6/17/16. Near-Infrared Spectroscopy (NIRS): Principles, Evidence and Clinical Applications. What is Near Infrared Spectroscopy (NIRS)?

Transcription

1 What is Near Infrared Spectroscopy (NIRS)? Near-Infrared Spectroscopy (NIRS): Principles, Evidence and Clinical Applications Krisa Van Meurs, MD Rosemarie Hess Professor of Neonatal and Developmental Medicine Stanford University School of Medicine Medical Director, Neuro NICU Lucile Packard Children s Hospital Stanford X Simpósio Internacional de Neonatologia do Rio de Janeiro Hotel Royal Tulip, Rio de Janeiro 23 de Junho 216 NIRS can be used as a non-invasive monitoring technique for cerebral and somatic oxygenation and hemodynamics. Data is acquired from vascular beds (cerebral, renal, and splanchnic) with varied flows and extraction ratios. While pulse oximetry provides a measure of arterial oxygen saturation reflecting oxygen supply to the tissues, NIRS-measured regional oximetry measures the balance between local oxygen delivery and consumption beneath the sensor. It provides a non-invasive measure of end-organ oxygenation and perfusion. NIRS Principles Biologic tissues absorb light in the near infrared spectrum (7-9 nm). This is called the window into living organisms Absorption of light in the infrared spectrum is mainly by oxygenated and deoxygenated hemoglobin. The change in oxyhb and deoxyhb concentration can be calculated by measuring the change in absorption at 2 or more wavelengths. How a NIRS sensor works Placement of NIRS sensor on Forehead. The two black circles are the light source and detector. Light passes from light source through the scalp, skull, and brain tissue then to the detector. Cerebral saturation (rso2) reflects a ratio of arterial to venous blood of 25%:75% Cerebral Somatic Cerebral and Somatic oximetry high-flow and high extraction compensatory mechanisms and autoregulation cerebral desaturation is a late indicator of shock if autoregulationis present variable-flow, lower extraction flow is highly influenced by sympathetic tone somatic desaturation is early indicator of shock Two-site NIRS can provide ongoing indications of oxygenation and perfusion changes in cerebral and somatic circulations Interrogation depth of the sensor is estimated to be ½ of light source to detector separation distance. 1

2 Regional saturation reflects oxygen balance rso2 increases with more oxygen delivery or less demand while rso2 decreases when delivery falls or rise in demand Oxygen delivery is influenced by: Hemoglobin concentration Hemoglobin saturation Cardiac output (HR, preload, contractility and afterload) Oxygen demand fever, shivering, cold stress, infection, seizures, pain Oxygen demand hypothermia, sedation/paralysis, decreased extraction Cerebral NIRS measures Regional mixed cerebral oxygen saturation (rsco2) HbO! rsco! = HbO! + HHb HHb = deoxy hemoglobin HbO 2 = oxy hemoglobin Cerebral fractional tissue oxygen extraction (FTOE) FTOE = SaO! rsco! SaO! SaO 2 = arterial oxygen saturation Validation of cerebral oximetry measures Normal NIRS values in newborns Objective: To validate cerebral oximetry b 1 measurements with cerebral oxygen 9 8 saturation directly from blood drawn from 7 cephalad catheter in internal jugular vein 6 5 in neonates on veno-venous ECMO Results: There is a high level of agreement between cerebral oximetry and co-oximetry measured by jugular venous saturation. Cerebral Oximetry SctO 2 % y = R =.73, p < IL Subjects n = 136 samples Bias =.4 Precision = Co-ox ref SctO 2 % (.3*SpO 2 +.7*SjvO 2 ) rso2 Term Preterm Cerebral (%) Renal (%) Mesenteric (%) Values differ by sensor type with neonatal sensors reading 1% higher Alderliesten T, et al. Ped Res (216) Rais-BahramiK et a l., J Perinatol(26) a 9 Cerebral saturation varies with gestational age and chronologic age Normal cerebral saturation during transition rsco 2 (%) White weeks Light gray weeks Dark gray weeks Black 3-31 weeks *Used small adult sensor with INVOS 51 or Postnatal age (h) Average rsco2 was 65% at admission, increased with GA by 1% per week and also with chronologic age peaking at around 36 hours of age. Alderliesten T, et al. J Pediatr (216) Pichler G, et al J Pediatr (213) 2

