Nonoperative treatment of brain abscesses in selected high-risk patients

Full access

✓ In a series of eight high-risk patients, presumed intraparenchymal pyogenic brain abscesses resolved or decreased markedly in size with antibiotic therapy and without surgery. Cases were selected for antibiotic trial because of multiplicity or location of the abscesses and complicating medical conditions. Patients were alert or lethargic, with stable or improving early clinical courses. Diagnosis was made by computerized tomography (CT) brain scans and corroborative laboratory and diagnostic studies. Cases were closely monitored with serial CT scans. In six cases, bacteria were cultured from blood, cerebrospinal fluid, or fluid aspirated from one of multiple abscesses. Antibiotics were selected for specificity against the cultured organism, or consisted of high-dose penicillin plus chloramphenicol; therapy continued for a median of 7 weeks.

Serial CT scans showed decreased abscess size by a mean of 2.4 weeks (range, 1 to 4 weeks) and resolution of contrast-enhancement and mass effects by a mean of 9.8 weeks (range, 8 to 14 weeks). Abscesses followed to resolution were significantly smaller (mean, 1.7 cm diameter, p < 0.01) than those not responding to similar courses of antibiotic therapy (mean, 4.2 cm diameter). Two patients died of causes unrelated to their abscesses; surviving patients have no neurological deficit (mean follow-up period, 25 months; range, 15 to 37 months). Guidelines are recommended for a treatment protocol to be cautiously applied for high-risk patients with small abscesses. Surgery is recommended in all other cases of pyogenic brain abscesses.

Abstract

✓ In a series of eight high-risk patients, presumed intraparenchymal pyogenic brain abscesses resolved or decreased markedly in size with antibiotic therapy and without surgery. Cases were selected for antibiotic trial because of multiplicity or location of the abscesses and complicating medical conditions. Patients were alert or lethargic, with stable or improving early clinical courses. Diagnosis was made by computerized tomography (CT) brain scans and corroborative laboratory and diagnostic studies. Cases were closely monitored with serial CT scans. In six cases, bacteria were cultured from blood, cerebrospinal fluid, or fluid aspirated from one of multiple abscesses. Antibiotics were selected for specificity against the cultured organism, or consisted of high-dose penicillin plus chloramphenicol; therapy continued for a median of 7 weeks.

Serial CT scans showed decreased abscess size by a mean of 2.4 weeks (range, 1 to 4 weeks) and resolution of contrast-enhancement and mass effects by a mean of 9.8 weeks (range, 8 to 14 weeks). Abscesses followed to resolution were significantly smaller (mean, 1.7 cm diameter, p < 0.01) than those not responding to similar courses of antibiotic therapy (mean, 4.2 cm diameter). Two patients died of causes unrelated to their abscesses; surviving patients have no neurological deficit (mean follow-up period, 25 months; range, 15 to 37 months). Guidelines are recommended for a treatment protocol to be cautiously applied for high-risk patients with small abscesses. Surgery is recommended in all other cases of pyogenic brain abscesses.

A patient with an accessible, encapsulated brain abscess who is deteriorating neurologically, or has an enlarging abscess with significant mass effect should be operated on as soon as possible. However, an initial trial of antibiotic therapy may be warranted in patients who: 1) are considered poor surgical candidates; 2) have multiple abscesses, especially if located distant from one another; 3) have abscesses in a deep or dominant location(s); 4) have concomitant meningitis or ependymitis; or 5) have concomitant hydrocephalus requiring a cerebrospinal fluid (CSF) shunt that might become infected during abscess surgery. The medical approach may also be beneficial to patients who preoperatively show partial abscess resolution attributable to antibiotic therapy.

Over the past 3 years, we have treated six carefully selected patients, diagnosed as having pyogenic brain abscesses, without surgical intervention. Two additional patients underwent surgery. All patients presented with the signs, symptoms, laboratory findings, and other diagnostic indications of a brain abscess. The diagnosis was established on the basis of characteristic computerized tomography (CT) findings.12,13,17 The CT scans* were performed before and after the administration of 300 ml of meglumine iothalamate (Conray 30) in adults, or of 0.5 gm/kg of iodine in children. All patients were either stable or improving during their early clinical course.

