Experimental xenon enhancement with CT imaging: cerebral applications

Prediction of optimal contrast times post-imaging agent administration to inform personalized fluorescence-guided surgery

SIGNIFICANCE

Fluorescence guidance in cancer surgery (FGS) using molecular-targeted contrast agents is accelerating, yet the influence of individual patients' physiology on the optimal time to perform surgery post-agent-injection is not fully understood.

AIM

Develop a mathematical framework and analytical expressions to estimate patient-specific time-to-maximum contrast after imaging agent administration for single- and paired-agent (coadministration of targeted and control agents) protocols.

APPROACH

The framework was validated in mouse subcutaneous xenograft studies for three classes of imaging agents: peptide, antibody mimetic, and antibody. Analytical expressions estimating time-to-maximum-tumor-discrimination potential were evaluated over a range of parameters using the validated framework for human cancer parameters.

RESULTS

Correlations were observed between simulations and matched experiments and metrics of tumor discrimination potential (p  <  0.05). Based on human cancer physiology, times-to-maximum contrast for peptide and antibody mimetic agents were <200  min, >15  h for antibodies, on average. The analytical estimates of time-to-maximum tumor discrimination performance exhibited errors of <10  %   on average, whereas patient-to-patient variance is expected to be greater than 100%.

CONCLUSION

We demonstrated that analytical estimates of time-to-maximum contrast in FGS carried out patient-to-patient can outperform the population average time-to-maximum contrast used currently in clinical trials. Such estimates can be made with preoperative DCE-MRI (or similar) and knowledge of the targeted agent's binding affinity.

[Application of collimator broad correction three dimensional ordered subsets expectation maximization for regional cerebral blood flow measurement]

Autoradiography (ARG) has been used for quantitative analysis of the cerebral blood flow using ¹²³I-IMP, and the regional cerebral blood flow (rCBF) can be assessed more accurately with scatter and attenuation correction. Currently, the filtered back projection (FBP) method is generally used for image reconstruction. However, we anticipate obtaining more accurate rCBF by the ordered subsets expectation maximization method with collimator broad correction three dimensional ordered subsets expectation maximization (3D-OSEM). In the present study, we optimized the processing conditions to quantify rCBF using the 3D-OSEM method and compared them with the FBP method. Regarding the method, we determined the subsets and iteration, compared rCBF values using a profile curve, and compared them with the rCBF values obtained by the XeCT (Xenon-enhanced computed tomography)/CBF method. We found that in the 3D-OSEM method using 90 direction collection and 1.72 mm/pixel, the most accurate image was obtained around subset 9 and iteration 10. In addition, as compared to the profile curve and the XeCT/CBF method, the thalamus rCBF was high in the 3D-OSEM method with a good correlation with that of the XeCT/CBF. Accordingly, we concluded that the 3D-OSEM method can improve the decrease in rCBF due to blurring of the distance between the source (i.e., a structure located in the central part of the brain such as the thalamus and the collimator).

Quantitative dependence of MR signal intensity on tissue concentration of Gd(HP-DO3A) in the nephrectomized rat

Cardiac-gated SE 20/224 +/- 20 MR images were obtained from nephrectomized rats before and after intravenously administering 153Gd-Gd(HP-DO3A). The concentration of Gd, [Gd], was linear in dose in myocardium, skeletal muscle, and blood. Under steady-state conditions, where d[Gd]/dt = 0, image intensities (IIN) in regions of interest were compared with the measured [Gd]. IIN was linear in myocardium at less than or equal to 0.61 mumol/g-myocardium (less than or equal to 0.5 mmol/kg dose) and in skeletal muscle at less than or equal to 0.63 mumol/g-muscle (less than or equal to 0.75 mmol/kg). Above 0.6 mumol Gd/g-tissue, IIN did not increase further. The in vivo data were consistent with measured ex vivo and in vivo relaxivities. A 29% greater slope for IIN versus [Gd] in myocardium [14,439 +/- 4350 IIN (mumol/g)] than in muscle [10,258 +/- 5,296 IIN/(mumol/g)] was attributed to a significant difference in blood content: 25% versus 2% weight blood in myocardium and skeletal muscle, respectively. Two components were apparent from plots of ex vivo 1/T1 versus [Gd] in myocardium and muscle, and only one for blood.