3 Hemispheric differences in cerebral oximetry What rsco2 values injure the brain? Figure 2. (a) Representative pattern of 2 h of left (blue line) and of right Purpose: To determine if cerebral oximetry is symmetrical Results: During stable and normal arterial saturations, there were only minor differences in rsco2 values. During periods of unstable saturation, <85% and >98%, transient differences in rsco2 values of ~ 1% were seen between R and L. Mitochondrial damage in CA1 region of hippocampus in newborn piglets subjected to graded anoxia. Hou X et al., Physiol Meas 27 rsco2 <4% New or worse ischemia on MRI in infants with hypoplastic left heart syndrome (HLHS). Dent C et al., J Thorac Cardiovasc Surg 22 rsco2 <45% for >18 minutes Functional impairments in newborn piglets. Kurth, Ped Res 27 rsco2 ranging 33-44% Abnormal high energy phosphates measured by MRI spectroscopy in brains of newborn piglets. (Kusaka T, Ped Res 29) rsco2 <4% Lemmers P, van Bel F, Ped Res (29) Target rsco2 ranges for newborns What can you do if the rsco2 is abnormal? rsco2 (%) Safe zone Danger zone If cerebral saturation is too low: Hypocarbia (decrease ventilation) Hypotension (treat with fluid or inotropes) Anemia (give packed red blood cell transfusion) Low arterial saturation (increase FiO2) If cerebral saturation is too high: Supranormal arterial saturation (wean FiO2) Hypercarbia (increase ventilation) Dix et a l., Pediatr Res (214); Alderliesten et a l., Pediatr Res 215 Two-site NIRS monitoring Mesenteric or Splanchnic saturation monitoring Cerebral Renal/Flank Cerebral sensor can be placed on right or left side of forehead Renal sensor on posterior flank below costal margin and above iliac crest (T1-L2) 3

4 Articles 6/17/16 Dix et al. Table 4. NIRS devices and sensors NIRS devices Sensors INVOS oximetera Small adult SomaSensor (SAFB-SM) (standard) Pediatric SomaSensor (SPFB) OxyAlert Neonatal Sensor (CNN) Desaturation events may defy logic, NIRS Devices they don t have to defy detection. Fore-Sight oximeterb Neonatal Sensor (small sensor) Equanox model 76c Adult sensor (Classic Sensor 8 CA) stable clinical episode, without interference due to, for ing or care. The INVOS (Covidien) adult sensor wa ence measurement. However, in the course of this rese limitation occurred in comparing the Equanox senso (Covidien) adult sensor. Strong interference between resulted in unreliable results. We therefore adjusted t and compared the Equanox sensor with the Fore-Sigh sor, where the interference problem did not occur. The binations were as follows: INVOS (Covidien) adult sensor (SomaSensor SA INVOS (Covidien) neonatal sensor (Oxyalert C INVOS (Covidien) adult sensor (SomaSensor SA INVOS (Covidien) Somanetics pediatric sensor SPFB) INVOS (Covidien) adult sensor ((SomaSensor S Fore-Sight neonatal sensor (small sensor) Equanox sensor (Classic Sensor 8CA) vs. Fo tal sensor (small sensor) NIRS, near-infrared spectroscopy. NIRS Sensors INVOS 51C (Covidien). bfore-sight (CAS Medical Systems). cequanox model 76 (NONIN Medical). a a Experience the enhanced performance and ease of Fore-sight, Casmed use with FORE-SIGHT ELITE b c d e to optimize brain protection r First and only FDA cleared tissue oximeter utilizing a 5th wavelength of near-infrared light to reduce patient variability r Enhanced FORE-SIGHT algorithm