Skip Navigation

Impact of Bacteremia on the Pathogenesis of Experimental Pneumococcal Meningitis

  1. Christian T. Brandt1,3,
  2. David Holm4,
  3. Matthew Liptrot4,
  4. Christian Østergaard1,5,
  5. Jens D. Lundgren3,
  6. Niels Frimodt-Møller1,
  7. Ian C. Skovsted2 and
  8. Ian J. Rowland4
  1. 1National Center for Antimicrobials and Infection Control, University of Copenhagen, Copenhagen, Denmark
  2. 2The Pneumococcus Laboratory, Statens Serum Institut, University of Copenhagen, Copenhagen, Denmark
  3. 3Copenhagen HIV Programme, Faculty of Health Sciences, University of Copenhagen, Copenhagen, Denmark
  4. 4Danish Research Centre for Magnetic Resonance, Copenhagen University Hospital Hvidovre, Copenhagen, Denmark
  5. 5Clinical Microbiological Department, University Hospital Herlev, Copenhagen, Denmark
  1. Reprints or correspondence: Dr. Christian T. Brandt, Copenhagen HIV Programme, Faculty of Health Sciences, University of Copenhagen, Panum Institute, Blegdamsvej 3, 2200 Copenhagen N, Denmark (ctb{at}ssi.dk).
 
Next Section

Abstract

Background. Bacteremia plays a major role in the outcome of pneumococcal meningitis. This experimental study investigated how bacteremia influences the pathophysiologic profile of the brain.

Methods. Rats with Streptococcus pneumoniae meningitis were randomized to 1 of 3 groups of infected study rats: (1) rats with attenuated bacteremia resulting from intravenous injection of serotype-specific pneumococcal antibody, (2) rats with early-onset bacteremia resulting from concomitant intravenous infection, or (3) a meningitis control group. The blood-brain barrier (BBB) breakdown, ventricle size, brain water distribution, and brain pathologic findings were analyzed using magnetic resonance morphological and functional imaging. Laboratory data and clinical disease scores were obtained.

Results. Attenuation of the bacteremic component of pneumococcal meningitis improved clinical disease symptoms and significantly reduced ventricle expansion and BBB breakdown (P < .05). Early-onset bacteremia did not further increase ventricle size or BBB leakage. Significantly increased brain edema developed among rats with both attenuated and early-onset bacteremia (P < .05). Focal brain pathologic findings were unaffected by bacteremia and were found to be associated with cerebrospinal fluid inflammation.

Conclusion. Although brain lesions appear to result from local meningeal infection, systemic infection significantly contributes to clinical disease presentation and the pathophysiology of BBB breakdown and ventricle expansion. The different end points affected by the systemic and local infectious processes should be addressed in future studies.

Pneumococcal bacteremia is present in the majority of persons with pneumococcal meningitis and is associated with a poor outcome [1]. Studies of experimental pneumococcal meningitis have demonstrated that attenuation of secondary bacteremia, which is an inevitable development in models of experimental meningitis, improves clinical disease presentation and increases survival [2, 3]. In contrast, increased bacteremia, resulting as a consequence of experimental treatments or genetic alterations of the host immune response, results in increased mortality [4–6]. Experimental studies have shown that septicemia without accompanying meningitis leads to neuronal dysfunction and injury [7, 8] and can be compared with the occurrence of encephalopathy, a well-known disease entity, in septicemic patients [9]. However, in studies of experimental meningitis, there is discordance between the development of neuronal injury and the clinical grading of disease, survival, or final outcome. Experimental manipulations resulting in a poorer outcome do not necessarily lead to increased neuronal injury [4, 10], and neuroprotection is not necessarily associated with improved clinical status [11, 12].

Systemic disease, therefore, appears to be an important factor in survival from meningitis. It is hypothesized that bacteremia, in addition to focal brain injury, participates significantly in the resultant pathophysiologic profile of the brain after meningitis. Consequently, the present experimental study aimed to investigate the role of bacteremia in pneumococcal meningitis by modulating the bacteremic component of the disease. The effects of early-onset and diminished bacteremia on blood-brain barrier (BBB) integrity, ventricle expansion, brain water distribution, and brain injury, together with clinical and laboratory results, were investigated. Such information is essential for the development of new treatment approaches that are needed to reduce disease-induced sequelae and improve outcome and survival.

Previous SectionNext Section

Animals, Materials, and Methods

The experimental protocol was approved by the Danish Animal Inspectorate (Dyreforsoegstilsynet). Adult, male Wistar rats (body weight, 300–320 g) were used in the experiments.

Experimental Study Design

A total of 43 rats were randomized to 1 of 4 groups: (1) rats that received intracisternal injection of pneumococci (hereafter referred to as the “cerebrospinal fluid [CSF]—infected group”; n = 11); (2) rats that were intracisternally and intravenously infected (hereafter referred to as the “early-onset bacteremia group” or “CSF/blood-infected group”; n = 10); (3) rats that were intracisternally infected and received intravenous injection of antipneumococcal antiserum (Ab) at the time of infection (hereafter referred to as the “Ab-CSF—infected group”; n = 14); or (4) rats injected intracisternally with sterile beef broth (control group; n = 8).