Vertebrobasilar insufficiency: stable xenon computed tomography-cerebral blood flow study in posterior circulation revascularization

Preoperative and postoperative local cerebral blood flow were measured by the stable xenon computed tomography-cerebral blood flow technique in 15 patients with vertebrobasilar insufficiency. The surgery included end-arterectomy or angioplasty of the vertebral artery (five cases), superficial temporal artery-superior cerebellar artery anastomosis (eight cases), and superficial temporal artery-posterior cerebral artery anastomosis (two cases). Fourteen (93.3%) of the 15 patients improved in the post-operative period. Low local cerebral blood flow in the ischemic area without infarction manifested a constant and significant increase postoperatively. In summary, the stable xenon computed tomography-cerebral blood flow technique is thought to be a simple and useful method for assessing local cerebral blood flow in posterior circulation perioperatively.

The role of noise in the measurement of cerebral blood flow and partition coefficient using xenon-enhanced computed tomography

Monte Carlo simulations have been used to study the accuracy which can be expected in the quantification of blood flow and the partition coefficient using xenon-enhanced computed tomography in the presence of noise. We have demonstrated that the markedly asymmetric frequency distribution of estimates increases in size rapidly with an increase in the standard error of the input CT data. On the basis of our results, we recommend that controllable sources of noise (eg. CT number drift) be corrected and that estimates be obtained by averaging CT data and then fitting, rather than averaging blood flow and partition coefficients derived from individual pixels, as the latter procedure results in the introduction of considerable bias.

Stable xenon versus radiolabeled microsphere cerebral blood flow measurements in baboons

Regional cerebral blood flow was simultaneously determined using the stable xenon computed tomographic and the radioactive microsphere techniques over a wide range of blood flow rates (less than 10-greater than 300 ml/100 g/min) in 12 baboons under conditions of normocapnia, hypocapnia, and hypercapnia. A total of 31 pairs of determinations were made. After anesthetic and surgical preparation of the baboons, cerebral blood flow was repeatedly determined using the stable xenon technique during saturation with 50% xenon in oxygen. Concurrently, cerebral blood flow was determined before and during xenon administration using 15-microns microspheres. In Group 1 (n = 7), xenon and microsphere determinations were made repeatedly during normocapnia. In Group 2 (n = 5), cerebral blood flow was determined using both techniques in each baboon during hypocapnia (PaCO2 = 20 mm Hg), normocapnia (PaCO2 = 40 mm Hg), and hypercapnia (PaCO2 = 60 mm Hg). Xenon and microsphere values in Group 1 were significantly correlated (r = 0.69, p less than 0.01). In Group 2, values from both techniques also correlated closely across all levels of PaCO2 (r = 0.92, p less than 0.001). No significant differences existed between the slopes or y intercepts of the regression lines for either group and the line of identity. Our data indicate that the stable xenon technique yields cerebral blood flow values that correlate well with values determined using radioactive microspheres across a wide range of cerebral blood flow rates.

Xenon computed tomography measuring cerebral blood flow in the determination of brain death in children

Local cerebral blood flow was measured using stable xenon computed tomography in 21 children, 10 of whom were clinically brain dead and had electrocerebral silence as determined by electroencephalography. Radioisotopic brain scanning in 9 patients showed no visible cerebral activity in all patients and minimal residual sagittal sinus activity in 4. In this population, mean cerebral blood flow as measured by xenon computed tomography was 1.3 +/- 1.6 ml/min/100 gm. Respiratory support was discontinued in 8 patients, and 2 patients had cardiac arrest. Eleven profoundly comatose children who did not meet all clinical criteria for brain death and who had markedly suppressed but not isoelectric electroencephalograms had an average cerebral blood flow of 33.5 +/- 16.3 ml/min/100 gm. There was no difference in cerebral blood flow in those children who survived (30.4 +/- 16.3 ml/min/100 gm; n = 7) compared with those who died acutely (38.3 +/- 14.3 ml/min/100 gm; n = 4). Two patients who survived had average total flows of only 11.8 and 12.1 ml/min/100 gm. Our findings suggest that in infants and children older than 1 month, (1) cerebral blood flow below approximately 10 ml/min/100 gm is consistent with clinical brain death, (2) cerebral blood flow of less than 5 ml/min/100 gm is consistent with no flow as demonstrated by radionuclide techniques, and (3) flow of more than 10 to 15 ml/min/100 gm is associated with the potential for survival.