Patients were selected for prolonged antibiotic trial because of either medical conditions that increased the risks of surgery, or the presence of multiple abscesses, or lesions located in high-risk regions of the brain. A patient with improving clinical symptoms was included whose treatment was initiated in the cerebritis phase of abscess evolution and who showed early CT evidence of decreased abscess size. The disappearance of all signs of contrast enhancement and mass effect on CT scanning was considered an indication of abscess resolution.

Before advocating antibiotic therapy as a possible alternative to surgery in the treatment of selected patients with brain abscesses, it is essential to establish judicious guidelines for its use. Recent reports have not elucidated either indications for therapy or treatment guidelines.1,10,15,16 The evaluation of our eight patients provides a basis for determining such guidelines.

Illustrative Case Reports
Case 1

This 56-year-old man was well until 16 days before admission to the University of California, San Francisco (UCSF), when he developed acute bacterial endocarditis with coagulase-producing Staphylococcus aureus and atrial fibrillation. He was then treated elsewhere with nafcillin and digitalis. Four days later, he developed right upper-extremity monoplegia and other systemic manifestations of septic emboli. A CT brain scan showed no abnormalities. A left focal seizure occurred the day before admission.

The patient was admitted in cardiogenic shock for emergency replacement of his mitral valve. Nafcillin (12 gm/day intravenously) was continued. His postoperative course was complicated by respiratory insufficiency, congestive heart failure, and L-4 vertebral body osteomyelitis. A CT scan was performed 11 days after admission because he developed dementia and confusion; the scan showed lesions characteristic of brain abscesses in the right frontal and left parieto-occipital regions (Fig. 1). The CT scans made 2 weeks later showed increasing edema around the right frontal lesion, but during the next 2 weeks both abscesses diminished in size. The left parieto-occipital lesion was no longer seen after 4 weeks. Antibiotics were continued for a total of 7 weeks. Eight weeks after the abscesses were initially diagnosed, the CT scan showed marked resolution of the remaining right frontal lesion; at 9 months the scan showed no abnormalities (Fig. 1). The patient's neurological problems have resolved; he was still well at 15 months after diagnosis of his disorder.

Fig. 1.
Fig. 1.

Case 1. Computerized tomography (CT) scans. The initial scan was performed 11 days after admission. Two slices are shown for this and the 3-week scan. The parieto-occipital lesion appeared to have decreased in size by 3 weeks after diagnosis; by 4 weeks, the frontal abscess was definitely smaller and the parieto-occipital lesion was visible. The frontal abscess had resolved markedly by 8 weeks and a CT scan made at 9 months showed no abnormalities.

Case 2

This 18-year-old man with cyanotic congenital heart disease felt well until 5 days before admission, when he developed bifrontal headaches and intermittent febrile episodes. He was admitted to another hospital complaining of a stiff neck, emesis, and a fever (39°C). Examination of CSF, obtained by lumbar puncture, disclosed an opening pressure of 150 mm H2O, 90 white blood cells (WBC's: 70 polymorphonuclear cells, 20 lymphocytes), 33 mg% protein, 71 mg% glucose, and negative Gram's stain and bacterial culture results. A CT scan showed a bilobed abscess in the right parieto-occipital lobe (Fig. 2). He was placed on a regimen of penicillin, chloramphenicol, and cephalothin, and transferred to UCSF.

Fig. 2.
Fig. 2.

Case 2. Initial computerized tomography (CT) scan shows a bilobed abscess in the right parieto-occipital lobe. Intravenous therapy with penicillin and chloramphenicol was initiated. A CT scan 1 week later showed a definite decrease in abscess size; CT abnormalities progressively resolved over the subsequent 3 weeks. Four weeks after abscess diagnosis, the CT scan demonstrated only a slight suggestion of contrast enhancement within an area of decreased attenuation in the right parieto-occipital lobe. The patient died from a cardiac disorder 1 day later. Postmortem evaluation disclosed a persistent, partially collapsed, well formed, bilobed abscess.