Magnetic resonance imaging (MRI) was performed for all rats at 28 h after infection. Rats pretreated with Ab presented with reduced clinical signs of infection at 28 h after infection. Consequently, 8 rats that were pretreated with Ab had imaging performed at 28 h and again at 38 h after infection (hereafter referred to as the “Ab-CSF—infected group at 38 h”).

All rats were clinically assessed, and a motor performance score was determined immediately before imaging. Blood and CSF samples were obtained after MRI. One rat in the CSF-infected group died of ophistotonic convulsions before MRI was performed.

Infection and Treatment with Serotype-Specific Antiserum

A Streptococcus pneumoniae type 3 strain (strain 68034; Statens Serum Institut) known to reproducibly induce severe disease was used [2]. The infectious inoculum was prepared by diluting the bacteria in sterile beef broth to obtain a concentration of ∼1 × 106 cfu/mL, as confirmed by quantitative cultures. Rats were anesthetized with Hypnorm/Dormicum/atropine and received intracisternal injection of 30 μL of the bacterial suspension or beef broth. Rats randomly assigned to the CSF/blood-infected group received intravenous injection of 0.6 mL of the infectious inoculum immediately after intracisternal injection.

A serotype-specific polyclonal rabbit antipneumococcal capsular serotype 3 antiserum (Pneumosera; Statens Serum Institut) was purified (15.0 mg of pneumococcal serotype 3 antibody/mL) and diluted 1:1 in PBS, as described elsewhere [3]. Treated animals received a single intravenous injection (1.0 μL/g) immediately after infection.

Sampling and Analysis of CSF and Blood

CSF and blood samples were obtained from each animal after MRI was performed for the control, CSF-infected, and CSF/blood-infected groups. MRI was performed twice for 8 Ab-CSF—infected rats. CSF samples from these rats were not obtained at 28 h after infection, to ensure that the course of disease was unaffected. However, blood samples were obtained from 6 of these rats at 28 h after infection.

The white blood cell (WBC) counts in 20-μL samples of CSF or blood were measured using an automatic cell counter (Medonic Vet-Autocounter 620; Boule Medical). Bacterial counts in CSF samples were determined by plating 10-fold serial dilutions of 20 μL of CSF. Fifty microliters of undiluted blood and a 20-fold dilution were also plated. CSF and blood samples obtained from uninfected controls were plated in undiluted form.

Assessment of Clinical Disease, Motor Performance, and Disease Severity

Clinical appearance and motor-function scores were determined before each scanning session. Scoring criteria are shown in table 1.

Figure 1
View larger version:
Figure 1

Breakdown of the blood-brain barrier (BBB). BBB leakage is measured as the fraction of the cerebral cortex area that has significant contrast enhancement. Rats that were intracisternally infected and received intravenous injection of antipneumococcal antiserum (Ab) at the time of infection (i.e., the Ab-cerebrospinal fluid (CSF)—infected group) had significantly reduced BBB leakage, compared with rats that received intracisternal injection of pneumococci (i.e., CSF-infected rats) and rats that were infected intracisternally and intravenously (i.e., the CSF/blood-infected group) (P < .05). All infected study groups had increased BBB leakage, compared with controls (P < .05).

Figure 2
View larger version:
Figure 2

Degree of blood-brain barrier (BBB) breakdown. The degree of BBB breakdown is measured as the fraction of the cerebral cortex in which the contrast agent failed to significantly wash out during the dynamic magnetic resonance imaging sequence. Rats that received intracisternal injection of pneumococci (i.e., cerebrospinal fluid [CSF]—infected rats) and rats that were intracisternally and intravenously infected (i.e., the CSF/blood-infected group) had significantly increased areas of the cortex that were without washout, compared with rats that were intracisternally infected and received intravenous injection of antipneumococcal antiserum (Ab) at the time of infection (i.e., the Ab-CSF—infected group) and controls (P < .05). Values obtained for Ab-CSF—infected rats were not different from those obtained for uninfected controls (P > .05).

Figure 3
View larger version:
Figure 3

Expansion of brain ventricles. Ventricle size was significantly increased among the rats that received intracisternal injection of pneumococci (cerebrospinal fluid [CSF]—infected rats), compared with the control rats (P < .05). Ventricle size was significantly decreased in rats that were intracisternally infected and received intravenous injection of antipneumococcal antiserum (Ab) at the time of infection (i.e., the Ab—cerebrospinal fluid [CSF]—infected group), compared with rats that received intracisternal injection of pneumococci (i.e., CSF-infected rats) and rats that were intracisternally and intravenously infected (i.e., the CSF/blood-infected group) (P < .05), but it increased to intermediate values at 38 h after infection.

