Optic neuritis: a novel presentation of Schilder's disease.

The clinical features of a 7-year-old girl who presented with unilateral optic neuritis are presented. Magnetic resonance imaging (MRI) showed lesions in the affected optic nerve and the centrum semiovale bilaterally. Biopsy of one of the cerebral lesions was consistent with a diagnosis of Schilder's disease. Visual acuity returned to normal, and the demyelinating MRI lesions improved markedly with corticosteroid treatment. Optic neuritis is a novel mode of presentation in Schilder's disease.

The surgical treatment of Parkinson's disease.

Surgical treatment of Parkinson's disease (PD) can provide gratifying symptomatic improvements for many individuals who suffer from persistent disabling symptoms despite the best available medical management. Current surgical therapies include ablative techniques (thalamotomy and pallidotomy), augmentative techniques (nondestructive) (deep brain stimulation), and restorative techniques (tissue transplantation and gene therapy). Ablative procedures can provide substantial clinical benefit, but the current trend is toward deep brain stimulation, which can provide similar symptomatic improvement in a nondestructive manner. Restorative techniques, such as tissue transplantation and gene therapy, are exciting but have significant obstacles to overcome before their promise can be realized. Until the underlying pathological defect of PD can be identified and treated, surgical intervention is likely to remain important in the symptomatic treatment of this disabling disease.

A prospective study of catheter-related complications of intrathecal drug delivery systems.

Intrathecal drug administration via implanted pump is an effective treatment for intractable pain and spasticity but can be compromised by catheter-related complications. To determine the etiology of catheter-related complications, we have conducted a multicenter, prospective study of the long-term performance of a one-piece catheter system. Data pertaining to catheter-related complications were collected at implant and at specified times during the follow-up period. Catheter implantation characteristics that might affect complications were assessed. Two hundred nine patients were studied at 22 participating centers, with 1764 cumulative patient-months of catheter experience. Forty-nine catheter system complications occurred in 37 patients (7 complications related to the catheter itself, and 42 complications related to the implantation procedure). The 9-month complication-free "survival" rate was 78.9% overall (95.5% for the catheter itself). No specific catheter implantation characteristics were associated with the occurrence of complications. These data indicate that the incidence of complications for a one-piece catheter system is similar to that of commercially available two-piece systems, and highlight the need for careful surgical technique during implantation.

The effect of spinal analgesia on visceral nociceptive neurons in caudal medulla of the rat.

UNLABELLED: A population of neurons resident in the caudal ventrolateral medulla are excited by noxious cutaneous and visceral stimuli from large portions of the body. These neurons act as monitors of ascending nociceptive information, and we hypothesized that they would be inhibited by spinally administered analgesics in a clinically relevant fashion. Rats were anesthetized with oxygen/ halothane. The caudal medulla was surgically exposed, and a catheter placed into the intrathecal space overlying the lower thoracic spinal cord via the surgical site. Single medullary neurons were characterized for responses to cutaneous and visceral (colorectal distension) stimuli. The effects of i.v. and intrathecally administered morphine and lidocaine were determined. The intrathecal infusion of morphine for 6 days before testing was also used as a pretreatment. Colorectal distension-evoked responses of medullary nociceptive neurons were inhibited in a dose-dependent, naloxonereversible fashion by intrathecal and i.v. morphine (50% effective dose values: 3.5 and 440 microg/kg, respectively). Intrathecal lidocaine abolished responses to colorectal distension and produced a spinal level at doses producing minimal effects when administered systemically. Prior treatment with an infusion of morphine produced tolerance to the effects of subsequent intrathecal morphine administration. These findings support the use of this preparation as a neurophysiologic model of spinal analgesia. IMPLICATIONS: Neurons in the brainstem, isolated electrophysiologically, were used as whole body monitors of pain-related activity in the rat. As a neurophysiologic model of nociception, this preparation may prove useful for the study of regionally administered analgesics and local anesthetics.