On admission, he was fully alert, but had a slightly stiff neck and a headache. A general examination showed a possible defect of the left visual field and a left extensor plantar response. He was placed on a daily regimen of aqueous penicillin G (21 million units intravenously), chloramphenicol (4.8 gm intravenously), and dexamethasone (16 mg orally). His headache disappeared, and a CT scan 1 week after admission showed his abscess was definitely smaller (Fig. 2). Two days later, hematuria and ecchymoses were noted. The platelet count was 14,000/cu mm, due to diffuse intravascular coagulopathy. Over the next 2 weeks it remained at approximately 20,000/cu mm despite multiple platelet transfusions; the hematocrit values decreased from 68% to 47%; and the patient developed intermittent episodes of junctional tachycardia with hypotension. Four weeks after the abscess was diagnosed, he suffered a cardiac arrest and could not be resuscitated. The day before his death, a CT scan showed continued abscess resolution and no other abnormalities (Fig. 2).

The postmortem examination disclosed a biloculated abscess in the right posterior parietal area, extending from the subcortical region to the occipital horn of the right lateral ventricle (Fig. 3); both loculations contained greenish, viscous material. The abscess wall appeared well formed, except in the area closest to the ventricle. The abscess contained necrotic debris, with only sparse acute inflammatory cells and no bacteria. There was no purulent material within the ventricular lumen. Histological evaluation, however, disclosed a small amount of necrotic debris identical to that in the abscess cavity, overlying an area of ependymal disruption adjacent to the abscess-ventricular interface. This finding suggested antemortem spillage of aseptic material from the abscess cavity. The ependyma, subependyma, and choroid plexus were normal. Examination of the leptomeninges revealed only diffuse areas of focal thickening with no inflammatory cells. These observations implied that the patient's death was not attributable to rupture of the abscess.

Fig. 3.
Fig. 3.

Case 2. Postmortem evaluation disclosed a persistent, well formed, bilobed abscess in the right parieto-occipital region. Histological evaluation showed a typical abscess wall with proliferation of mesodermal cells, collagen deposits, and neovascularity.

Case 3

This 38-year-old man suffered a compound, depressed skull fracture in the left parietal region from a hammer blow. The wound was debrided immediately. He received a 5-day course of aqueous penicillin G (24 million units daily, intravenously). Four days after penicillin was discontinued, he developed marked right hemiparesis and expressive dysphasia. A CT scan showed two adjacent, ring-like lesions suggestive of brain abscesses (Fig. 4). The larger lesion was aspirated percutaneously; a culture yielded Propionibacterium species. The patient was placed on chloramphenicol (4 gm daily, intravenously). He remained alert and his symptoms resolved by 2 weeks, when another CT scan disclosed that both abscesses were smaller (Fig. 4). Chloramphenicol was discontinued after a 5-week course, and CT scans documented his progressive improvement (Fig. 4). Both processes resolved completely by 14 weeks. The patient remains clinically well after 26 months of follow-up review.

Fig. 4.
Fig. 4.

Case 3. Serial computerized tomography scans demonstrate a definite decrease in the size of both lesions by 2 weeks, and resolution of all evidence of contrast-enhancing abnormalities by 14 weeks after abscess diagnosis.

Case 4

This 34-year-old woman with a history of heroin abuse presented with acute bacterial endocarditis with coagulase-producing Staphylococcus aureus, and a mild right hemiparesis which was presumed secondary to septic emboli. A radionuclide brain scan (technetium-99m-pertechnetate) showed no abnormality. Treatment with methicillin was initiated. She left the hospital against medical advice on the 6th day. Two months later, she had a grand mal convulsion and was readmitted with mild right hemiparesis and expressive dysphasia. A radionuclide scan, an electroencephalogram (EEG), and a cerebral arteriogram showed an abnormality in the left parietal lobe. A CT scan revealed a typical abscess cavity in the same location (Fig. 5). Intravenous treatment with methicillin (16 gm daily) was instituted again. A CT scan 1½ weeks later disclosed a decrease in abscess size (Fig. 5); her neurological deficits diminished. A CT scan at 2½ weeks showed progressive resolution. At 4 weeks she again left the hospital against advice. A radial nerve palsy prompted her readmission 1 week later. The next day, she underwent a left parietal craniotomy and an abscess was totally removed. Gram-positive bacteria were seen in the purulent contents, but bacterial and fungal cultures of both the contents and the well formed abscess wall showed no growth. Two and 6 months after diagnosis, CT scans showed no significant abnormality; her neurological status was normal. She subsequently died from a different episode of acute bacterial endocarditis.