Chronic cerebrovascular insufficiency on the xenon CT scan

Several investigators have described CT-negative low flow areas in TIA and stroke patients in the chronic phase. The emission tomographic SPECT image they employed has, in contrast to the xenon CT method, no direct relation to the x-ray transmission CT scan. The aim of our study was to study the phenomenon of CT-negative low flow areas using the xenon CT method, a method especially well suited for such cases. 57 xenon CT examinations were performed in 40 TIA patients. Flow data from brain tissue which appeared to be anatomically intact in a slice 5 cm above the canthomeatal plane were analyzed. In the TIA group, the flow in the gray matter was found to be significantly lower on the clinically affected side: symptomatic side, 61.8 +/- 14.7 ml/100 g/min; asymptomatic side, 66.4 +/- 15.8 ml/100 g/min (p less than 0.001). In the stroke group, the flow in the white matter was also affected; symptomatic side, 31.2 +/- 9.8 ml/100, g/min; asymptomatic side, 35.3 +/- 11.1 ml/100 g/min (p less than 0.01). Gray matter: symptomatic side, 56.1 +/- 11.4 ml/100 g/min; asymptomatic side, 66.0 +/- 11.0 ml/100 g/min (p less than 0.001). The findings indicate that the appearance of CT-negative low flow areas in TIA and stroke patients during the chronic phase is the rule rather than the exception. Flow adaptation to anatomic changes not discernible by CT can be differentiated from clinically relevant flow impairment only by testing the cerebrovascular reserve.

Cerebral blood flow in normal and abnormal sleep and dreaming

Measurements of regional or local cerebral blood flow (CBF) by the xenon-133 inhalation method and stable xenon computerized tomography CBF (CTCBF) method were made during relaxed wakefulness and different stages of REM and non-REM sleep in normal age-matched volunteers, narcoleptics, and sleep apneics. In the awake state, CBF values were reduced in both narcoleptics and sleep apneics in the brainstem and cerebellar regions. During sleep onset, whether REM or stage I-II, CBF values were paradoxically increased in narcoleptics but decreased severely in sleep apneics, while in normal volunteers they became diffusely but more moderately decreased. In REM sleep and dreaming CBF values greatly increased, particularly in right temporo-parietal regions in subjects experiencing both visual and auditory dreaming.

Clinical experience with the use of xenon-enhanced CT blood flow mapping in cerebral vascular disease

Cerebral blood flow mapping with the xenon-enhanced/CT method has become a useful clinical tool in the management of patients with occlusive cerebral vascular disease. Studies involving 4-5 minutes of inhaling a xenon/oxygen mixture (less than or equal to 35%) can now be performed routinely with acceptable patient tolerance and compliance. Four cases with acute and chronic ischemic injuries are reported here to illustrate the manner in which this method has been used to characterize flow pattern in such patients and the relevance of this flow information to clinical patient management.

Progress in cerebrovascular disease: local cerebral blood flow by xenon enhanced CT

A noninvasive technique for measuring local cerebral blood flow (LCBF) by xenon enhanced x-ray transmission computed tomography (CT) has been developed an reported quite extensively in recent years. In this method, nonradioactive xenon gas in inhaled and the temporal changes in radiographic enhancement produced by the inhalation are measured by sequential computed tomography. Time dependent xenon concentrations within various tissue segments in the brain are used to derive both local partition coefficient (lambda) and LCBF. An assessment of this method reveals that although it provides functional mapping of blood flow with excellent anatomic specificity, there are distinct limitations. The assumptions underlying this methodology are examined and problems associated with various potential applications of this technique are discussed.