The effect of morphine on responses of nucleus ventroposterolateralis neurons to colorectal distension in the rat.

In 71 halothane-anesthetized rats, we characterized the responses of single neurons in the nucleus ventroposterolateralis (VPL) of the thalamus to a noxious visceral stimulus (colorectal balloon distension; CRD) and studied the effects of intravenous morphine on these responses using standard extracellular microelectrode recording techniques. One hundred nine neurons were isolated on the basis of spontaneous activity. Sixty-four (59%) responded to CRD, of which 52 (81 %) had excitatory and 12 (19%) had inhibitory responses. Neurons showed graded responses to graded CRD pressures (20-100 mmHg), with maximum excitation or inhibition occurring at 80 mmHg. Responses to noxious (pinch, heat) and innocuous (brush, tap) cutaneous stimuli were studied in 95 of the VPL neurons isolated. Eighty-three of these neurons (48 CRD responsive and 35 CRD nonresponsive) (87%) had cutaneous receptive fields, of which 96% were small and contralateral and 4% were large and contralateral or bilateral. Ninety-four percent of these neurons responded to both noxious and innocuous cutaneous stimulation, and 6% responded to only noxious stimulation. No neurons responded solely to innocuous stimulation. Cumulative doses of morphine (0.125, 0.25, 0.5, 1, and 2 mg/kg, i.v) produced statistically significant dose-dependent attenuation of neuronal responses to CRD. Naloxone (0.4 mg/ kg, i.v.) reversed the effects of morphine. Morphine and naloxone had no significant effects on spontaneous activity. These data support the involvement of VPL neurons in visceral nociception and are consistent with a role of VPL in sensory-discriminative aspects of nociception.

The effect of morphine on responses of ventrolateral orbital cortex (VLO) neurons to colorectal distension in the rat.

In 49 halothane-anesthetized rats, we characterized the responses of single neurons in the ventrolateral orbital cortex (VLO) to a noxious visceral stimulus (colorectal balloon distension, CRD), and studied the effects of intravenous morphine on these responses using standard extracellular microelectrode recording techniques. One hundred and four neurons were isolated on the basis of spontaneous activity. Fifty-seven (55%) responded to CRD, of which 32% had excitatory and 68% had inhibitory responses. Neurons showed tendencies toward graded responses to graded CRD pressures (20-100 mmHg), with maximum excitation or inhibition occurring at 80 or 100 mmHg, respectively. Responses to noxious (pinch, heat) and innocuous (brush, tap) cutaneous stimuli were studied in 80 of the VLO neurons isolated. Thirty-three (41%) of these neurons (21 CRD-responsive and 12 CRD-nonresponsive) had cutaneous receptive fields, of which 79% were large and bilateral, 18% were small and bilateral, 3% were small and ipsilateral. Ninety-four percent of these neurons responded only to noxious cutaneous stimulation, 6% responded to both noxious and innocuous stimulation. No neurons responded solely to innocuous stimulation. Cumulative doses of morphine (0.0625, 0.125 and 0.25 mg/kg i.v.) produced statistically significant dose-dependent attenuation of neuronal responses to CRD. Naloxone (0.4 mg/kg i.v.) reversed the effects of morphine. Morphine and naloxone had no significant effects on spontaneous activity. These data support the involvement of VLO neurons in visceral nociception.

Characterization of neurons in the area of the medullary lateral reticular nucleus responsive to noxious visceral and cutaneous stimuli.