Fig. 5.
Fig. 5.

Case 4. Serial computerized tomography scans demonstrate a definite decrease in abscess size at 11 days, and progressive resolution the following week. Surgical extirpation of a well formed abscess was carried out 5 weeks after diagnosis; bacterial and fungal cultures of its purulent contents showed no growth despite the presence of Gram-positive organisms.

Three of the other four cases included in this analysis were described briefly elsewhere.1,15 The eight cases are summarized in Tables 1, 2, and 3.

TABLE 1

Abscess location(s) and neurological evaluation for patients demonstrating partial or complete resolution on antibiotics

Case No.Age SexNo. of AbscessesLocationAverage Diam (cm)Neurological Grade*
AdmissionEarly Course      
156 yrs M2rt frontal,2.0IIstable
 lt occipital0.8
218 yrs M1rt occipital3.0IIimproved
338 yrs M2lt parietal,3.0Istable
 lt parietal2.0
434 yrs F1lt parietal4.0Istable
55½ yrs M1rt caudate1.8IIimproved
64½ mos M1lt frontal2.2Istable
722 yrs F2rt occipital,2.5IIimproved
 lt frontal1.0
83 yrs F1rt occipital1.5IIimproved

Grade I = alert; Grade II = lethargic or obtunded.

The abscess was aspirated or excised.

TABLE 2

Presumed etiological factor, presumed bacterial organism, and antibiotic therapy for patients demonstrating abscess resolution without surgery*

Case No.Presumed Etiological FactorPresumed OrganismAntibiotic Therapy
TypeSourceType(s)Daily Dose & Route  Duration (wks)
Individual AgentTotal      
1endocarditisStaphylococcus aureusbloodnafcillin12 gm, iv77
2congenital heart diseaseunknownpenicillin, chloramphenicol21 mil U, iv, 4.8 gm, iv4 44
3traumaPropionibacterium spabscesspenicillin, chloramphenicol24 mil U, iv, 4 gm, iv/po1 56
4endocarditisStaphylococcus aureusbloodmethicillin16 gm, iv88
5unknownα hemolytic StreptococcusCSFampicillin, penicillin8 gm, iv, 7.2 mil U, iv2 57
6meningitisProteus mirabilisCSFchloramphenicol, penicillin0.5 gm, iv, 1.8 mil U, iv1 45
7endocarditisStaphylococcus aureusbloodoxacillin, dicloxacillin15 gm, iv, 2 gm, po6 915
8unknownunknownpenicillin, chloramphenicol4.5 mil U, iv, 1 gm, iv6 66

iv = intravenously, po = by mouth, CSF = cerebrospinal fluid.

Patient died 4 weeks after abscess diagnosis of cause unrelated to abscess.

TABLE 3

Chronology of clinical course*

Case No.First Symptom to Abscess Diagnosis (wks) Abscess Diagnosis to:Follow-Up Period (mos)
First Decr. Size (wks)CT Resolution (wks)    
13 48† 15
21 1died died
3 214 26
49 1.5died died
51 2.58 37
612 311 33
74 §9 25
83 2.59 16
median3 2.59 25.5
mean4.7 2.49.8 25

Duration from first symptom to abscess diagnosis, from abscess diagnosis to first computerized tomography (CT) evidence of decreased abscess size, and to CT resolution of abnormalities and of follow-up period. In Cases 2 and 4, death was unrelated to abscess.

Approximate date of resolution.

Information not relevant, head trauma etiology.

