Measurements of Optical Pathlength Using Phase-Resolved Spectroscopy in Patients Undergoing Cardiopulmonary Bypass

Preterm infants variability in cerebral near-infrared spectroscopy measurements in the first 72-h after birth

BACKGROUND

Cerebral near-infrared spectroscopy is a non-invasive tool used to measure regional cerebral tissue oxygenation (rScO2) initially validated in adult and pediatric populations. Preterm neonates, vulnerable to neurologic injury, are attractive candidates for NIRS monitoring; however, normative data and the brain regions measured by the current technology have not yet been established for this population.

METHODS

This study's aim was to analyze continuous rScO2 readings within the first 6-72 h after birth in 60 neonates without intracerebral hemorrhage born at ≤1250 g and/or ≤30 weeks' gestational age (GA) to better understand the role of head circumference (HC) and brain regions measured.

RESULTS

Using a standardized brain MRI atlas, we determined that rScO2 in infants with smaller HCs likely measures the ventricular spaces. GA is linearly correlated, and HC is non-linearly correlated, with rScO2 readings. For HC, we infer that rScO2 is lower in infants with smaller HCs due to measuring the ventricular spaces, with values increasing in the smallest HCs as the deep cerebral structures are reached.

CONCLUSION

Clinicians should be aware that in preterm infants with small HCs, rScO2 displayed may reflect readings from the ventricular spaces and deep cerebral tissue.

IMPACT

Clinicians should be aware that in preterm infants with small head circumferences, cerebral near-infrared spectroscopy readings of rScO2 displayed may reflect readings from the ventricular spaces and deep cerebral tissue. This highlights the importance of rigorously re-validating technologies before extrapolating them to different populations. Standard rScO2 trajectories should only be established after determining whether the mathematical models used in NIRS equipment are appropriate in premature infants and the brain region(s) NIRS sensors captures in this population, including the influence of both gestational age and head circumference.

International survey of neuromonitoring and neurodevelopmental outcome in children and adults supported on extracorporeal membrane oxygenation in Europe

BACKGROUND

Adverse neurological events during extracorporeal membrane oxygenation (ECMO) are common and may be associated with devastating consequences. Close monitoring, early identification and prompt intervention can mitigate early and late neurological morbidity. Neuromonitoring and neurocognitive/neurodevelopmental follow-up are critically important to optimize outcomes in both adults and children.

OBJECTIVE

To assess current practice of neuromonitoring during ECMO and neurocognitive/neurodevelopmental follow-up after ECMO across Europe and to inform the development of neuromonitoring and follow-up guidelines.

METHODS

The EuroELSO Neurological Monitoring and Outcome Working Group conducted an electronic, web-based, multi-institutional, multinational survey in Europe.

RESULTS

Of the 211 European ECMO centres (including non-ELSO centres) identified and approached in 23 countries, 133 (63%) responded. Of these, 43% reported routine neuromonitoring during ECMO for all patients, 35% indicated selective use, and 22% practiced bedside clinical examination alone. The reported neuromonitoring modalities were NIRS (n = 88, 66.2%), electroencephalography (n = 52, 39.1%), transcranial Doppler (n = 38, 28.5%) and brain injury biomarkers (n = 33, 24.8%). Paediatric centres (67%) reported using cranial ultrasound, though the frequency of monitoring varied widely. Before hospital discharge following ECMO, 50 (37.6%) reported routine neurological assessment and 22 (16.5%) routinely performed neuroimaging with more paediatric centres offering neurological assessment (65%) as compared to adult centres (20%). Only 15 (11.2%) had a structured longitudinal follow-up pathway (defined followup at regular intervals), while 99 (74.4%) had no follow-up programme. The majority (n = 96, 72.2%) agreed that there should be a longitudinal structured follow-up for ECMO survivors.

CONCLUSIONS

This survey demonstrated significant variability in the use of different neuromonitoring modalities during and after ECMO. The perceived importance of neuromonitoring and follow-up was noted to be very high with agreement for a longitudinal structured follow-up programme, particularly in paediatric patients. Scientific society endorsed guidelines and minimum standards should be developed to inform local protocols.