In halothane-anesthetized rats, 283 caudal medullary neurons responsive to colorectal distension (CRD) were characterized using extracellular electrodes. Neurons inhibited by CRD (n = 82) were in the area dorsal to the lateral reticular nucleus (LRN). Most neurons excited by CRD (n = 130) were located within or immediately adjacent to the LRN, were excited by noxious heat and/or noxious pinch of at least half the body surface and were called bilateral nociceptive specific (bNS) neurons. bNS neurons had accelerating responses to graded CRD (threshold: 20 +/- 2 mmHg). Ten of twelve bNS neurons tested could be antidromically activated by electrical stimulation of the midline cerebellum. Other neurons excited by CRD (n = 71) had mixed responses to cutaneous stimuli and were generally located in the area dorsal to the LRN. Increases in blood pressure due to intravenous phenylephrine did not significantly alter the spontaneous activity of neurons excited by CRD, but altered spontaneous activity (12 excited, four inhibited) in all neurons tested which were inhibited by CRD. Decreases in blood pressure produced by intravenous nitroprusside produced a reciprocal response in most neurons inhibited by CRD and had a delayed onset (20-30 s after bolus administration) excitatory effect on 21 of 27 units excited by CRD. Combined with other studies, these data suggest a role for neurons within and adjacent to the LRN in the modulation of visceral nociception. They also implicate a role for the cerebellum in visceral nociceptive processing.

The development of tolerance to intrathecal morphine in rat models of visceral and cutaneous pain.

The development of tolerance to intrathecal morphine was studied in rats chronically implanted with intrathecal catheters connected to osmotic minipumps. Measures of cutaneous nociception were the hot plate (HP) and tail flick (TF) tests. Measures of visceral nociception were visceromotor (VM) responses to ramped colorectal distension (CRD) and cardiovascular (CV) responses to phasic colorectal distension. Tolerance to a continuous infusion of 6 or 20 nmol/h of morphine sulfate developed over 6 days. A significant reduction in the dose-dependent effects of intrathecal morphine in the TF and HP tests and VM and CV responses to CRD occurred in rats receiving continuous infusions of morphine. The development of tolerance to intrathecal morphine was similar in both cutaneous and visceral models.

Management options in thoracolumbar burst fractures.

BACKGROUND: Both surgery and recumbency have been adopted in the treatment of spinal fractures. Herein we present the indications for each, and our experience with thoracolumbar junction (T12, L1 and L2) burst fractures. METHODS: Sixty-eight patients with thoracolumbar burst fractures were treated operatively in 36 cases, and nonoperatively in 32 with recumbency for 1-6 weeks. Treatment was based on clinical and radiological criteria. Eighty-one percent of the recumbency patients, but only 14% of the surgical patients were intact on admission. Patients were followed for a mean+/-SD of 9+/-10 months in the recumbency group, and 21+/-21 months in the surgical group. RESULTS: Neurological improvement and progressive angular deformity occurred in both groups. The cost of recumbency in our patients was nearly half that of those who required surgery, though the length of hospitalization between the two groups was similar at 1 month +/-2 weeks. CONCLUSION: The above study emphasizes that the selection of operative versus nonoperative treatment in burst fractures should not be random but based on clinical as well as radiological criteria. Recumbency is favored in patients who are intact, with angular deformity less than 20 degrees , a residual spinal canal greater than 50% of normal, and an anterior body height exceeding 50% of the posterior height. Surgical intervention is generally indicated in patients with partial neurological deficit, and those with severe instability.

Computed imaging-assisted stereotactic brain biopsy: a risk analysis of 225 consecutive cases.