improves accuracy to unprecedented levels (3.5% A rms)1 r Eliminates the need for pre-induction baseline reading r Detects otherwise unnoticed cerebral desaturation events INVOS 51C, Somanetics/Medtronic r Wide range of connectivity options with VGA output, Philips IntelliBridge, and EMR Systems 4. The different sensors used for comparison. (a) INVOS adult a. Figure INVOS Adult NIRS sensor, Covidien, 2 wavelengths sensor (SomaSensor SAFB-SM) (Covidien). (b) INVOS neonatal sensor Monitor Specifications Absolute Tissue Oxygen Saturation Range: to 99% Preamp Cable Large Sensor P/N: P/N: (2 sensors/case) Adult Cerebral: 4kg Product is Weight: 6. kg (13.3 lbs) SenSmart x-1, Nonin Marked ISO Certificate Number FM The INVOS 51C (Covidien) sensors use light-emitting diodes to emit near-infrared light of two wavelengths (73 and 81 nm). The nature and quantity of the recaptured near-infrared light reflects the amount of HHb and O2Hb, used to calculate rsco2. Two detectors are located next to the light-emitting diodes. By subtracting the shallow (shorter) signal from the deeper (further) signal, surface interference contamination is minimized (31,32). Clinical applicability of the INVOS device in neonates has been researched (33). The INVOS 51C (Covidien) can be used with three different sensors: the adult (SomaSensor SAFB-SM), pediatric (SomaSensor SPFB), and neonatal sensor (Oxyalert CNN). As the adult sensor has been exclusively used at the Wilhelmina Children s Hospital in the clinical setting, this sensor serves as reference measurement. The Equanox model 76 uses two light-emitting diodes (Classic Sensor 8CA), sending out a near-infrared signal composed of three wavelengths (73, 81, and 88 nm). The two light-emitting diodes are in the middle of the sensor, flanked by two photo diodes to capture the reflected light. Double detectors reduce intervening tissue and surface effects. At the time of the study, no neonatal sensor for the Equanox device was available. The Equanox device has thus far only one (adult) sensor. The Fore-Sight tissue oximeter and its neonatal sensor (small sensor) use four different wavelengths in the near-infrared light spectrum (from nm). One light emitting source is placed next to an absorbing diode. An overview of the different devices and their sensors is shown in Table East Industrial Road, Branford, CT 645 USA CASMED reserves the right to make changes to this sheet and the product at any time without notice. All rights reserved. CASMED, FORE-SIGHT and FORE-SIGHT ELITE are registered trademarks of CAS Medical Systems, Inc. All other trademarks belong to the companies indicated. U.S. Patents information at 1) MacLeod D, Ikeda K, Cheng C, Shaw A. Validation of the next generation FORE-SIGHT ELITE Tissue Oximeter for adult cerebral tissue oxygen saturation. Anesth Analg 213;116(SCA Suppl)-182,#4 from Rev 4 Sensor application procedure 1. Recommend place NIRS sensor onto Mepitel or other translucent skin dressing positioned. Do not apply pressure (e.g. headbands, wrap, tape). 2. Make sure signal strength bar is green 3. Check for erythema or irritation of skin around the sensor at least every 24 hours. Avoid lifting up sensor unless removing. 