Cerebral oximetry-guided pulmonary artery banding for end-stage heart failure in a child with left ventricular noncompaction cardiomyopathy: a case report

Pulmonary artery banding (PAB) may reduce the need for left ventricular assist devices and heart transplantation in children with end-stage heart failure. However, excessive banding may increase the right ventricular afterload, leading to worsening of heart failure. The estimated right ventricular pressure and the shifting of the interventricular septum by transesophageal echocardiography (TEE), pulmonary artery pressure, right atrial and ventricular pressure, percutaneous oxygen saturation, and mixed venous oxygen saturation are utilized to determine the optimal circumference for PAB. Here, we report the case of a 5-month-old patient with end-stage heart failure due to left ventricular noncompaction cardiomyopathy (LVNC), with a gene mutation of MYH7, who underwent successful PAB. The exact PAB placement was additionally guided by using cerebral regional oxygen saturation (rSO₂) measurement to achieve a tolerable and optimal PAB effect. We monitored rSO₂ and other hemodynamic parameters while surgeons banded the pulmonary artery to achieve both highest rSO₂ levels and stable hemodynamics. rSO₂ was 68% before banding, and increased and remained at over 90% after the banding at same FiO₂. Patient's heart failure improved gradually, and the child was discharged home at 6 months after PAB. The rSO₂ is a simple and non-invasive monitor for the measurement of oxygen delivery to the brain tissue. rSO₂ alone would not be able to guide PAB placement in the vulnerable DCM patients, but it may be of one further monitoring value for the optimal pulmonary artery circumference while patients are undergoing PAB.

Cerebral oxygen saturation (rSO2) during cardiopulmonary bypass (CPB) measured using the INVOS oximeter closely correlates with baseline rSO2

Baseline cerebral regional saturation (rSO₂) measured using the INVOS 5100C (Medtronic, MN, USA) varies widely among patients with cardiac and/or renal diseases. To identify significant correlates of baseline rSO₂ and to investigate intraoperative rSO₂ changes, we conducted a retrospective study in 494 patients undergoing on-pump cardiovascular surgery. Correlations between preoperative blood laboratory test variables and baseline rSO₂ before anesthesia were examined. Intraoperative rSO₂ changes were analyzed. Of all the variables examined, log-transformed B-type natriuretic peptide (BNP) most significantly and negatively correlated with baseline rSO₂ (r = - 0.652, p < 0.0001). Intraoperatively, rSO₂ showed the lowest value during cardiopulmonary bypass (CPB) (median rSO₂: 56.2% during CPB vs. 63.9% at baseline, p < 0.0001). Although rSO₂ during CPB correlated positively with hemoglobin concentration and oxygen delivery during CPB (r = 0.192, p < 0.0001; and r = 0.172, p = 0.0001, respectively), it correlated much more closely with baseline rSO₂ (r = - 0.589, p < 0.0001). Thus, patients showing low baseline rSO₂ primarily associated with preoperatively high BNP continued to show low rSO₂ even during CPB independent of hemodynamics artificially controlled by CPB. Our findings suggest that low baseline rSO₂ in patients with high BNP due to cardiac and/or renal diseases is more likely to result from tissue edema causing alterations in optical pathlength and thus in calculated rSO₂ values, not readily modifiable with CPB, rather than actual cerebral hemodynamic alterations readily modifiable with CPB. These may partly explain why the INVOS oximeter is a trend monitor requiring baseline measures.

Anatomical and physiological variables influencing measurement of regional cerebral oxygen saturation by near infrared spectroscopy using the Sensmart Model X-100TM

The Sensmart Model X-100 (Nonin Medical Inc, Plymouth, MN, USA) is a relatively new device that possesses two sets of emitters and detectors and uses near infrared spectroscopy (NIRS) to measure regional cerebral oxygen saturation (rSO₂). The value of rSO₂ obtained by other NIRS devices is affected by physiological and anatomical variables such as hemoglobin concentration, area of cerebrospinal fluid (CSF) layer and skull thickness. The effects of these variables have not yet been determined in measurement of rSO₂ by Sensmart Model X-100. We examined the effects of area of CSF, hemoglobin concentration, and skull thickness on the values of rSO₂ measured by Sensmart Model X-100 and tissue oxygen index (TOI) measured by NIRO-200NX (Hamamatsu Photonix, Hamamatsu, Japan). Forty neurosurgical, cardiac and vascular surgical patients who underwent preoperative computed tomographic (CT) scan of the brain were enrolled in this study. Regional cerebral oxygen saturation (rSO₂) at the forehead was measured sequentially by NIRO-200NX and by Sensmart Model X-100. Simultaneously, mean arterial pressure, hemoglobin concentration, and partial pressure of carbon dioxide in arterial blood (PaCO₂) were measured. To evaluate the effects of anatomical factors on rSO₂, we measured skull thickness and area of CSF layer using CT images of the brain. Multiple regression analysis was used to examine the relationships between the rSO₂ values and anatomical and physiological factors. The area of the CSF layer and hemoglobin concentration had significant associations with rSO₂ measured by the Sensmart Model X-100, whereas none of the studied variables was significantly associated with TOI. The measurement of rSO₂ by Sensmart Model X-100 is not affected by the skull thickness of patients. Area of the CSF layer and hemoglobin concentration may be the main biases in measurement of rSO₂ by Sensmart Model X-100.