BACKGROUND: Treatment strategies for intracranial mass lesions are most effective when based upon histopathological diagnoses. Image-guided stereotaxy has provided the means to sample tissue from small or deeply seated intraparenchymal lesions with a relatively high degree of safety and accuracy. Although procedural complications are infrequent, devastating neurological sequelae may result from hemorrhage or direct trauma. This study was undertaken to identify factors that may confer an increased risk of morbidity from stereotactic brain biopsy. METHODS: Two hundred twenty-five consecutive computer-assisted stereotactic brain biopsy procedures were reviewed. Patient age averaged 47.4 years (range, 3-84 years); gender ratio was approximately 2:1 (male:female). Pre-existing medical conditions were identified in nearly half of the cohort. 61.3% of biopsied lesions were lobar; the remainder (38.7%) were "deep-seated" (thalamus, basal ganglia, pineal, hypothalamus, cerebellum, brainstem). Glial tumors accounted for the majority (44.4%) of biopsied lesions; metastases (12.9%) and lymphoma (11.6%) were also relatively common. Demographical, anatomical, surgical, and histological data were compiled and putative risk factors for morbidity identified. These variables were then subjected to univariate and logistic regression analyses to determine their significance as independent predictors of operative risk. RESULTS: Twelve patients suffered complications as a consequence of the biopsy procedure (eight from hemorrhage, four from direct trauma). Major morbidity (hemiparesis, aphasia, obtundation) occurred in eight patients (3.6%). Three patients (1.3%) suffered minor morbidity (transient, mild neurological deficits). One operative fatality occurred (0.4%). An increased risk of morbidity was associated with the preoperative use of antiplatelet agents, chronic corticosteroids, deep-seated lesions, malignant gliomas, and a greater number of biopsy attempts (p < 0.05). Factors not conferring increased morbidity included gender, age, pre-existing illness, extracranial malignancy, cardiac disease, hypertension, diabetes, HIV status, and instrument used to procure the specimen. CONCLUSIONS: Complications arising from stereotactic brain biopsy are infrequent but can be disastrous. Operative risk is a function of several independent variables, including lesion properties (location, histology), preoperative pharmacological therapy (corticosteroids, antiplatelet agents), and operative technique. This analysis suggests that the morbidity of stereotactic brain biopsy may be minimized by risk factor modification.

Aging and impression formation: the impact of processing skills and goals.

Two studies assessed age differences in representations and judgments about people. Our specific interest was in examining how presumed age-related changes in processing efficiency and motivation affected performance in an impression formation task. Consistent with age-related declines in processing efficiency, we found that increasing age was associated with: (a) no change in the processing of evaluative information; (b) less use of specific traits to organize impressions; (c) poorer memory for behavioral information, especially when it contradicted expectations; and (d) less systematic relationships between memory and judgments. We also found, however, that more meaningful task goals and a focus on individual behaviors resulted in reduced age differences in the nature of representations about the target person.

Complications of posterior articular mass plate fixation of the subaxial cervical spine in 43 consecutive patients.

STUDY DESIGN: A retrospective study evaluating the complications in 43 consecutive patients treated with posterior lateral mass plate fixation of the subaxial cervical spine. OBJECTIVES: To determine the surgical complications of applying posterior lateral mass plates, to correlate complication rate with surgical technique, and to determine fusion rate. SUMMARY OF BACKGROUND DATA: Posterior cervical plate stabilization is a viable alternative to more traditional wiring techniques. Advantages include superior internal stability and no requirement for intact posterior spinal elements. METHODS: Records of 43 consecutive patients who underwent posterior articular mass plate fixation were independently reviewed to identify associated complications. The average follow-up was 25 months (range, 1 to 63 months). The most common indications for surgery were posttraumatic instability (n = 22) and instability after multilevel laminectomy (n = 14). Four patients had difficult spinal disorders requiring a combined anterior and posterior approach. RESULTS: Two hundred eighty-one screws were inserted in the cervical spine (average, 7 screws per patient). There were no complications associated with placement of the screws (i.e., root injury or vertebral artery injury). Complications included two cases in which patients had superficial wound infections and one in which the patient had a spinal epidural hematoma requiring evacuation. In one patient, hardware failure required an anterior approach to correct a progressive angulation. CONCLUSIONS: With the authors' surgical technique, lateral mass plate fixation is safe and effective. There were no nerve root or vertebral artery injuries.

The effect of morphine on responses of mediodorsal thalamic nuclei and nucleus submedius neurons to colorectal distension in the rat.