4. Sensor instructions state leave in place for 48 hours however we keep in place for 4-7 days. 5. Use adhesive remover or warm moist cloth to remove. Hypocarbia during mechanical ventilation Artificial Ventilation can influence cerebral Oxygenation 26 4/7 weeks 925 g, chorioamnionitis, day 1 of life 26 4/7 wks;gestation, 925 g; SMIV; chorioamnionitis; day 1; 1 9 SaO2 Preterm infants < 29 weeks gestation Infants with suspected hemodynamically significant PDA Study Design Hypoxic ischemic encephalopathy Five different NIRS sensors from the three NIRS devices were compared (Figure 4). Two sensors at the time were applied to the fronto- parietal part of the head of the neonate, one on each side symmetri Grade III/IV intraventricular hemorrhage cally. Sensors were fixated with an opaque elastic bandage to shield the optodes from ambient light. After a period of at least 1 h, sensors were switchedcongenital to the contralateralheart side to collect two periods of 6 min of a Complex disease 562 Pediatric diaphragmatic RESEARCH Volume 74 Number 5 November 213 Congenital hernia Copyright 213 International Pediatric Re Critically ill infants with hemodynamic instability (pre-ecmo or ECMO) HFOV can affect cerebral saturation Artificial Ventilation (HFO) can influence Systemic Hemodynamics 2626week gestation, 78 grams, on HFOV for RDS 1/7 wks; 78 g; day 1; HFO for RDS SaO2 MAP (18 16 cm H2O) REFERENCES 1. Wolf M, Greisen G. Advances in near-infrared spectros brain of the preterm and term neonate. Clin Perinatol 2 2. van Bel F, Lemmers P, Naulaers G. Monitoring neona bral oxygen saturation in clinical practice: value and pitf 28;94: Naulaers G, Meyns B, Miserez M, et al. Use of tissue oxyg fractional tissue oxygen extraction as non-invasive param oxygenation. A validation study in piglets. Neonatology 4. Nagdyman N, Fleck T, Schubert S, et al. Comparison bet sue oxygenation index measured by near-infrared spectro jugular bulb saturation in children. Intensive Care Med 2 5. Weiss M, Dullenkopf A, Kolarova A, Schulz G, Frey Near-infrared spectroscopic cerebral oxygenation rea and infants is associated with central venous oxygen sa Anaesth 25; Yoshitani K, Kawaguchi M, Tatsumi K, Kitaguchi K, Furu son of the INVOS 41 and the NIRO 3 near-infrared eters. Anesth Analg 22; Greisen G. Is near-infrared spectroscopy living up to its Fetal Neonatal Med 26;11: Petrova A, Mehta R. Near-infrared spectroscopy in regional tissue oxygenation during hypoxic events in undergoing critical care. Pediatr Crit Care Med 26;7:4 9. Toet MC, Lemmers PM. Brain monitoring in neonates 29;85: Who may benefit from NIRS monitoring per minute, because the different NIRS devices use diff rates and to exclude the influence of -values (artifa cessive values of each sensor (and if representative of b analyzed with the Signal Base program. No signals w given less weight during the calculations. We used SP Chicago, IL) for statistical analysis Statistical Analysis Data were summarized as mean values ± SD or as CNN)neonatal (Covidien). (c)sensor INVOS pediatric sensor (SomaSensor SPFB) b. (Oxyalert INVOS, Covidien and ranges where appropriate. Simple linear regressi Fore-Sight neonatal sensor (small sensor) (CAS Medical c. (Covidien). INVOS(d)pediatric sensor, Covidien used to analyze the correlation between the different Systems). (e) Equanox sensor (Classic Sensor 8CA) (NONIN Medical). signals. Bland Altman statistics compares the diffe d. Fore-sight neonatal sensor, CASMed, 5 wavelengths the signals with the average rsco2 (35). Representativ used NIRS devices in the NICU the INVOS 51C (Covidien), the were converted in median values with a sampling ra e. Equanox Equanox NONIN, wavelengths model sensor 76, and, the Fore-Sight4systems were compared. Dimensions: 29.7 cm (11.7 in) X 32.5 cm (12.8 in) X 17. cm (6.7 in) Call our Customer Service toll-free number to contact your local sales representative. To correct for the 7% difference between left and we measured both the devices bilaterally resulting in periods (34). Of the 55 included neonates, 1 result monitoring period of an hour. In our experience, 5 min to produce a reliable signal after application. W not include the first 5 min into the analysis of the resu (a program especially designed at the Wilhelmina Chi for NIRS signal analysis) was used to convert and obtained rsco2 signals. 6 rsco2 rsco2 5 MABP pco2 (mmhg) MABP Van Bel et al, Neonatology 29 van Bel F, Brain monitoring conference (215) 6 minutes van Bel F, Brain monitoring conference (215) 4