Clinical and Technical Limitations of Cerebral and Somatic Near-Infrared Spectroscopy as an Oxygenation Monitor

Cerebral and somatic near-infrared spectroscopy monitors are commonly used to detect tissue oxygenation in various circumstances. This form of monitoring is based on tissue infrared absorption and can be influenced by several physiological and non-physiological factors that can induce error in the interpretation. This narrative review explores those clinical and technical limitations and proposes solutions and alternatives in order to avoid some of those pitfalls.

A Cross-Sectional Survey of Near-Infrared Spectroscopy Use in Pediatric Cardiac ICUs in the United Kingdom, Ireland, Italy, and Germany

OBJECTIVES

Despite the increasing use of near-infrared spectroscopy across pediatric cardiac ICUs, there is significant variability and equipoise with no universally accepted management algorithms. We aimed to explore the use of near-infrared spectroscopy in pediatric cardiac ICUs in the United Kingdom, Ireland, Italy, and Germany.

DESIGN

A cross-sectional multicenter, multinational electronic survey of one consultant in each pediatric cardiac ICU.

SETTING

Pediatric cardiac ICUs in the United Kingdom and Ireland (n = 13), Italy (n = 12), and Germany (n = 33).

INTERVENTIONS

Questionnaire targeted to establish use, targets, protocols/thresholds for intervention, and perceived usefulness of near-infrared spectroscopy monitoring.

RESULTS

Overall, 42 of 58 pediatric cardiac ICUs (72%) responded: United Kingdom and Ireland, 11 of 13 (84.6%); Italy, 12 of 12 (100%); and Germany, 19 of 33 (57%, included all major centers). Near-infrared spectroscopy usage varied with 35% (15/42) reporting that near-infrared spectroscopy was not used at all (7/42) or occasionally (8/42); near-infrared spectroscopy use was much less common in the United Kingdom (46%) when compared with 78% in Germany and all (100%) in Italy. Only four units had a near-infrared spectroscopy protocol, and 18 specifically used near-infrared spectroscopy in high-risk patients; 37 respondents believed that near-infrared spectroscopy added value to standard monitoring and 23 believed that it gave an earlier indication of deterioration, but only 19 would respond based on near-infrared spectroscopy data alone. Targets for absolute values and critical thresholds for intervention varied widely between units. The reasons cited for not or occasionally using near-infrared spectroscopy were expense (n = 6), limited evidence and uncertainty on how it guides management (n = 4), difficulty in interpretation, and unreliability of data (n = 3). Amongst the regular or occasional near-infrared spectroscopy users (n = 35), 28 (66%) agreed that a multicenter study is warranted to ascertain its use.

CONCLUSIONS

Although most responding units used near-infrared spectroscopy for high-risk patients, the majority (31/35 [88%]) did not have any protocols or guidelines for intervention. Target thresholds and intervention algorithms are needed to support the use of near-infrared spectroscopy in pediatric cardiac ICUs; an international multicenter study is warranted.

Continuous correction of differential path length factor in near-infrared spectroscopy

In continuous-wave near-infrared spectroscopy (CW-NIRS), changes in the concentration of oxyhemoglobin and deoxyhemoglobin can be calculated by solving a set of linear equations from the modified Beer-Lambert Law. Cross-talk error in the calculated hemodynamics can arise from inaccurate knowledge of the wavelength-dependent differential path length factor (DPF). We apply the extended Kalman filter (EKF) with a dynamical systems model to calculate relative concentration changes in oxy- and deoxyhemoglobin while simultaneously estimating relative changes in DPF. Results from simulated and experimental CW-NIRS data are compared with results from a weighted least squares (WLSQ) method. The EKF method was found to effectively correct for artificially introduced errors in DPF and to reduce the cross-talk error in simulation. With experimental CW-NIRS data, the hemodynamic estimates from EKF differ significantly from the WLSQ (p < 0.001). The cross-correlations among residuals at different wavelengths were found to be significantly reduced by the EKF method compared to WLSQ in three physiologically relevant spectral bands 0.04 to 0.15 Hz, 0.15 to 0.4 Hz and 0.4 to 2.0 Hz (p < 0.001). This observed reduction in residual cross-correlation is consistent with reduced cross-talk error in the hemodynamic estimates from the proposed EKF method.