Figure 4
View larger version:
Figure 4

Brain pathologic findings observed using postcontrast T1-weighted (T1W) (A) and T2-weighted (T2W) (B) magnetic resonance (MR) imaging sequences. Representative MR images showing the brain of healthy, uninfected control rat are shown in panel A (i–iii) and panel B (i–iii). Negligible contrast enhancement is observed in the meninges, cortex, or ventricles on the postcontrast T1W images. The T2W images show the homogeneity of brain parenchyma and nondilated ventricles. Varying degrees of meningeal enhancement were observed on T1W images of rats that received intracisternal injection with pneumococci (i.e., cerebrospinal [CSF]—infected rats) and rats that were intracisternally and intravenously infected (i.e., the CSF/blood-infected group) (A, iv–xviii ). An extreme level of enhancement in CSF/blood-infected rats is shown in panel A (xiii–xviii). This level of enhancement was observed only among CSF-infected rats (4 of 11 rats) and CSF/blood-infected rats (5 of 10 rats) and was associated with the highest levels of BBB leakage and leakage severity. Brain pathologic findings were primarily observed as focal enhancement extending into the cortex. The localization, distribution, and “streaklike” appearance suggested that the lesions were of vascular origin. On T2W images, pathologic findings were observed as bright, wedge-shaped areas in the cortex (B, viii and ix ), dark ring lesions (B, xii), or hyperintense lesions as seen in the white matter (B, vii). Individual lesions were often apparent on both the contrast-enhanced T1W images and the T2W images (A, v; B, v). Other pathologic findings in the brains of infected rats were dilation of the lateral and third ventricles (B, iv), and irregularity of the cortex edge (B, xv). This irregularity was considered to be related to venous congestion and was associated with the dark vessel bands observed in the cortex (panel B, subpanel xiii).

Table 1
View larger version:
Table 1

Guidelines for clinical and motor performance scoring.

MRI

MRI was performed using an Inova 4.7-T imaging and spectroscopy system (Varian). Rats were positioned in a stereotactic device inside a home-built quadrature coil. Animals were kept warm with the use of a blanket and circulating warm water. Anesthetized rats underwent T1-weighted (T1W), T2-weighted (T2W), quantitative diffusion, dynamic MRI (dMRI), and post-contrast T1W imaging.

Apparent diffusion coefficient (ADC) mapping and T2W imaging. Quantitative diffusion measurements (along x, y, and z) were performed before the administration of contrast agent (repetition time [TR], 1500 ms; echo time [TE], 65 ms; matrix size [MA], 128 × 128; field of view [FOV], 35 × 35 mm; slice thickness [SL], 1; number of transients [NT], 1 [with b values of 0, 185, 740, and 1665 s/mm2]; 16 contiguous slices). ADCmaps were calculated for all 16 slices, and measurements of region of interest (ROI), which were performed in the cerebral neocortex, white matter, and jaw muscles in 3 coronal slices (frontal, midfrontal, and midbrain), were obtained. The mean ADC values from the selected regions are reported here. Importantly, the meningeal layer was not included in the ROI measurement in the cortex. T2W images were acquired as described above with b = 0.

T1W imaging and dMRI. T1W images (TR, 300 ms; TE, 14 ms; SL, 1 mm; MA, 128 × 128; FOV, 35 × 35 mm; NT, 4; 16 contiguous slices) were acquired before bolus administration of contrast agent (0.5 mmol/kg Gd-DTPA [Magnevist; Schering AS]) via a cannulated tail vein. Bolus passage was followed using a dMRI protocol in which 999 sets of FLASH T1W images (1 slice) were obtained with 100 images acquired before and 899 images acquired after injection of contrast agent (TR, 10 ms; TE, 4 ms; flip angle, 60°; SL, 2.5 mm; MA, 128 × 64; FOV, 35 × 35 mm; NT, 1). BBB integrity was determined by measuring the number of enhancing voxels, which were classified according to whether significant T1 enhancement occurred within the voxel after Gd-DTPA administration. Enhancing voxels resulting from BBB leakage, as well as the severity of this leakage (as assessed by slow washout of contrast agent), were measured in ROIs covering the cerebral neocortex (excluding the meninges). Dynamic data for Ab-CSF—infected rats at 38 h after infection were influenced by the first injection of contrast agent at 28 h after infection and were not analyzed further. Results from the dynamic data are presented as the fraction of the cortex with enhancing voxels and the fraction of voxels without significant contrast agent washout. Higher-resolution T1W images were acquired after bolus administration (TR, 300 ms; TE, 14 ms; SL, 1 mm; MA, 256 × 256; FOV, 35 × 35 mm; NT, 16; 16 contiguous slices).

Measurement of brain ventricle size. Measurement of the size of the third and lateral ventricles was performed by finding the threshold value of the ADC in the calculated ADC maps so that the CSF could be readily identified. The ventricle size was determined by semiautomated counting of pixels, with an ADC of >100 mm2/s in each of the 5 anatomically equivalent slices selected per animal, by use of ImageJ (National Institutes of Health; available at: http://rsb.info.nih.gov/ij/; 1997–2006).

Brain Pathologic Findings

Evaluation, description, and classification of pathologic findings were performed using T2W and high-resolution postcontrast T1W images. A quantitative score was obtained as described elsewhere [2], with each coronal brain slice divided into 20 segments (10 segments in the outer cortex and 10 segments in the inner cortex) and with 1 point assigned when pathologic findings were observed in 1 segment. Thirteen of the 16 acquired coronal sections per rat (excluding the cerebellum and the 2 most frontal sections) were included in this analysis.

Statistical Analysis

Data are presented as median values and interquartile ranges (IQRs). Comparisons were performed using the nonparametric Mann-Whitney U test and Fisher's exact test, and P < .05 was considered to be statistically significant. Spearman's rank correlation analysis was performed, and P < .01 was considered to be statistically significant.