A simplified in vivo autoradiographic strategy for the determination of regional cerebral blood flow by positron emission tomography: theoretical considerations and validation studies in the rat

A simplified mathematical model is described for the measurement of regional cerebral blood flow by positron emission tomography in man, based on a modification of the autoradiographic strategy originally developed for experimental animal studies. A modified ramp intravenous infusion of radiolabeled tracer is used; this results in a monotonically increasing curvilinear arterial activity curve that may be accurately described by a polynomial of low degree (= zeta). Integrated cranial activity CB is measured in regions of interest during the latter portion of the tracer infusion period (times T1 to T2). It is shown that (See formula: text) where each of the terms A chi is a readily evaluated function of the blood flow rate constant kappa, the brain:blood partition coefficient for the tracer, the cranial activity integration limits T1 and T2, the coefficients of the polynomial describing the arterial curve, and an iteration factor n that is chosen to yield the desired degree of precision. This relationship permits generation of a table of CB vs. kappa, thus facilitating on-line computer solution for blood flow. This in vivo autoradiographic paradigm was validated in a series of rats by comparing it to the classical autoradiographic strategy developed by Kety and associates. Excellent agreement was demonstrated between blood flow values obtained by the two methods: CBF in vivo = CBF classical X 0.99 - 0.02 (units in ml g-1 min-1; correlation coefficient r = 0.966).

Mapping local blood flow of human brain by CT scanning during stable xenon inhalation

Non-invasive methods are described for estimating local cerebral blood flows (LCBF) and partition coefficients (L lambda) during inhalation of 35% stable xenon gas (Xes) in oxygen during CT scanning. After denitrogenation by 100% oxygen breathing, 35% Xes is breathed for 7-8 minutes to minimize subanesthetic effects. Mean changes in brain Hounsfield units extrapolated to 15 minutes were 7.7 units for white matter and 5.3 units for gray matter. They were measured from volumes 80 cubic mm (10 mm2 area x 8 mm), or larger with an EMI 1010 scanner at 1 minute intervals. These data were used for computing LCBFs and L lambdas. Irradiation measured at the center of brain slices was 1 rad per minute. To calculate L lambdas about 6 exposures are necessary, thereafter, each 1 minute scan provides LCBF measurements for 2 adjacent 8 mm slices. Reproducibility for LCBF was r = 0.85 (P less 0.001). Mean L lambdas were 0.86 +/- 0.08 for gray and 1.34 +/- 0.10 for white matter. Normative mean flows (mls/100 g brain/min) were: basal ganglia = 79.6 +/- 9.3, cortex = 82.3 +/- 8.5, white matter = 29.2 +/- 5.9, midbrain tegmentum = 94.3 +/- 14.8, cerebellar cortex = 80.1 +/- 10.9, dorsal pons = 89.3 +/- 4.7, brachium pontis = 35.0 +/- 4.2. Subject finger exercises produced increases of LCBF in contralateral pre-central and post-central gyri. Eye closure decreased flow values limited to the visual system. Gray matter flow values diffusely decreased in non-REM sleep but increased above normal in REM sleep. Cerebral infarction and hemorrhage resulted in zones of zero flow with borders having reduced lambdas and low flows attributed to edema.

Local cerebral blood flow values as estimated with diffusible tracers: validity of assumptions in normal and ischemic tissue

A theoretical assessment is made of the validity of assumptions underlying the theory for estimating local cerebral blood flow with diffusible tracer in the tissue under normal and ischemic conditions. First, Kety's derivation of equations that have commonly been used for calculating local cerebral blood flow values is examined in order to define the problems and assumptions. Second, the brain:blood partition coefficient of diffusible tracer, lambda, and the diffusion-limited factor, m, under normal and ischemic conditions are reviewed. An examination of the literature suggested that contrary to common belief, lambda appears to change very little in ischemia if the tissue constituents remain unchanged, whereas m does change with ischemia if the diffusible tracer used is greatly diffusion-limited in the exchange between brain and blood. Even when a gas with an m value close to unity is used as the diffusible tracer, the prolonged mean transit time of blood through the ischemic tissue would make it difficult to maintain the exponential assumptions. As part of the ischemic tissue became infarcted, which is the case with most stroke patients, so the assumptions of homogeneous perfusion would become invalid. This inevitably renders it difficult to estimate local cerebral blood flow with diffusible tracer in ischemic tissue containing an infarcted mass.