Yichao Teng

Monitoring cerebral oxygen saturation during cardiopulmonary bypass using near-infrared spectroscopy: the relationships with body temperature and perfusion rate

During cardiopulmonary bypass (CPB) because of weak arterial pulsation, near-IR spectroscopy (NIRS) is almost the only available method to monitor cerebral oxygenation noninvasively. Our group develops a NIRS oximeter to monitor regional cerebral oxygenation especially its oxygen saturation (rScO2). To achieve optimal coupling between the sensor and human brain, the distances between the light source and the detectors on it are properly chosen. The oximeter is calibrated by blood gas analysis, and the results indicate that its algorithm is little influenced by either background absorption or overlying tissue. We used it to measure the rScO2 of 15 patients during CPB. It is shown that rScO2 is negatively correlated with body temperature and positively with perfusion rate. There are two critical stages during CPB when rScO2 might be relatively low: one is the low-perfusion-rate stage, the other is the early rewarming stage. During cooling, the changes of total hemoglobin concentration (C(tHb)) compared with its original value is also monitored. It is shown that C(tHb) decreases to a small extent, which may mainly reflect cerebral vasoconstriction induced by cooling. All these results indicate that NIRS can be used to monitor cerebral oxygenation to protect cerebral tissue during CPB.

Non-invasive monitoring of human cerebral oxygen saturation by near infrared spectroscopy: instrumentation, calibration and application in cardiopulmonary bypass

Cerebral oxygenation especially regional cerebral oxygen saturation (rSO2) needs to be monitored to avoid irreversible damages induced by hypoxia. Based on steady-state spatially resolved spectroscopy (SRS), rSO2 was obtained by the NIRS oximeter developed by our group. The oximeter was calibrated by blood gas analyzer using the liquid tissue model made up by our group. The results indicate that the correlation between rSO2 and the real oxygen saturation of the model is excellent (R>0.99), and the influences by the background absorption and overlaying tissues are little. The rSO2 of 14 patients were monitored during cardiopulmonary bypass (CPB) non-invasively by our oximeter. The results indicate that first, NIRS is almost the only available method to monitor rSO2 non-invasively during CPB; second, rSO2 is negatively correlated with body temperature and positively with blood perfusion rate, so it reflects the balance between cerebral oxygen supply and consumption; third, the low perfusion rate stage and the rewarming stage are relatively critical because rSO2 decreases significantly and may be relatively low

NIRS study of cerebral oxygenation and hemodynamics in neonate at birth

To study the changes of cerebral oxygenation and hemodynamics in normal neonates at 2–5 min post-birth and understand the effects of pregnancy-induced hypertension (PIH) upon cerebral oxygenation and hemodynamics in newborn neonates. The near infrared spectroscopy (NIRS) was employed to measure the absolute quantity of brain tissue oxygen saturation (rSO2) in newborn neonates and the changes of concentrations of deoxyhemoglobin (Hb) and oxygenation hemoglobin (HbO2) with time relative to initial values to further obtain the changes of total hemoglobin (tHb) and cerebral perfusion (denoted by HbD). In normal neonates at 2–5 min post-birth, rSO2 increased while tHb remained relatively stable and HbD increased. In neonates born of PIH mothers at 3–5 min post-birth, the changes of tHb were markedly higher than those in the normal infants, p<0.05; at 2–5 min post-birth, the changes were markedly lower than the normal term infants. We concluded that NIRS can detect the changes of cerebral oxygenation and blood flow in a non-invasive and effective way.

Physiological meaning of cerebral oxygen saturation for piglet with hypoxia-ischemia

The physiological meaning of cerebral oxygen saturation absolute values and the oxygen metabolism of piglet with hypoxia-ischemia (HIE) were researched. The subjects were two piglets. During the total experiment of hypoxia then recovery, the regional cerebral tissue oxygen (rScO2), pulse oxygen saturation (SpO2) were detected non-invasively and the jugular oxygen saturation (SjO2), arterial oxygen saturation (SaO2) were given invasively. The results show that because SjO2 was equal to or larger than rScO2 and SaO2 > ScO2, rScO2 cannot be determined by the weighted sum of SjO2 and SaO2 which had been presented in some papers. According to above-mentioned analysis, the ecchymoma and pathological changes of the vessels due to HIE may be another contribution of rScO2. SjO2 was correlated with SaO2 (R=0.996 and 0.962 for two piglets) and the values of (SaO2-SjO2) are close to constants (29.3±8% and 30.3±8%).It means that because the subjects were under anesthesia, the oxygen consumption of cerebral tissue kept constants.