Previous SectionNext Section

Results

Clinical and laboratory data were compared for Ab-CSF—infected, CSF/blood-infected, and CSF-infected rats. The control group was included for comparison of the MRI-generated data.

Effects of Attenuated or Early-Onset Bacteremia Concomitant to Pneumococcal Meningitis 28 h after Infection

Clinical scores, motor performance scores, and laboratory data obtained 28 h after infection for all infected groups and also for the control group are presented in table 2.

Table 2
View larger version:
Table 2

Clinical scores, motor performance scores, and laboratory data obtained 28 h after infection in cerebrospinal fluid (CSF)—infected, CSF/blood-infected, and antipneumococcal antiserum (Ab)—CSF—infected rats, as well as in uninfected control rats.

Disease scores. Ab-CSF—infected rats had significantly lower clinical and motor performance scores than did CSF-infected and CSF/blood-infected rats (clinical score, P = .003 and P = .002, respectively; motor performance score, P = .002 and P = .003, respectively). The clinical and motor performance scores of CSF-infected rats did not differ significantly from those of CSF/blood-infected rats (P > .05).

Bacterial loads and WBC counts in the CSF. No significant differences were found when pairwise comparisons of the Ab-CSF—infected, CSF-infected, and CSF/blood-infected rats were performed (P > .05).

Bacterial loads and WBC counts in blood. The blood bacterial load was significantly reduced in Ab-CSF—infected rats, compared with CSF-infected and CSF/blood-infected rats (P = .001 and P = .001, respectively). CSF/blood-infected rats did not present with increased blood bacterial loads, compared with CSF-infected rats (P > .05), at 28 h after infection, despite bacteremia occurring earlier in the former group.

WBC counts in blood did not differ significantly between the Ab-CSF—infected, CSF—infected, and CSF/blood-infected groups. Only among Ab-CSF—infected rats were blood WBC counts significantly increased, compared with those noted for controls (P = .004).

BBB Leakage and Contrast Agent Washout 28 h after Infection

BBB leakage. The fraction of the cerebral cortex with BBB leakage was significantly increased among Ab-CSF—infected, CSF-infected, and CSF/blood-infected rats, compared with that noted among control rats (P = .04, P = .0004, and P < .0001, respectively) (figure 1).

Of the infected study rats, Ab-CSF—infected rats had a significantly reduced area with BBB leakage, compared with both CSF-infected and CSF/blood-infected rats (P = .0008 and P = .0002, respectively). BBB leakage was comparable in the latter 2 groups (P > .05).

Washout of contrast agent in the cerebral cortex—severity of BBB leakage. The fraction of the cerebral cortex in which the injected contrast agent accumulated without washing out during the acquisition of the dynamic sequence was significantly lower in Ab-CSF—infected rats than in CSF-infected and CSF/blood-infected rats (P = .016 and P = .0004, respectively) (figure 2). Compared with control rats, the CSF/blood-infected and CSF-infected rats both had significantly increased areas with contrast accumulation (P = .004 and P < .0001, respectively), whereas the increase was not significant in Ab-CSF—infected rats (P > .05).

Expansion of the Brain Ventricles Measured 28 h after Infection

Compared with that noted in control rats, the brain ventricle size in Ab-CSF—infected (P = .017), CSF/blood-infected (P < .0001), and infected control rats (P = .0006) was significantly larger (figure 3). Ventricle size did not differ significantly between CSF-infected and CSF/blood-infected rats (P > .05); however, in both groups, it was significantly enlarged compared with the ventricle size in Ab-CSF—infected rats (P = .004 and P = .04, respectively).

Considering that Ab-CSF—infected rats exhibited improved clinical disease scores, it is noted that this parameter was significantly associated with ventricle size among CSF-infected and CSF/blood-infected rats (Spearman's rank correlation, ρ = 0.71; P = .0003). Motor performance score was not significantly associated with ventricle size among these groups of rats (P > .01).

Measurement of Edema 28 h after Infection

Edema in the cerebral cortex, white matter, and jaw muscle was measured 28 h after infection, by use of the ADC (table 3).

Table 3
View larger version:
Table 3

Measurements of the apparent diffusion coefficient in the cerebral cortex, white matter, and jaw muscle.

Cerebral cortex. Compared with controls, the ADC in the cerebral cortex of Ab-CSF—infected (P = .009) and CSF/blood-infected rats (P = .034)—but not CSF-infected rats (P > .05)—was significantly increased. No significant differences were found between the Ab-CSF—infected, CFS-infected, or CSF/blood-infected rats (P > .05).

White matter. The ADC in white matter was not found to differ significantly when controls were compared with Ab-CSF—infected, CSF-infected, or CSF/blood-infected rats (P > .05). Among the infected study groups, the ADC was increased for CSF/blood-infected rats, compared with the ADC for CSF-infected rats (P = .032); however, it was not increased, compared with the ADC noted for Ab-CSF—infected rats (P > .05).

Jaw muscle. The ADC in the jaw muscle was higher in controls than in the infected study groups. However, this was significant only when compared with the ADC for CSF/blood-infected rats (P = .022) and not when compared with the ADC for CSF-infected or Ab-CSF—infected rats (P > .05). No significant differences were found between the infected study groups (P > .05).