In halothane-anesthetized rats, we characterized the responses of single neurons in the nuclei of medial thalamus (MT), specifically the mediodorsal thalamic nucleus (MD) and the nucleus submedius (Sm), to a noxious visceral stimulus (colorectal balloon distension, CRD), and studied the effects of intravenous morphine (Mor) on these responses using standard extracellular microelectrode recording techniques. 62 MD and 46 Sm neurons were isolated on the basis of spontaneous activity. 47 of the MD neurons (76%) responded to CRD, of which 70% had excitatory and 30% had inhibitory responses. 38 of the Sm neurons (83%) responded to CRD, of which 89% had excitatory and 11% had inhibitory responses. Responses of MD and Sm neurons excited by CRD were related significantly to distension pressure (20-100 mmHg), with maximum excitation occurring at 60 and 100 mmHg, respectively. MD neurons inhibited by CRD also had graded responses to graded CRD, with maximum inhibition occurring at 80 mmHg. The responses to noxious (pinch, heat) and nonnoxious (tap, brush) cutaneous stimuli were studied in 59 of the MD and 44 of the Sm neurons isolated. 22 of the MD neurons (37%) studied had cutaneous receptive fields, of which 59% were large and bilateral, 41% were small and usually contralateral receptive fields. 55% of these neurons were nociceptive-specific, 45% responded to both noxious and nonnoxious cutaneous stimulation. 29 of the Sm neurons (66%) studied had cutaneous receptive fields, of which 72% were large and usually bilateral, 14% were small and bilateral, 14% were small and contralateral receptive fields. 90% of these neurons were nociceptive-specific, 10% responded to both noxious and nonnoxious stimulation. No MD or Sm neurons responded exclusively to nonnoxious cutaneous stimulation. Mor (0.125, 0.25, 0.5 and 1 mg/kg I.V.) attenuated MD and Sm neuronal excitatory responses to CRD in a dose-dependent fashion, abolishing evoked activity with a dose of 0.5 mg/kg (p < 0.05) and 1 mg/kg (p < 0.05), respectively. Naloxone (0.4 mg/kg I.V.) reversed the effects of Mor. Mor and naloxone had no effects on spontaneous activity. These data support the involvement of MD and Sm neurons in visceral nociception, and are consistent with a role of Sm in affective-motivational, and MD in both sensory-discriminative and affective-motivational aspects of nociception.

Idiopathic spinal epidural lipomatosis.

OBJECTIVE: Spinal epidural lipomatosis (SEDL) is a rare disorder often associated with the administration of exogenous steroids or the elevation of endogenous steroids. Spinal epidural lipomatosis develops in some patients in the absence of elevated steroid levels. The limited information known about idiopathic SEDL comes predominantly from isolated case reports. We proposed to study our experience with idiopathic SEDL and to review the literature. METHODS: We identified eight symptomatic patients with idiopathic SEDL treated at our institution, which is the largest series reported. All patients were male and obese by body mass index (> 27.5 kg/m2). The mean age of the patients was 35.4 years. Idiopathic SEDL was equally distributed between the thoracic and lumbar spine. Six patients underwent laminectomy and fat debulking with good postoperative results; two patients were treated with a weight loss diet, which resulted in the relief of symptoms after losing > 15 kg each. RESULTS AND CONCLUSION: A review of our patients in conjunction with other reported cases reveals the following: 1) idiopathic SEDL occurs almost exclusively in the obese population; 2) idiopathic SEDL seems to occur with equal frequency between the thoracic and lumber spine; 3) a strong male predominance exists; 4) thoracic SEDL presents at an earlier age compared with lumbar SEDL; 5) surgical decompression remains the treatment of choice for the immediate relief of symptoms. Our experience suggests that idiopathic epidural lipomatosis may be a pathological entity that has been underdiagnosed.

Chymopapain-induced reduction of proinflammatory phospholipase A2 activity and amelioration of neuropathic behavioral changes in an in vivo model of acute sciatica.