Early CT scan not obtained.

Summary of Cases

Headaches, focal neurological deficits, and seizures were the presenting symptoms in approximately one-half of the patients in this study. One-half of the patients had an elevated peripheral WBC count. Five had lumbar punctures for CSF evaluation: in three, CSF examination revealed increased WBC counts and protein concentrations; in two, glucose concentrations were below normal limits. The frequency of presenting complaints and laboratory abnormalities is essentially identical to those in cases reported previously.15 Abnormalities were seen on all radionuclide brain scans (five), cerebral arteriograms (two), and EEG's (four) obtained for patients treated without surgery.

The patients ranged from 4½ months to 56 years of age; the men outnumbered the women in a ratio of 2 to 1. Five patients had single abscesses, and three had multiple lesions. The abscesses ranged from 0.8 cm to 4.0 cm in diameter. Neurological evaluation was graded as follows: Grade I indicates an alert patient; Grade II, a patient with lethargy or obtundation; Grade III, a patient who responded to painful stimulus only; and Grade IV, a patient with no appropriate response to pain.6,7 On admission, four patients were in Grade I and four in Grade II. The clinical course of all patients was either stable or showed improvement during the antibiotic trial (Table 1).

Bacteria were cultured from the blood, CSF, or one of multiple abscesses in six patients, and in these cases, the antibiotic selected was specific for the organism. In two cases in which an organism was not identified, a combination of penicillin and chloramphenicol was used in large dosages. The median duration of antibiotic therapy was 7 weeks (Table 2).

Symptoms attributable to the abscess or its initiating event were first evident a median of 3 weeks before the abscess was diagnosed. The first definite evidence of a decrease in abscess size was seen within a mean of 2.4 weeks (95% confidence limits of 0 to 4.8 weeks) after diagnosis of the abscess. Abscess resolution was noted on CT scans by a mean of 9.8 weeks (95% confidence limits of 3.8 to 15.8 weeks). No death was attributable to an abscess; surviving patients have remained well for a mean of 25 months (Table 3).

To determine why medical therapy fails in some cases but is successful in others, we compared the sizes of eight abscesses (in six patients) that resolved on medical therapy alone, with four cases in which medical therapy failed. Cases 2 and 4 were excluded from analysis because the patients died or required surgical extirpation before their abscesses resolved completely.

During the past 3 years, four patients treated at UCSF for an intracerebral abscess have received antibiotics for a prolonged period before undergoing a definitive operative procedure. All four had unsuccessful operations shortly after admission. Surgery resulted in partial abscess removal in two cases, aspiration in one, and subdural abscess removal in one case. Bacterial organisms were cultured from specimens taken during this initial operation, and appropriate antibiotics were administered for an additional 5 to 7 weeks in each case. Despite this regimen, the intracerebral abscess enlarged progressively, as shown by serial CT scans. Therefore, these four patients failed to show abscess resolution despite sufficient antibiotic therapy and operative intervention. Subsequent surgery was curative in all cases. In the group treated successfully with antibiotics, the mean diameter of the abscesses before treatment was 1.7 cm; in the failure group the mean diameter was 4.2 cm (Table 4); the difference was significant (p < 0.001 by Student's t-test).

TABLE 4

Effect of abscess size on result of medical treatment*

Result of TreatmentNo. of AbscessesNo. of CasesPretreatment Diameter (cm)
MeanRange   
success861.70.8–2.5
failure444.22.0–6.0

Comparison of initial abscess sizes between patients demonstrating abscess resolution on medical therapy alone (success), and cases with no change or enlargement of abscess despite 5 to 7 weeks of appropriate antibiotics (failure).

Significantly different, p < 0.01 (Student's t-test).

Discussion
Results with Antibiotic Therapy

Several factors may have contributed to the success of antibiotic therapy in this series. First, in four cases therapy was initiated before the abscess had become well encapsulated. In two (Cases 1 and 3), antibiotic treatment was initiated soon after the events probably precipitating the abscesses. In two others (Cases 7 and 8), medical therapy was initiated before a well formed abscess wall was observed on the CT scan. Nevertheless, CT results in all eight cases indicated a well formed abscess wall at some time before the abscess resolved. Heineman, et al.,8 suggested that intracerebral infective processes are probably eradicated more readily when therapy is initiated during the cerebritis phase of abscess evolution. Therefore, administration of antibiotic treatment early in the course of the disease may have contributed to the success of medical therapy in four patients.