Reevaluation of Pneumococcal Meningitis inAb-CSF—Infected Rats at 38 h after Infection

The improved clinical status of Ab-CSF—infected rats permitted this group to be studied at a later point in time (at 38 h after infection). The clinical score, motor performance score, bacteremia levels, CSF bacterial counts, and WBC counts were comparable to those noted for CSF-infected and CSF/blood-infected rats evaluated at 28 h after infection (P > .05). Ventricle expansion also was not significantly different from that noted in the other infected study groups (P > .05).

Compared with the Ab-CSF—infected rats studied 10 h earlier, Ab-CSF—infected rats studied at 38 h after infection had clinical and motor performance scores that were significantly increased (P = .003 and P = .0008). Furthermore, the ADC in the cerebral cortex was increased, compared with that noted in Ab-CSF—infected rats at 28 h after infection (P = .013), in controls (P = .0003), and in CSF-infected rats (P = .0008), but not in CSF/blood-infected rats (P > .05). The ADC in white matter was significantly increased compared with that noted in the CSF-infected group (P = .015) but not in comparison with that observed in the other study groups, including the control group (P > .05). The ADC in jaw muscle decreased significantly, compared with that noted in Ab-CSF—infected rats evaluated 10 h earlier (P = .022) and in controls (P = .007) but not compared with that noted in CSF-infected or CSF/blood-infected rats (P > .05).

MRI Evaluation of Brain Pathologic Findings

Gross findings. Gross pathologic findings revealed multiple petechial hemorrhages that were readily visible on the outer cortical surface of infected animals. As illustrated in figure 4, consistent findings characteristic of focal brain pathologic findings were identified on coronal postcontrast-enhanced T1W (figure 4A) and T2W images (figure 4B). These findings included (1) focal increased enhancement (T1W) extending inward from the cortex edge; (2) focal hyperintense “wedges” extending inward from cortex edge (T2W); (3) enhancing single vessels (T1W), which were often observed as the core of the focal enhancements patterns described above; (4) focal dark areas in contrast-enhanced T1W images that appeared to be hyperintense on T2W images; and (5) focal dark areas within a region of enhancement (T1W). Two additional, more-diffuse pathologic findings not included in the pathology score were also observed: hyperintensity in the outer cortex layer (T2W) and broad, diffuse cortex enhancement resulting in a “glowing” appearance (T1W).

Focal findings. The number of rats with focal pathologic findings was similar among Ab-CSF—infected, CSF-infected, CSF/blood-infected, and Ab-CSF—infected rats who had imaging performed again at 38 h after infection (7 of 14 rats, 5 of 10 rats, 7 of 11 rats, and 3 of 8 rats, respectively; P > .05, for groupwise comparison by use of the Fisher's exact test). Brain pathology scores were not significantly different between the infected study groups (median [IQR]: for Ab-CSF—infected rats, 1 [0–5]; for CSF/blood-infected rats, 0 [0–4]; for CSF-infected rats, 1 [0–7]; and for Ab-CSF—infected rats that had imaging performed again at 38 h after infection, 0 [0–3]) (P > .05, for groupwise comparison by use of the Mann-Whitney U test).

Focal pathologic findings and CSF laboratory data. CSF WBC counts differed significantly between rats with and without brain pathologic findings that were detectable by MRI (median [IQR], 6331 × 109 cells/L [3700–11,506 109 cells/L] vs. 2500–109 cells/L [995–5345 109 cells/L]) (P = .016, by Mann-Whitney U test). CSF bacterial counts were not associated with the pathology score (P > .05).

Previous SectionNext Section

Discussion

The multiparameter data presented in the current study are, to our knowledge, the first data to demonstrate that the intracranial pathophysiology in pneumococcal meningitis is significantly related to bacteremia and, therefore, systemic disease. Rats administered Ab, which has previously been shown to efficiently reduce bacteremia in experimental pneumococcal meningitis [3], had significantly attenuated ventricular expansion and reductions in both the area of the cerebral cortex with BBB leakage and the severity of leakage, compared with rats that had bacteremia induced at the same time that CSF infection occurred or in which bacteremia was allowed to develop secondarily. Importantly, pretreatment with Ab significantly improved clinical disease and motor performance scores. Despite these improvements in the pathophysiologic profile of meningitis, CSF inflammation, focal brain pathologic findings, and edema in the cerebral cortex (as determined using the ADC) were unaffected by the attenuation of bacteremia. At 38 h after infection, rats that had received pretreatment with Ab had levels of bacteremia comparable to those in the CSF infected control and blood/CSF-infected rats and exhibited characteristics of advanced disease, limiting follow-up of “pure” meningeal infection. It was observed that, despite the development of bacteremia, none of the Ab-pretreated rats displayed the vivid cortical contrast enhancement—as illustrated in figure 4A (xvi–xviii)—that was seen in the CSF-infected and CSF/blood-infected groups.

Although the additional intravenous infection in CSF/blood-infected rats did not result in significantly increased measurements of ventricle size or BBB leakage, compared with such measurements noted for CSF-infected rats, the degree of cortex edema was increased, compared with that in controls, as was also observed in Ab-CSF—infected rats. The lack of significant findings in comparisons of rats with earlyonset and secondary bacteremia (i.e., the CSF/blood-infected and CSF-infected rats) was probably related to the development of secondary bacteremia early in the course of disease, as was observed elsewhere [13].