The mechanism of action underlying chymopapain (Chymodiactin) chemonucleolysis remains obscure. Radiographic studies suggest that chymopapain does not alter disc fragment size acutely; nonetheless, patients often report symptom resolution within a few days, even hours, of treatment. The authors postulate that, in addition to its chemonucleolytic action, chymopapain may possess antiinflammatory properties. To test this hypothesis, the authors assessed the ability of chymopapain to modulate the activity of the proinflammatory enzyme phospholipase A2 (PLA2) and to ameliorate behavioral changes associated with inflammatory neuropathy in an in vivo model of sciatica. Thirty-nine male Fischer rats were randomly assigned to one of three treatment groups: 1) saline, 2) betamethasone, or 3) chymopapain. All of the rats underwent unilateral sciatic nerve ligation with loose chromic gut suture to induce inflammatory mononeuropathy. The animals were tested for thermal and mechanical hyperalgesia on Days 0 (preoperation), 7 (pretreatment), and 14 (prior to death). Three animals were killed on Day 0 to determine the baseline PLA2 activity within unmanipulated rat sciatic nerves. On Day 7, three animals from each group were killed to assess PLA2 activity prior to treatment. The remainder were given a single infusion of saline, betamethasone (0.3 mg/kg), or chymopapain (100 pKat U) around the inflamed nerve. On Day 14, the remaining animals were killed and their sciatic nerves were removed. The tissue was homogenized and the PLA2 activity was determined using [14C]arachidonate-labeled Escherichia coli phospholipid membrane as a substrate. Lipids were extracted and separated by thin-layer chromatography. All animals developed behavioral changes consistent with inflammatory mononeuropathy 24 to 72 hours postoperatively; these included gait disturbance, flexion deformity, and hyperalgesia of the involved hindlimb. The degree of mechanical and thermal hyperalgesia was comparable between groups at Day 7. By Day 14, the thermal hyperalgesia had resolved; the mechanical hyperalgesia was less evident in the betamethasone- and chymopapain-treated groups than in the saline-treated controls (p = 0.003; saline- vs. chymopapain-treated groups p = 0.004; saline- vs. betamethasone-treated groups p = 0.008). The mean PLA2 activity at baseline (Day 0) was 11.6 +/- 4.9 nmol phospholipid hydrolyzed per minute per milligram of protein. The PLA2 activity at Day 7 was 74.4 +/- 18.2 (ligated side) and 21.2 +/- 11.7 (nonligated side). At Day 14, PLA2 activity was reduced in the chymopapain- (47.8 +/- 12.3) and betamethasone- (39.7 +/- 9.5) treated groups compared with the saline control group (62.3 +/- 11.2), (saline- vs. chymopapain-treated groups p < 0.05; saline- vs. betamethasone-treated groups p < 0.01). The PLA2 activity in nonligated specimens was 18.6 +/- 10.1. These data indicate that chymopapain exhibits antiinflammatory properties in vivo, reducing PLA2 activity and ameliorating mechanical hyperalgesia in this model of inflammatory sciatic neuropathy.

Symptomatic intrasellar hemangioblastoma in a child treated with subtotal resection and adjuvant radiosurgery. Case report.

The first documented case of a symptomatic intrasellar hemangioblastoma is described, occurring in an 11-year-old girl with stigmata of von Hippel-Lindau disease who presented with headaches, progressive bitemporal hemianopsia, and adenohypophysial dysfunction. A subtotal resection of the lesion was achieved with two separate surgical procedures: a transsphenoidal approach and a subfrontal craniotomy. Subsequent growth of residual tumor was treated with combined conventional radiotherapy and stereotactic radiosurgery. Two years following completion of these adjuvant therapies, no residual tumor was evident on magnetic resonance imaging. Previous experience with hemangioblastoma in this region, as well as the rationale for radiotherapy in the treatment of incompletely resected lesions, is reviewed.

Spinal epidural hematoma following coronary thrombolysis with tissue plasminogen activator. Report of two cases.

