operative recuperations, circulatory problems, central nervous system disorders or FUO, and did not need antibiotics at the first encounter. For those with wounds and abscesses (8%), generalised antibiotic treatment can also be questioned. Bacteraemia was found in 71 (11.3%) patients; in the HIV patients, however, 5 (23%) of 22 had bacteraemia. This is a minimum incidence, since culture techniques were not optimal for the isolation of fastidious microorganisms. The most prevalent organisms isolated were Salmonella, Klebsiella/Enterobacter and S. aureus. Resistance (intrinsic and extrinsic) in the Gram- bacteria was high: 31-100% were resistant to amoxycillin, 0-80% to cotrimoxazole, 15-95% to chloramphenicol and 9-15% to gentamicin. The need for cultures and sensitivity tests for patients with prolonged or undiagnosed fever is stressed. Specific treatment should be given only when infections, whether malarial or bacterial, have been positively diagnosed.

Typhim Vi vaccine against typhoid fever: a clinical trial in Kenya.

Safety, tolerance and immunogenicity of the purified Vi polysaccharide vaccine (Typhim Vi) against typhoid fever was evaluated in primary school children aged 5-15 years. A total of 435 children were vaccinated, each with a single intramuscular injection in the left deltoid muscle. One hundred and ten children were randomly selected for blood samples on day 0 (pre vaccination) and day 30 (post vaccination). Vi antibodies studied by Radio immuno assay (RIA) on 97(88%) paired sera showed a seroconversion rate of 76.2% and seroprotection rate after vaccination was 74.2%, while 6.2% of children already had protective immunity before vaccination. The vaccine was well tolerated. Most commonly reported reactions were mild pain at site of injection (83%), and a few complained of mild swelling (4.6%), induration (1.1%), itching (1.1%) and headaches (1.4%). All reactions were of mild severity and disappeared within 24 to 48 hours.

Enteric pathogens in malnourished children with diarrhoea.

Enteric pathogens were determined from stools of 273 children aged less than 5 years at Kenyatta National Hospital (KNH), 43.6% (119/273) of whom were malnourished according to the Wellcome criteria. Rotavirus was detected by ELISA test, Salmonella, Shigella and E. coli by culture on MacConkey and Salmonella-Shigella agar at 37 degrees C overnight and Campylobacter on Skirrow's selective media at 42 degrees C for 48 hrs. These were identified by biochemical tests and serotyping using specific antisera. Whereas isolation rate for Campylobacter (0.0% vs 5.0%, p = 0.006), well malnourished ETEC-LT (0.6% vs 5.0%, p = 0.003) and T. hominis (0.0% vs 3.4%, p = 0.03) was higher in the malnourished children, EPEC (30.5% vs 10.1%, p < 0.001) and Salmonella+ETEC-LT (7.8% vs 1.7%, p = 0.02) was higher in children. The other enteric pathogens were equally isolated from normal and malnourished children. A larger proportion of malnourished children had diarrhoea of unknown aetiology compared to the well nourished (26.6% vs 50.4%, p < 0.001). Campylobacter and T. hominis may be opportunistic infections due to immuno-suppression in malnutrition. Diarrhoea of unknown aetiology may be due to aetiological agents that were not determined in this study.

Effect of human immunodeficiency virus on local immunity in children with diarrhoea.

The purpose of this study was to determine the relationship between intestinal mucosal immunity and diarrhoea. Stools were tested for total IgA by radial immunodiffusion, cultured for bacteria and examined for ova/cysts by microscopy. Peripheral blood was screened for HIV-1 antibody by ELISA, CD4 and CD8 enumerated by flow cytometry and phagocytic activity by C. albicans engulfment. A total of 271 children were enrolled with a mean age of 20.3 m (range 0.3-60.0 m). HIV exposed (born to HIV seropositive mothers) had more episodes of diarrhoea than HIV unexposed (born to HIV seronegative mothers) children in the first six months of life (26.0% versus 5.5%, p = 0.002). Exposed children had severe (16/44 versus 6/29, p = 0.02) and prolonged diarrhoea lasting more than nine days (11.0% versus 1.4%, p = 0.03) than unexposed. CD8 counts were significantly higher in exposed than unexposed children (1837.0 versus 1373.0 cells/mm3, p = p.01). Among children aged 15 months and over, HIV seropositive children had severe diarrhoea (4/6 versus 11/32, p<0.01), reduced phagocytic activity (phagocytic index 15.4 versus 28.9, p<0.01), total intestinal IgA (0.2 versus 0.7 mg/ml, p = 0.04) and CD4 counts (624.2 versus 1345.1 cells/mm3, p = 0.01) than seronegative. Reduction of CD4 was more significant in HIV seropositive children with severe diarrhoea (298.7 versus 1318.5 cells/mm3, p = 0.01). Isolation of enteric pathogens was independent of either maternal or child's HIV serostatus although E. coli was more frequent in children with low CD4 counts. These results highlight the importance of mucosal immunity in the intestinal infections. Exposure to HIV, reduced CD4 counts and IgA were associated with diarrhoea probably due to impaired intestinal mucosal immunity.