The second possible factor was that the abscesses that resolved with antibiotic therapy alone were markedly smaller than those that either fail to resolve or that enlarge during the course of treatment, which suggests that successful medical treatment might be anticipated primarily in the case of abscesses smaller than 3 cm in diameter. Black's study, in which antibiotics did not eradicate the responsible organism(s), despite the accumulation of sufficient concentrations of drug,4 reported significantly larger abscesses than those in the present study.3 It seems probable that the smaller size of the abscesses in our series accounts for our contrasting success.

Third, the current understanding of pharmacokinetics and microbial sensitivity enabled us to select antibiotics that penetrate brain abscesses in sufficient concentrations to be effective against potentially susceptible organisms.4,11 Finally, the CT scan played a pivotal role in our therapy decisions; it is essential to the selection, treatment, and follow-up of patients with brain abscesses.15

The analysis of this series implies that, 95% of the time, the abscesses that will resolve with antibiotic therapy alone demonstrate definite CT evidence of decreased size by 1 month after the initiation of treatment, and resolution of contrast-enhancing abnormalities by 3½ months. The long-term follow-up findings, for a mean period of 25 months, imply that these patients have probably been “cured” of their infective process. However, further observation is necessary to ensure that none of the abscesses recur.

The patient described in Case 2 died 1 day after a CT scan demonstrated resolution of all evidence of contrast-enhancing abnormalities. Nevertheless, postmortem examination disclosed a persistent brain abscess with a well formed wall containing neovascularity. The difference between the CT and postmortem findings demonstrates the limitation of CT scans in documenting abscess resolution.

After antibiotic therapy was completed in our patients, serial CT scans documented continuing abscess resolution in five of six patients followed to cure on medical treatment alone. Therefore, our limited experience suggests that antibiotics should be administered for at least 6 to 8 weeks as primary therapy in unoperated cases, but may then be discontinued even if all abnormalities seen on the CT scan have not resolved entirely.

Proposed Treatment Scheme

Our present protocol for the treatment of pyogenic brain abscesses is as follows: A patient who presents in a Grade III or IV neurological state, or who shows clinical deterioration attributable to the intracranial mass lesion, is operated on as soon as possible. However, if the patient is in Grade I or II, and is stable or improving, antibiotics are administered for 2 weeks to maximize the chance for a well formed abscess wall to develop.15

Whenever possible, antibiotic therapy for bacterial infections should be specific for the responsible organism. A major consideration is the ability of the drug to penetrate the blood-brain barrier. Both penicillin (dosage greater than 10 million units daily, intravenously) and chloramphenicol (dosage of 3 to 4 gm daily, intravenously or orally) are appropriate for treating intracerebral suppurative disease in adults.2,4,5 For children the recommended dosages are 300,000 units/kg/day for penicillin and 50 to 100 mg/kg/day for chloramphenicol.5 In combination, these two drugs are effective against most responsible aerobic and anaerobic bacteria. When a penicillinase-producing organism is suspected, methicillin or a less toxic cogener (preferably nafcillin at a dosage greater than 12 gm daily, intravenously) may be added to, or substituted for, the penicillin.2,5 Our experience suggests that an antibiotic specific for the organism isolated from the blood or the CSF is appropriate as the sole agent in the treatment of a presumed brain abscess.

Corticosteroid therapy is to be avoided whenever possible. Dexamethasone may inhibit the inflammatory reaction in brain surrounding an infective process and decrease the likelihood of fibrous abscess encapsulation.14 Compared to results achieved using the antibiotic alone, dexamethasone and an appropriate antibiotic administered concurrently appeared to be less effective in eradicating an area of cerebritis caused by intracerebral inoculation of bacteria in an animal study.14 Additionally, it has been suggested that dexamethasone reduces antibiotic penetration into the brain abscess.9,12 For these reasons, we administer corticosteroids to patients with purulent intracerebral lesions only if a marked mass effect is evident either clinically or on diagnostic studies. Corticosteroids are continued for as short a time as possible in both operated and unoperated patients.