Two cases of spinal epidural hematoma following intravenous administration of recombinant tissue-type plasminogen activator are presented. Both patients received thrombolytic therapy for acute myocardial infarction; back pain and progressive neurological dysfunction ensued, secondary to spinal cord compression caused by epidural hematoma. Both individuals underwent emergency surgery for decompression and hematoma evacuation, resulting in improvement in neurological function. The current status of thrombolytic therapy is reviewed, with emphasis on complications of therapy that require neurosurgical intervention.

Responses of neurons in ventrolateral orbital cortex to noxious visceral stimulation in the rat.

In pentobarbital-anesthetized rats, responses of single neurons in ventrolateral orbital cortex (VLO) to noxious visceral (colorectal distension, CRD) and cutaneous stimulation were recorded. Of 71 neurons identified on the basis of spontaneous activity, 44 responded to CRD. CRD caused inhibition of neuronal activity in 38, facilitation of activity in four and 'mixed' responses in two of these cells. Cutaneous receptive fields were identified in 31 CRD-responsive and 10 CRD-non-responsive neurons. Cutaneous receptive fields were large and bilateral. 25 CRD-responsive cells responded only to noxious cutaneous stimulation, six had wide dynamic range responses. Six CRD-non-responsive cells responded only to noxious stimuli, four had wide dynamic range responses. No VLO neuron responded only to innocuous stimuli. These data are consistent with involvement of VLO in visceral nociception, possibly in non-discriminative aspects of nociception.

Diagnosis of recurrent brain tumor: value of 201Tl SPECT vs 18F-fluorodeoxyglucose PET.

OBJECTIVE: This prospective study was designed to compare the sensitivity and specificity of a relatively simple examination, 201Tl chloride single-photon emission CT (SPECT), with a more complex examination, 18F-fluorodeoxyglucose (FDG) positron emission tomography (PET), in patients thought to have recurrent brain tumor. Because both agents have been shown to be markers of viable tumor, we hypothesized that their sensitivity and specificity should be the same. SUBJECTS AND METHODS: Nineteen patients with evidence of recurrent tumor on CT or MR images were studied with both 201Tl SPECT and FDG PET imaging. Two patients were examined twice, so a total of 21 studies were evaluated. The 201Tl SPECT and FDG PET examinations were performed on the same day in 17 patients, and the remaining four examinations were done within 1 week of one another. Three reviewers independently interpreted each Tl SPECT and PET scan. Inappropriate regional increases in 201Tl or FDG activity were considered indicative of tumor recurrence. Sensitivity and specificity values were based on biopsy results and clinical follow-up. The final diagnosis was tumor recurrence in 16 cases and radiation necrosis in 5 cases. The relationship of scan results to survival was analyzed. RESULTS: The sensitivity and specificity of the 201Tl examination for detecting tumor recurrence were 11 (69%) of 16 and two (40%) of five, respectively; values for the FDG PET examination were 13 (81%) of 16 and 2 (40%) of 5, respectively. In patients with recurrent tumors less than 1.6 cm in size, results were false-negative in four 201Tl SPECT examinations and three FDG PET studies. All tumor lesions 1.6 cm or larger (n = 8) were detected. Agreement among the three nuclear medicine specialists was complete for each of the 201Tl SPECT scans. There was disagreement on the interpretation of five (24%) of the 21 FDG PET scans, which was resolved by consensus. Scintigraphic findings did not correlate with patients' survival times. CONCLUSION: We were unable to detect a statistically significant difference in sensitivity or specificity between the 201Tl SPECT and FDG PET scans. Both techniques were sensitive for tumor recurrence with lesions less than 1.6 cm or larger. However, given the greater availability, simplicity, and ease of interpretation and the lower cost of the 201Tl SPECT studies, this technique should be considered for detection of tumor recurrence with lesions that are demonstrated to be 1.6 cm or larger on CT or MR examinations.