In our previous study, in which Ab was evaluated as adjunctive treatment for pneumococcal meningitis, pretreatment with Ab resulted in significant attenuation of the CSF bacterial load without affecting CSF pleocytosis, possibly as a result of a reduced bacterial exchange between the CSF and blood compartments [2, 14]. However, increased phagocytosis in the CSF resulting from Ab passage to the CSF has been suggested by Sande and Täuber [15]. This could indicate that our results are caused by postponed infection rather than by diminished bacteremia; however, the CSF inflammatory response (including WBC counts and the production of interleukin-8) and the extent of focal brain pathologic findings are unaffected by Ab pretreatment/immunization [3, 13]. In addition, CSF bacterial counts in immunized rabbits were comparable to counts determined for unimmunized rabbits until the latter group developed bacteremia [13]. Thus, we believe that the local inflammatory and infectious processes within the brain are not directly influenced by Ab treatment and that systemic effects of Ab treatment appear, therefore, to be of significant importance.

Measurement of the ADC in jaw muscle was performed to indirectly assess a number of physiologic parameters, including muscle perfusion and interstitial volume. When compared with the ADC noted in control animals, the ADC in muscle from infected rats was decreased but the decrease was significant only in the CSF/blood-infected rats and in the Ab-CSF—infected rats that had imaging performed again at 38 h after infection.

Subarachnoid WBC accumulation has been shown to contribute significantly to increased BBB permeability [12, 16]. These findings have been supported by the association between attenuated CSF WBC counts and reduced BBB permeability [17–19]. The present results appear to contrast with previous findings, because the extent of BBB leakage and contrast agent washout was markedly reduced in our Ab-CSF—infected rats, whereas CSF WBC counts were equal to those in study groups with a markedly increased BBB leakage [20]. Our data showing the contribution of bacteremia to BBB breakdown are supported by recent experimental meningitis studies in rabbits and in Toll-like receptor—2—deficient mice [13, 21, 22], and they emphasize the need to consider septicemia as a significant component of the disease.

In agreement with previous results, vasogenic edema was observed in the cortex [23, 24], with the highest ADC values observed among CSF/blood-infected and Ab-CSF—infected rats, despite differences between the 2 groups with regard to the cortical areas with BBB leakage, leakage severity, ventricle size, and clinical appearance. These results are not considered to be contradictory, because, in this acute phase of disease, reduced ventricle size is probably directly associated with reduced intracranial pressure and decreased CSF outflow resistance, thus allowing increased accumulation of water in brain parenchyma as a result of higher perfusion pressure [25, 26]. Furthermore, it appears that vasogenic edema alone may not be directly related to disease severity (as determined by disease score) but may become an important or even critical factor when it develops in combination with ventricular expansion and other pathophysiologic processes. Although improved systemic circulation in Ab-CSF—infected rats is likely to contribute to the pathophysiology described above, indirect assessment of systemic circulation via measurement of the ADC in muscle did not yield significant differences between infected groups.

The presence of visible focal pathologic findings on MRI was not affected by either early-onset bacteremia or Ab treatment. The investigation of parameters that were not influenced by pretreatment with Ab yielded a significant association between high CSF WBC counts and the presence of focal pathologic findings. This is in agreement with suggestions published elsewhere [27, 28]. Importantly, the findings presented in the current study indicate a lack of association between brain injury and the severity of clinical disease, suggesting that the latter is predominantly caused by the systemic disease component. Pretreatment with Ab did result in apparently healthier animals, despite intracranial pathologic findings, whereas rats with early-onset bacteremia presented with increased disease severity, even in the absence of brain pathologic findings that were visible on MRI. This discovery is in accord with previous findings indicating that the administration of an experimental leukocyte blocker increased bacteremia and mortality but did not affect the extent of injury to the cortex [4]. Similar dissociations have been observed elsewhere [11, 12, 29].

Clinically, the outcome of meningitis is commonly reported on a continuum ranging from death to survival with neurological sequelae to good outcome without sequelae. The present study suggests that improved survival or the severity of attenuated clinical disease can be achieved without reducing the incidence of brain injuries and resulting sequelae. Incorporation of radiological methods, especially MRI, can provide refined assessments of adjunctive treatments and their effects on meningitis. Thus, outcome and sequelae and the underlying disease entity (local, systemic, or “mixed”) may be reported, providing further insight into the clinical disease. Furthermore, clinical, paraclinical, and diagnostic measures related to these disease entities should be incorporated into clinical and experimental studies.

In summary, the present study shows that the systemic infection associated with pneumococcal meningitis contributes significantly to BBB breakdown, the dynamics of ventricular expansion, and brain water distribution. Brain injury appears to be caused predominantly by localized infectious and inflammatory factors. Hence, a multicomponent disease process may explain the dissociation between disease severity and brain injury. The results presented in this study indicate that future preclinical and clinical studies should address the different underlying disease mechanisms and perform evaluations that reflect individual systemic and localized disease components.

Previous SectionNext Section

Acknowledgments

We thank Lise Vejby Søgaard for technical support. Statens Serum Institut (available at: http://www.ssi.dk) produces pneumococcal serotype-specific antibodies for diagnostic purposes only (Pneumosera).