5 Change in rsco2 and FTOE with red blood cell transfusion ScO2 increases and FTOE decreases following transfusion in preterm infants. A poor correlation of pre-transfusion hematocrit with rsco2 is seen suggesting that hematocrit alone is a poor predictor of tissue oxygenation. rsco2, perfusion and symptoms of anemia improve in infants with rsco2 <55% but not in infants with rsco2 55% rsco2 or FTOE may be better indicators of need for transfusion Sood B et a l., J Nea r InfraRed Spectrosc (214) Hypotension in preterm infants Objective: To co m p are n eu ro d evel o p m en tal (N D) o utco m e, m ean arterial BP, and rsco2 between neonates treated for low mean arterial pressure (MAP) and controls Results: Infants treated for low MAP spent more time with MAP < gestational age than controls (9 versus %, p<.1) but there were no differences in ND outcome or rsco2 rso2 <5% for >1% time was associated with lower ND outcome Conclusion: This suggests that rsco2 is a surrogate marker for cerebral blood flow and could be used in hypotension treatment protocols. Alderliesten T et a l., J Pediatr (214) NIRS and the PDA rsco2 and hemodynamically significant PDA A hemodynamically significant PDA is associated with increased pulmonary blood flow and decreased systemic blood flow due to ductal steal. This is associated with lower blood pressure as well as decreased brain and other organ perfusion. NIRS has been used to study cerebral and somatic effects of PDA as well as response to medical and surgical treatment. Variable hspda (n= 2) Controls (n=2) P value Mean blood pressure (mm Hg) <.5 Mean rsco2 (%) <.5 FTOE <.5 Conclusions: In infants with PDA, mean blood pressure and cerebral saturation were lower and FTOE higher compared with control infant without a PDA. Lemmers P et a l., Pediatrics (28) Renal saturation and hemodynamically significant PDA Is the PDA significant in this infant? Variable hspda (n=21) No PDA (n=14) P value Renal saturation (%) Low renal saturation is associated with hs PDA. Also see increased variability of tracing. Renal saturation <66% was associated with hspda with sensitivity of 81% and specificity of 77%, AUC =.786, p<.1. Chock VY et a l., E-PAS

6 What about the PDA in this infant? NIRS and response to medical management of PDA Cerebral saturation (blue) and mean blood pressure in infant with PDA being treated with indomethacin. rsco2 is extremely low with PDA and decreases with each dose of indomethacin. Median rscos for babies with PDA. (white bars) and controls (black bars). rsco2 returned to normal values after treatment. Lemmers PMA et a l., Pediatrics (28) Renal saturation changes with medical management After Treatment: Renal sats higher with less variability Before Treatment: Renal sats significantly depressed at baseline with extreme variability PDA and cerebral autoregulation Cerebral autoregulation Brain maintains constant perfusion pressure despite fluctuations in systemic blood pressure Impaired cerebral autoregulation Pressure passive circulation Measured as concordance (r>.5) between mean arterial blood pressure (MAP) and rso 2 Common in preterm infants. Soul et al., Pediatr Res 27 Associated with mortality, severe IVH/PVL. Tsuji et al., Pediatrics 2 Cerebral autoregulation Cerebral autoregulation Intact Autoregulation MAP (mmhg) or rso2 (%) "!! *! )! (! '! &! %! $! #!!"# "! $%&'!! #! %! '! )! "!! "#! "%! "'! Time (min) MAP rso2 Lack of concordance between MAP and rso 2, r= Impaired Autoregulation MAP (mmhg) or rso2 (%) MAP Concordance between MAP and rso 2, r=.82 1 rso Time (min) Chock VY et a l., J Pediatr 212 6

