Meningitis following cochlear implantation: pathomechanisms, clinical symptoms, conservative and surgical treatments.

Pneumococcal otogenic meningitis is a rare postsurgical complication that can develop following stapedectomy or after cochlear implantation. The bacterial infection can be fatal in some instances. A recent increase in the incidence of otogenic meningitis among cochlear implant wearers is of concern. The majority of meningitis cases are associated with a 2-component electrode manufactured by one cochlear implant company. The device with the added 'positioner' component has been withdrawn from the market (FDA Public Health Web Notification: Cochlear Implant Recipients may be at Greater Risk for Meningitis, Updated: August 29, 2002, www.fda.gov/cdrh/safety/cochlear.html). Not all cases have been subsequent to otitis media and symptoms have developed from less than 24 h up to a few years after implantation. The purpose of this paper is to review and discuss the pathogenesis, pathology/bacteriology and to elaborate on some clinical features of otogenic meningitis in implanted children and adults. Essential aspects of surgery, electrode design, and cochleostomy seal are discussed. Conclusions are drawn from the available data and recommendations are made for good practice in cochlear implantation and follow-up.

Identification of Pneumocystis carinii in the lungs of infants dying of sudden infant death syndrome.

BACKGROUND: Recently Pneumocystis carinii has been identified in a significant number of infants diagnosed as having died from sudden infant death syndrome (SIDS) in South America and Europe. METHODS: We examined lung sections of 79 infants who died with a diagnosis of SIDS in Rochester, NY, and Connecticut for the presence of P. carinii. RESULTS: Organisms with a characteristic silver stain appearance for P. carinii were identified in 14% of the lung sections. CONCLUSIONS: These data suggest that a possible link between some cases of SIDS and infection with P. carinii should be further evaluated and that infection of young infants may serve as an important reservoir for human P. carinii.

Quinolone arthropathy in animals versus children.

The use of quinolones in children and accumulation of data on the pharmacodynamics of these drugs have been limited and delayed by concern regarding their chondrotoxicity. A comprehensive review of the findings in animals compared with the cumulative published findings in children and adolescents (>7,000 to date) allows the conclusion that such concern is not justified. Prospective controlled studies in children are justifiable in view of a continuing lack of correlation between findings in juvenile animals and those in children and because of the selected therapeutic advantages of the current and newer quinolones.

Secretory anti-Giardia lamblia antibodies in human milk: protective effect against diarrhea.

OBJECTIVE: To determine whether anti-Giardia lamblia secretory IgA (sIgA) antibodies in human milk protect infants from acquisition of or symptoms associated with Giardia infection. METHODS: One hundred ninety-seven Mexican mother/infant pairs were followed weekly from birth for diarrheal disease and feeding status. Infant stool specimens were collected weekly and were cultured for bacterial pathogens and tested for Giardia and rotavirus by enzyme-linked immunosorbent assay. Maternal milk samples were collected weekly for 1 month postpartum and monthly thereafter. To determine the protective effect of anti-Giardia sIgA in milk against infection and against diarrhea due to Giardia, milk samples from mothers of infected infants and appropriately matched controls were assayed for anti-Giardia sIgA by enzyme-linked immunosorbent assay. RESULTS: Asymptomatic, infected infants ingested significantly (P = .046) higher amounts of milk anti-Giardia sIgA compared with symptomatic, infected infants. However, milk anti-Giardia sIgA concentrations did not differ between Giardia-infected and noninfected infants. CONCLUSION: The amount of anti-Giardia sIgA in human milk was associated with prevention of symptoms of diarrhea due to Giardia, but not with acquisition of the organism.

Effect of subinhibitory concentrations of antibiotics on extracellular Shiga-like toxin I.

Patients with diarrhea due to strains of enterohemorrhagic Escherichia coli (EHEC) (e. g. O157:H7) might be at a higher risk of developing hemolytic uremic syndrome when treated with antimicrobial agents. It has been suggested that this might be due to an increase of release or production of vero or shiga-like toxin from such organisms, possibly as a stress response to antimicrobial agents. The aim of this study was to detect such increases in extracellular toxin in vitro with a newly developed method that exposed EHEC to high sublethal concentrations followed by a recovery phase at progressively lower concentrations. Five strains of EHEC were exposed to continuously changing concentrations of ciprofloxacin, co-trimoxazole, cefixime and tetracycline. The amount of free shiga-like toxin I (SLT-I) released was compared to the amount released from inocula that were not exposed to antibiotics. There were significant differences between the five EHEC strains in the amount of toxin detected after exposure to antimicrobial agents (p less than 0.001). Equally important was the type of antibiotic (p less than 0.001), with ciprofloxacin inducing the largest increase ranging from 169 to 436%, followed by co-trimoxazole, cefixime and tetracycline. In addition, the increases in free toxin correlated with the concentration of the antibiotics (p less than 0.001). The association between antibiotic-induced increases in SLT-I produced by strains of EHEC and certain classes of antibiotics might influence the analysis of future epidemiological studies on risk factors for HUS.