Previous SectionNext Section

Footnotes

  • Potential conflicts of interest: none reported.

  • Financial support: Lundbeck Foundation; Augustinus Foundation; Copenhagen HIV Programme; Statens Serum Institut; Direktør Jacob Madsen og Hustru Olga Madsens Fond; Dagmar Marshalls Fond; Scandanavian Society for Antimicrobial Chemotherapy Fondet.

  • Received May 20, 2007.
  • Accepted August 7, 2007.
Previous Section
 

References

  1. ↵
    1. van de Beek D,
    2. de Gans J,
    3. Spanjaard L,
    4. Weisfelt M,
    5. Reitsma JB,
    6. Vermeulen M
    . Clinical features and prognostic factors in adults with bacterial meningitis. N Engl J Med 2004;351:1849-59.
    CrossRefMedlineWeb of Science
  2. ↵
    1. Brandt CT,
    2. Lundgren JD,
    3. Lund SP,
    4. et al
    . Attenuation of the bacterial load in blood by pretreatment with granulocyte-colony-stimulating factor protects rats from fatal outcome and brain damage during Streptococcus pneumoniae meningitis. Infect Immun 2004;72:4647-53.
    Abstract/FREE Full Text
  3. ↵
    1. Brandt CT,
    2. Frimodt-Moller N,
    3. Lundgren JD,
    4. et al
    . Evaluation of antipneumococcal capsular antibodies as adjunctive therapy in experimental pneumococcal meningitis. J Antimicrob Chemother 2006;58:1291-4.
    Abstract/FREE Full Text
  4. ↵
    1. Brandt CT,
    2. Lundgren JD,
    3. Frimodt-Moller N,
    4. et al
    . Blocking of leukocyte accumulation in the cerebrospinal fluid augments bacteremia and increases lethality in experimental pneumococcal meningitis. J Neuroimmunol 2005;166:126-31.
    CrossRefMedlineWeb of Science
    1. Koedel U,
    2. Rupprecht T,
    3. Angele B,
    4. et al
    . My D88 is required for mounting a robust host immune response to Streptococcus pneumoniae in the CNS. Brain 2004;127:1437-45.
    Abstract/FREE Full Text
  6. ↵
    1. Rupprecht TA,
    2. Angele B,
    3. Klein M,
    4. et al
    . Complement C1q and C3 are critical for the innate immune response to Streptococcus pneumoniae in the central nervous system. J Immunol 2007;178:1861-9.
    Abstract/FREE Full Text
  7. ↵
    1. Orihuela CJ,
    2. Fillon S,
    3. Smith-Sielicki SH,
    4. et al
    . Cell wall-mediated neuronal damage in early sepsis. Infect Immun 2006;74:3783-9.
    Abstract/FREE Full Text
  8. ↵
    1. Bogdanski R,
    2. Blobner M,
    3. Becker I,
    4. Hanel F,
    5. Fink H,
    6. Kochs E
    . Cerebral histopathology following portal venous infusion of bacteria in a chronic porcine model. Anesthesiology 2000;93:793-804.
    CrossRefMedlineWeb of Science
  9. ↵
    1. Papadopoulos MC,
    2. Davies DC,
    3. Moss RF,
    4. Tighe D,
    5. Bennett ED
    . Pathophysiology of septic encephalopathy: a review. Crit Care Med 2000;28:3019-24.
    CrossRefMedlineWeb of Science
  10. ↵
    1. Wellmer A,
    2. von Mering M,
    3. Spreer A,
    4. et al
    . Experimental pneumococcal meningitis: impaired clearance of bacteria from the blood due to increased apoptosis in the spleen in Bcl-2-deficient mice. Infect Immun 2004;72:3113-9.
    Abstract/FREE Full Text
  11. ↵
    1. Leib SL,
    2. Kim YS,
    3. Ferriero DM,
    4. Tauber MG
    . Neuroprotective effect of excitatory amino acid antagonist kynurenic acid in experimental bacterial meningitis. J Infect Dis 1996;173:166-71.
    Abstract/FREE Full Text
  12. ↵
    1. Leib SL,
    2. Leppert D,
    3. Clements J,
    4. Tauber MG
    . Matrix metalloproteinases contribute to brain damage in experimental pneumococcal meningitis. Infect Immun 2000;68:615-20.
    Abstract/FREE Full Text
  13. ↵
    1. Østergaard C,
    2. O'Reilly T,
    3. Brandt C,
    4. Frimodt-Moller N,
    5. Lundgren JD
    . The influence of the blood bacterial load on the meningeal inflammatory response in Streptococcus pneumoniae meningitis. BMC Infect Dis 2006;6:78-X.
    CrossRefMedline
  14. ↵
    1. Brandtzaeg P,
    2. Ovstebøo R,
    3. Kierulf P
    . Compartmentalization of lipopolysaccharide production correlates with clinical presentation in meningococcal disease. J Infect Dis 1992;166:650-2.
    Abstract/FREE Full Text
  15. ↵
    1. Sande MA,
    2. Tauber MG
    . Pneumococcal meningitis: current pathophysiologic concepts. Microb Drug Resist 1999;5:295-9.
    CrossRefMedlineWeb of Science
  16. ↵