A CT scan is performed weekly, or more frequently if new symptoms develop. After 2 weeks, the patient is operated on if neurological deterioration, abscess enlargement, or significant mass effect is noted. Operation should be attempted also in all other cases involving patients whose abscess is unchanged in size when there are no relative contraindications to surgery.

If the clinical course is stable or improving, and if there is a relative contraindication to surgery, then antibiotics may be continued for 2 additional weeks (a total of 4 weeks). A prolonged antibiotic trial should be considered only for patients who are poor surgical candidates, or for those whose condition fulfills the following criteria: multiple abscesses (especially those located distant from one another); abscesses in deep or dominant locations; the concurrent presence of ependymitis or meningitis; hydrocephalus requiring a CSF shunt; or CT scan evidence of a decrease in the size of the abscess attributable to antibiotics in the first 2 weeks. The CT scans are obtained weekly and whenever any new symptom is noted. An operation is performed at 4 weeks if no change in the size of the abscess is seen on the CT scans.

Patients who have a demonstrated decrease in abscess size and mass effect should receive antibiotics for a total of 6 to 8 weeks. A CT scan should be performed every 2 to 4 weeks until resolution is observed, then every 2 to 4 months for 1 year to ensure that the abscess has not recurred. Subsequently, a CT scan should be made when any new symptom of disturbance in the central nervous system is noted.

Cautionary Note

Despite the promise of antibiotics in the treatment of patients such as those reported here, the selection of patients for a trial of medical therapy must be monitored carefully to avoid the inclusion of patients who might be more properly treated by surgical intervention. The medical approach offers a noninvasive alternative in the initial treatment of intracranial mass lesions in high-risk patients, but does not subrogate the role of neurosurgical management. The application of surgical techniques and adjuvant medical therapies to control perioperative brain swelling have reduced considerably the risks involved in the neurosurgical treatment of brain abscesses, and, together with the rapid diagnosis and improved localization of brain abscesses now possible using the CT scan, have decreased the mortality rates associated with surgical management of the disease.15 The patient must be followed meticulously, and a neurosurgeon must be prepared to operate when the indication arises.

Acknowledgment

The authors wish to thank Susan Eastwood for editorial assistance in the preparation of this paper.

References

EMI Mark I 160 × 160 matrix scanner manufactured by EMI Tronics, Inc., 3605 Woodhead Drive, Northbrook, Illinois.

This work was supported in part by NIH Trauma Center Program Project Grant GM18470, and NIH Brain Edema Program Project Grant NS14543-01.

Article Information

Address reprint requests to: Mark L. Rosenblum, M.D., The Editorial Office, Department of Neurological Surgery, University of California Medical Center, 350 Parnassus Avenue, Suite 807, San Francisco, California 94143.

© AANS, except where prohibited by US copyright law.

Headings

Figures

  • View in gallery

    Case 1. Computerized tomography (CT) scans. The initial scan was performed 11 days after admission. Two slices are shown for this and the 3-week scan. The parieto-occipital lesion appeared to have decreased in size by 3 weeks after diagnosis; by 4 weeks, the frontal abscess was definitely smaller and the parieto-occipital lesion was visible. The frontal abscess had resolved markedly by 8 weeks and a CT scan made at 9 months showed no abnormalities.

  • View in gallery

    Case 2. Initial computerized tomography (CT) scan shows a bilobed abscess in the right parieto-occipital lobe. Intravenous therapy with penicillin and chloramphenicol was initiated. A CT scan 1 week later showed a definite decrease in abscess size; CT abnormalities progressively resolved over the subsequent 3 weeks. Four weeks after abscess diagnosis, the CT scan demonstrated only a slight suggestion of contrast enhancement within an area of decreased attenuation in the right parieto-occipital lobe. The patient died from a cardiac disorder 1 day later. Postmortem evaluation disclosed a persistent, partially collapsed, well formed, bilobed abscess.