Non-invasive detecting cerebral hemoglobin concentration of newborns during oxygen inhalation using near-infrared spectroscopy

Human tissue oxygenation status can be completely described by the absolute values of the tissue hemoglobin concentrations. To detect them non-invasively and conveniently using continuous-wave near infrared spectroscopy (NIRS), the “in vivo calibration” algorithm was adopted. Three newborns were discussed during oxygen inhalation. By comparing the absorption coefficient before and after the change of tissue oxygenation induced by oxygen inhalation, the cerebral reduced scattering coefficient was calculated when the correlation coefficient R was above 0.75. Then the cerebral oxy- and deoxy-hemoglobin concentrations were derived based on the solution of steady-state diffusion equation. The total hemoglobin concentration of brain tissue was 50.1±6.1µM, which was consistent with most of the recent reports. In summary, even though further discussions are needed, the “in vivo calibration” method is effective and relatively accurate to detect human tissue hemoglobin concentrations.

Influence of high-flow modified ultrafiltration on brain oxygenation and perfusion during surgery for children with ventricular septal defects: a pilot study

Background: Modified ultrafiltration (MUF) can be performed in infants with ventricular septal defects (VSDs) after cardiopulmonary bypass (CPB) to reduce haemodilution and its potential adverse effects. High-flow MUF might reduce ultrafiltration duration and hasten the necessary correction of haemodilution during CPB. However, its influence on brain oxygenation remains controversial. Objective: This non-randomized, prospective, pilot study aimed to investigate the influence of high-flow MUF on brain oxygenation in infants with VSDs. Methods: High-flow MUF (≥20 mL/kg/min) was performed in twenty infants. Brain oxygen saturation (rSO2) and tissue haemoglobin index (tHI) were non-invasively and continuously measured intraoperatively using near-infrared spectroscopy (NIRS). Transcranial Doppler non-invasively detected the mean flow velocity of the middle cerebral artery (Vmean). Results: rSO2 increased significantly during MUF, as did tHI, Vmean, mean arterial pressure and haematocrit (all p<0.05). No correlation was found between changes in rSO2 and changes in other parameters (all p≥0.05). Conclusion: In infants with ventricular septal defects managed with CPB during VSDs repair, high-flow MUF did not reduce brain oxygenation.

Non-invasive detecting human tissue hemoglobin concentrations using near-infrared spectroscopy with in vivo calibration

Human tissue oxygenation status can be completely indicated by its hemoglobin concentrations. To detect them non-invasively using continuous-wave near infrared spectroscopy (NIRS), the ‘in vivo calibration’ algorithm, which was comparing tissue absorption before and after the change of tissue oxygenation, was adopted to obtain its reduced scattering coefficient. The hemoglobin concentrations of forearm muscle of four adults during occlusion, and those of brain cortex of three newborns during oxygen inhalation were detected. For the four adults, the of the forearm muscle were 6.01~7.87 cm–1 under 758 nm, and its total hemoglobin concentration (CtHb) was 37.5 ± 3.1 μM at the beginning of occlusion. For the three newborns, the cerebral were 5.46~8.42 cm–1 under 758 nm, and CtHb was 50.1 ± 6.1 μM at the beginning of oxygen inhalation. All the results were consistent with most recent reports. In summary, the in vivo calibration algorithm is effective to detect human tissue hemoglobin concentrations.

Investigation on biomedical engineering education of the universities in US and UK

To enhance the student training and scientific research of biomedical engineering (BME) department of Tsinghua University, we investigated the top three universities on BME in US, which are John Hopkins University (JHU), Georgia Tech and Emory University (GTEU) and the Harvard & MIT Division of Health Science and Technology (HST). We also investigated the department of medical physics and bioengineering of University College London (UCL) in UK. According to the investigations, many useful experiences are obtained. In this paper, the following four aspects are specified: first, the research fields being concerned in biomedical engineering in the above universities; second, the characteristics of the training programs to the undergraduate and graduate students of BME in these universities; third, the experiences of these universities; fourth, the suggestions to us