    1. Wispelwey B,
    2. Lesse AJ,
    3. Hansen EJ,
    4. Scheld WM
    . Haemophilus influenzae lipopolysaccharide-induced blood brain barrier permeability during experimental meningitis in the rat. J Clin Invest 1988;82:1339-46.
    MedlineWeb of Science
  17. ↵
    1. Tuomanen EI,
    2. Saukkonen K,
    3. Sande S,
    4. Cioffe C,
    5. Wright SD
    . Reduction of inflammation, tissue damage, and mortality in bacterial meningitis in rabbits treated with monoclonal antibodies against adhesion-promoting receptors of leukocytes. J Exp Med 1989;170:959-69.
    Abstract/FREE Full Text
    1. Sáez-Llorens X,
    2. Jafari HS,
    3. Severien C,
    4. et al
    . Enhanced attenuation of meningeal inflammation and brain edema by concomitant administration of anti-CD18 monoclonal antibodies and dexamethasone in experimental Haemophilus meningitis. J Clin Invest 1991;88:2003-11.
    MedlineWeb of Science
  19. ↵
    1. Ostergaard C,
    2. Yieng-Kow RV,
    3. Benfield T,
    4. Frimodt-Møller N,
    5. Espersen F,
    6. Lundgren JD
    . Inhibition of leukocyte entry into the brain by the selectin blocker fucoidin decreases interleukin-1 (IL-1) levels but increases IL-8 levels in cerebrospinal fluid during experimental pneumococcal meningitis in rabbits. Infect Immun 2000;68:3153-7.
    Abstract/FREE Full Text
  20. ↵
    1. Paul R,
    2. Winkler F,
    3. Bayerlein I,
    4. Popp B,
    5. Pfister HW,
    6. Koedel U
    . Urokinase-type plasminogen activator receptor regulates leukocyte recruitment during experimental pneumococcal meningitis. J Infect Dis 2005;191:776-82.
    Abstract/FREE Full Text
  21. ↵
    1. Echchannaoui H,
    2. Frei K,
    3. Schnell C,
    4. Leib SL,
    5. Zimmerli W,
    6. Landmann R
    . Toll-like receptor 2.deficient mice are highly susceptible to Streptococcus pneumoniae meningitis because of reduced bacterial clearing and enhanced inflammation. J Infect Dis 2002;186:798-806.
    Abstract/FREE Full Text
  22. ↵
    1. Koedel U,
    2. Angele B,
    3. Rupprecht T,
    4. et al
    . Toll-like receptor 2 participates in mediation of immune response in experimental pneumococcal meningitis. J Immunol 2003;170:438-44.
    Abstract/FREE Full Text
  23. ↵
    1. Koedel U,
    2. Scheld WM,
    3. Pfister HW
    . Pathogenesis and pathophysiology of pneumococcal meningitis. Lancet Infect Dis 2002;2:721-36.
    CrossRefMedlineWeb of Science
  24. ↵
    1. Brandt CT,
    2. Simonsen H,
    3. Liptrot M,
    4. et al
    . BMC Med Imaging. 2008. In vivo study of experimental pneumococcal meningitis using magnetic resonance imaging. (in press).
  25. ↵
    1. Scheld WM,
    2. Dacey RG,
    3. Winn HR,
    4. Welsh JE,
    5. Jane JA,
    6. Sande MA
    . Cerebrospinal fluid outflow resistance in rabbits with experimental meningitis: alterations with penicillin and methylprednisolone. J Clin Invest 1980;66:243-53.
    MedlineWeb of Science
  26. ↵
    1. Malipiero U,
    2. Koedel U,
    3. Pfister HW,
    4. et al
    . TGFβ receptor II gene deletion in leucocytes prevents cerebral vasculitis in bacterial meningitis. Brain 2006;129:2404-15.
    Abstract/FREE Full Text
  27. ↵
    1. Braun JS,
    2. Novak R,
    3. Herzog KH,
    4. Bodner SM,
    5. Cleveland JL,
    6. Tuomanen EI
    . Neuroprotection by a caspase inhibitor in acute bacterial meningitis. Nat Med 1999;5:298-302.
    CrossRefMedlineWeb of Science
  28. ↵
    1. Weber JR,
    2. Tuomanen EI
    . Cellular damage in bacterial meningitis: an interplay of bacterial and host driven toxicity. J Neuroimmunol 2007;184:45-52.
    CrossRefMedlineWeb of Science
  29. ↵
    1. Loeffler JM,
    2. Ringer R,
    3. Hablützel M,
    4. Tauber MG,
    5. Leib SL
    . The free radical scavenger α-phenyl-tert-butyl nitrone aggravates hippocampal apoptosis and learning deficits in experimental pneumococcal meningitis. J Infect Dis 2001;183:247-52.
    Abstract/FREE Full Text
| Table of Contents

This Article

  1. J Infect Dis. (2008) 197 (2): 235-244. doi: 10.1086/524874
  1. AbstractFree
  2. » Full Text (HTML)Free
  3. Full Text (PDF)Free

Classifications

Share

  1. Email this article

Navigate This Article

Current Issue

  1. March 1, 2012 205 (5)
  1. Journal of Infectious Diseases
  1. Alert me to new issues

Most