  • View in gallery

    Case 2. Postmortem evaluation disclosed a persistent, well formed, bilobed abscess in the right parieto-occipital region. Histological evaluation showed a typical abscess wall with proliferation of mesodermal cells, collagen deposits, and neovascularity.

  • View in gallery

    Case 3. Serial computerized tomography scans demonstrate a definite decrease in the size of both lesions by 2 weeks, and resolution of all evidence of contrast-enhancing abnormalities by 14 weeks after abscess diagnosis.

  • View in gallery

    Case 4. Serial computerized tomography scans demonstrate a definite decrease in abscess size at 11 days, and progressive resolution the following week. Surgical extirpation of a well formed abscess was carried out 5 weeks after diagnosis; bacterial and fungal cultures of its purulent contents showed no growth despite the presence of Gram-positive organisms.

References

1.

Berg BFranklin GCuneo RNonsurgical cure of brain abscess: early diagnosis and follow-up with computerized tomography. Ann Neurol 3:4744781978Ann Neurol 3:

2.

Black P: Brain abscessKagan BM (ed): Antimicrobial Therapy. Philadelphia: WB Saunders1970244252Antimicrobial Therapy

3.

Black P: Personal communication1979

4.

Black PGraybill JRCharache P: Penetration of brain abscess by systemically administered antibiotics. J Neurosurg 38:7057091973J Neurosurg 38:

5.

Brewer NSMacCarty CSWellman WE: Brain abscess: a review of recent experience. Ann Intern Med 82:5715761975Ann Intern Med 82:

6.

Carey MEChou SNFrench LA: Experience with brain abscesses. J Neurosurg 36:191972J Neurosurg 36:

7.

Garfield J: Management of supratentorial intracranial abscess: a review of 200 cases. Br Med J 2:7111969Garfield J: Management of supratentorial intracranial abscess: a review of 200 cases. Br Med J 2:

8.

Heineman HSBraude AIOsterholm JL: Intracranial suppurative disease. Early presumptive diagnosis and successful treatment without surgery. JAMA 218:154215471971JAMA 218:

9.

Jawetz EMerrill ER: Effect of cortisone on the therapeutic efficacy of antibiotics in experimental infections. Arch Intern Med 93:8508611954Arch Intern Med 93:

10.

Kirmani NTuazon CUOcuin JAExtensive cerebral nocardiosis cured with antibiotic therapy alone. Case report. J Neurosurg 49:9249281978J Neurosurg 49:

11.

Kramer PWGriffith RSCampbell RL: Antibiotic penetration of the brain. A comparative study. J Neurosurg 31:2953021969J Neurosurg 31:

12.

New PFJDavis KRBallantine HT Jr: Computed tomography in cerebral abscess. Radiology 121:6416461976Radiology 121:

13.

Newton THNorman DAlvord ECThe CT scan in infectious disease of the CNSNorman DKorobkin MNewton TH (eds): Computed Tomography. San Francisco: University of California Printing1977317337Computed Tomography

14.

Quartey GRCJohnston JARozdilsky B: Decadron in the treatment of cerebral abscess. An experimental study. J Neurosurg 45:3013101976J Neurosurg 45:

15.

Rosenblum MLHoff JTNorman DDecreased mortality from brain abscesses since advent of computerized tomography. J Neurosurg 49:6586681978J Neurosurg 49:

16.

Rotheram EB JrKessler LA: Use of computerized tomography in nonsurgical management of brain abscess. Arch Neurol 36:25261979Arch Neurol 36:

17.

Stevens EANorman DKramer RAComputed tomographic brain scanning in intraparenchymal pyogenic abscesses. AJR 130:1111141978AJR 130:

TrendMD

Metrics

Metrics

All Time Past Year Past 30 Days
Abstract Views 4 4 4
Full Text Views 94 94 72
PDF Downloads 64 64 38
EPUB Downloads 0 0 0

PubMed

Google Scholar