Criteria for better detection of brucellosis in the Narok District of Kenya.

Monthly disease summary sheets from 1986-1992 of 60 dispensaries, clinics and hospitals in Narok district, Kenya were reviewed for the occurrence of brucellosis and other diseases with "flu-like symptoms". Diseases with these symptoms accounted for about 52% of the 1,037,875 cases reported for the time period. These were classified as malaria (79.3%), rheumatism (7.1%), PUO (2.4%), and brucellosis (0.8%). Brucellosis was diagnosed by a positive Rose Bengal (RB) test routinely conducted in seven out of the 60 health units. In these units, 55% of flu-like cases were classified as malaria and 21.2% as brucellosis. Individual case records of patients at four dispensaries using the RB test during 1991-92 were assessed for specific predictor symptoms. For 625 RB tested patients, a positive test result was associated with joint pain, headache, and the combinations of joint pain with headache and lameness with headache. A logistic regression model correctly predicted the RB test result in 62.3% of the time. For the 465 patients examined by the blood smear examination, identification of malaria parasites was associated with, headache, joint pain and combinations of emesis with pale mucous membranes. This regression model correctly predicted positive results 67.2% of the time. Both models indicate that selected clinical predictors represented significantly increased odds of being positive to the respective tests. However, for both diseases, clinical signs alone appear insufficient for reliable diagnosis and differentiation probably due to resemblance in symptomatology between these two and other diseases.

Breastfeeding and immunity to intestinal infections.

The purpose of this study was to compare immune response in breast and non breastfed children presenting with diarrhoea at Paediatric Observation Ward, Kenyatta National Hospital (KNH-POW) and Maternal and Child Health Clinic, Pumwani Maternity Hospital (PMH-MCH). Blood and stool samples were collected from the first four consecutive children aged 5 years and below per day, presenting with or without diarrhoea from January to December, 1992. The stools were tested for total IgA by single radial immunodiffusion (SRID) and specific IgA by enzyme linked immunosorbent assay (ELISA). Peripheral blood CD4 and CD8 enumeration was done by flow cytometry. Stools were cultured for bacteria on selective media while ova and cysts of parasites were identified by wet preparation microscopy. A total of 457 children were enrolled into the study, 69.6% of whom presented with diarrhoea. Breastfed children tended to have a shorter duration of diarrhoea than either mixed fed or bottle fed (8.3 vs 9.8 vs 11.2 days, p = 0.2). In general, E. coli were more commonly isolated from breastfed than mixed fed or bottle fed (56.7% vs 43.9% vs 28.9%, p = 0.004) while intestinal parasites were mostly in bottle fed than mixed or breastfed children (28.8% vs 8.2 vs 0.8, p < 0.004). However, when children with diarrhoea were considered, E. coli was more frequently isolated from bottle fed children who presented with diarrhoea than without (26.7% vs 7.7%, p = 0.03).(ABSTRACT TRUNCATED AT 250 WORDS)

Bacterial meningitis in children admitted in hospitals within Nairobi.

Four hundred and ninety nine children (aged between one month and five years) admitted with clinical features of meningitis were recruited in cross-sectional survey of bacterial meningitis in hospitals within Nairobi. Lumbar punctures were done on all of them and the cerebrospinal fluid (CSF) analysed bacteriologically and serologically for the common causative organisms. Two hundred and fifty (50.1%) cases were diagnosed clinically as having meningitis. Of these, 132 (52.8%) had turbid CSF specimens, while 118 (47.2%) were clear. When turbid CSF specimens were cultured, 83 (62.8%) yielded three common bacterial micro-organisms namely; Neisseria meningitidis, Streptococcus pneumoniae and Haemophilus influenzae in that order of frequency. The implications of these findings in paediatric meningitis together with the drug sensitivity patterns is presented and discussed.

Mixed infections in childhood diarrhoea: results of a community study in Kiambu District, Kenya.

One thousand four hundred and twenty diarrhoea specimens from 846 children aged 0 to 60 months were collected and analysed for bacteria, parasites and rotavirus over a 16 month period, from June 1985 to September 1986 inclusive. The study was conducted in 4 villages situated in Kiambu District, Kenya. All the specimens were analysed for rotavirus and parasites, including Cryptosporidium. The majority of the specimens were analysed for enteropathogenic Escherichia coli (EPEC), Shigella, Salmonella, Campylobacter and Aeromonas. Only 387 specimens were analysed for enterotoxigenic E. coli (ETEC). However, of this proportion analysed for ETEC, 33% were positive. A total of 344 specimens were negative for any organisms while a further 140 were only positive for parasites which have been implicated as being pathogenic, including Ascaris lumbricoides, Hymenolepis nana, Trichomonas hominis and Blastocysts hominis were considered to be at least potentially pathogenic and capable of causing diarrhoea. An average of only 29.4% of these organisms occurred as single isolates. The remaining infections were mixed, with a maximum of 7 potentially pathogenic organisms occurring together in a single specimen. The associations of certain organisms were significant, notably Campylobacter with Giardia lamblia. Campylobacter with EPEC, EPEC with Ascaris, and G. lamblia with rotavirus. The latter was a negative association.