Design and performance of blanket heat shields for neonates.

BACKGROUND: Several designs of plastic blanket heat shields are in use. This study was done to compare different designs for their efficiency in reducing heat loss. METHODS: Four heat shield designs were tested by sequentially covering each of 14 infants (wt. 640-2,030 g) cared for under radiant warmers. The power consumption of the radiant warmers was measured as a surrogate for heat loss. All designs were tested for a total of 20 min on all infants. Results were calculated as percent change in power consumption from shield to shield. The most efficient design was further modified and evaluated in another group of 14 infants (wt. 700-1,180 g). RESULTS: The relative reductions in power consumption were: no shield (control) -0%, a plastic foil over the side rails: -17%, a single layer close to the infant but excluding the head: -34%, the same as double layer -37% and the most efficient one, a single layer covering the whole infant -42%. A modification of this design, tested in the second group of infants, reduced power consumption by 13% (95% CI -5.9/-19.7), (p less than 0.004) when compared to the single layer covering the whole infant. It was tucked under the connecting tubes to the ventilator. It also reduced the risk for displacement and allowed for the endotracheal tube to be suctioned without removing the blanket. CONCLUSION: Modifications of the design of heat shield blankets for infants resulted in significant increases in efficiency.

Comparative inactivation of isepamicin, amikacin, and gentamicin by nine beta-lactams and two beta-lactamase inhibitors, cilastatin and heparin.

This study was undertaken to compare the susceptibility to inactivation of isepamicin with amikacin and gentamicin when exposed to different beta-lactams, beta-lactamase inhibitors, and heparin. The aminoglycosides (5, 10, 20, and 50 micrograms/ml) were incubated in human serum with ampicillin, azlocillin, aztreonam, carbenicillin, ceftazidime, piperacillin, and ticarcillin (100 and 600 micrograms/ml) and with clavulanate, cilastatin, 1:1 imipenemcilastatin, oxacillin, and sulbactam (20 and 120 micrograms/ml) for 48 h at 37 degrees C. Aminoglycoside concentrations were measured by fluorescence polarization immunoassay (FPI) after 0, 8, and 48 h of incubation and by radial diffusion bioassay after 48 h of incubation. Each of the three aminoglycosides was also added to whole blood containing either heparin (100 U/ml) or 0.5% EDTA as a control and assayed after 6 h by FPI. The degree of inactivation of isepamicin by the beta-lactams was significantly less than that by amikacin (P less than 0.003) and gentamicin (P less than 0.0002) when determined by bioassay. Piperacillin, carbenicillin, and azlocillin produced the greatest amount of inactivation, and cilastatin and oxacillin produced the least. A similar pattern was observed when the degree of inactivation was measured by FPI. A significant difference in the degree of inactivation was noted between isepamicin and gentamicin (P less than 0.003 at 8 h and P less than 0.006 at 48 h) but not between isepamicin and amikacin (P greater than 0.7 at 8 h and P greater than 0.08 at 48 h). Aminoglycoside determinations by FPI were not influenced by the presence of heparin. In summary, isepamicin was found to be at least as stable as amikacin against inactivation by beta-lactam compounds and beta-lactamase inhibitors. Heparin (100 U/ml) did not influence aminoglycoside determinations by FPI.

Adjunctive treatment with low-dose polymyxin B improves operant response rates and latency in infant rats with Haemophilus influenzae type b meningitis.

Polymyxin B given in conjunction with ampicillin protects infant rats against death from overwhelming Haemophilus influenzae type b infection. This study was undertaken to examine whether polymyxin B would mitigate the effects of brain damage caused by meningitis. Six- to 7-d-old Sprague-Dawley rats were infected subcutaneously into the nape with 10(7) cfu Haemophilus influenzae type b strain Eagan. This dose consistently caused bacteremia (1.2 X 10(5) cfu/mL) and meningitis (0.5 X 10(5) cfu/mL) in pilot studies. Twenty-four h after infection, all animals received intraperitoneal treatment consisting in either ampicillin alone (400 mg/kg X 4 q 3 h) repeated 12 h later, n = 15 or combined with polymyxin B (0.1 mg/kg/dose) n = 16. At age 2 mo, they were taken off ad libitum feeding and maintained at 80% of their wt. They were then conditioned to receive a food pellet by pressing a lever (continuous reinforcement). The next day, the time lapse between placement and pressing the lever for the first time was recorded (conditioned operant response or latency). Three wk later, the animals were put in the test chamber again and the time to press the lever (latency), as well as the time required to obtain 100 pellets (rate) were recorded. Animals who received polymyxin B had a significantly shorter reaction time; mean 34 s, SEM +/- 5.7 versus mean 88 s, SEM +/- 26.3, P less than or equal to 0.05 and performed significantly faster in obtaining 100 pellets; mean 925 s, SEM +/- 72.1 versus mean 1283 s, SEM +/- 126.3, p less than or equal to 0.02 (analysis of variance, Scheffé test).(ABSTRACT TRUNCATED AT 250 WORDS)