7 PDA and loss of cerebral autoregulation Use of cerebral oximetry to predict outcome in HIE a PPI = Pressure Passivity Index N=28 VLBW *p=.4 Cerebral saturation (rsco2) is higher and fractional tissue oxygen extraction (FTOE) is lower by 24 hours and onward in neonates with HIE with adverse outcomes FTOE = SaO2 rsco2/sao2 Mean rsco 2 (%) b * * * * Time postnatal (h) Average PPI in infants following PDA ligation is increased for 2 hours and then normalizes by 6 hours. High rsco2 and low FTOE reflects secondary energy failure with reduced oxygen consumption by severely injured neuronal cells Solid line = good outcome Dashed line = poor outcome Mean cftoe * * * Time postnatal (h) Chock VY et al., J Pediatr 212 Lemmers P, et al. Ped Res (213) High rsco2 at 24 hours is associated with poor neurodevelopmental outcome NIRS Monitoring in Congenital Heart Disease What can pre-operative NIRS monitoring tell us? Indirect measure of Qp:Qs Regardless of systemic oxygenation (SpO2), cerebral or somatic oxygenation may be inadequate How do NIRS values correlate with other indicators of poor systemic perfusion (e.g. high lactate, prolonged capillary refill, cold extremities, low urine output) Effectiveness or need for additional nterventions Ventilator changes, diuretics, change in PGE dose, need for blood transfusion, earlier surgical intervention Infant with Juxtaductal Aortic Coarctation 3-day old 31 week infant with hypoplastic left heart SpO 2 stable- 95% pre-ductal and 71% post-ductal 1 9 Csat Rsat SpO 2 86% SpO 2 99% Cerebral sats dropped to 35-45% PGE restarted 1 Decreased UOP noted. Cr.6 Started Lasix and Milrinone 3:6: :224 6:2 78 8:45: : : :4 17: :22 2:4:28 212:25 22:2 23: : :73 65:41 7:239 81:2 99 1: :23 14: : :418 19:49:49 27: :21 1:28:49 26:41 3:447 42:24 6: 7

8 NIRS and Necrotizing Enterocolitis (NEC) Splanchnic oximetry is more challenging due to marked variability in readings (16%) when compared to renal (6%), cerebral (3%) and pulse oximetry (1-2%). Splanchnic oximetry may help identify babies with NEC. Low rsso2 is seen with NEC and feeding intolerance as shown here. Dotted line = normal, dashed = feeding intolerance, solid= NEC Can NIRS monitoring improve outcomes? Premature infants rsco2 targets to improve outcomes Congenital heart disease Management of HLHS Cortez J et a l., J Matern Fetal Neonatal Med (211) SafeBoosC II: Phase 2 Study (Randomized) Infants enrolled in: Lyon Madrid Copenhagen Cork Utrecht Graz Milan Cambridge (SafeboosC II Group, BMJ 215) <28 wks GA (n=166) NIRS- monitored Standard Treatment Avoid: Hyperoxia:rScO2 >85% Hypoxia: rsco2< 55% Sample Size: n=86/8 What could be done? When cerebral rsco2 is low (<55%), consider: Low pco2 (Increase pco2) hspda (Close) Hypotension (treat) Anemia (Erytrocyte transfusion) Low arterial saturation (Increase FiO2) When cerebral rsco2 is high (>85%), please consider: Supranormal Art Sat (Decrease FiO2 if possible) Too high pco2 (Decrease pco2) Low glucose (Treat low blood glucose) SafeBoosC II Infants enrolled in: Lyon Madrid Copenhagen Cork Utrecht Graz Milan Cambridge (SafeboosC II Group, BMJ 215) 81% hrs (No NIRS) N = 8 GA = 26.8 wks p<.1 Hyperoxia:rScO2 >85% Hypoxia: rsco2<55% (Yes NIRS) N = 86 GA = 26.6 wks 36% hrs (under preparation) Use of NIRS in pre-op newborns with HLHS Hypothesis: Pre-operative use of NIRS would reduce the need for invasive therapies including controlled ventilation and inspired gas manipulation Methods: Retrospective review of infants who had stage1 palliation for HLHS from 2-26 Historical cohort from 2-22 without pre-op NIRS monitoring (n=47) NIRS cohort from had cerebral and somatic NIRS measures recorded hourly (n=45) J Johnson BA et a l., An n Th o ra c Surg (29) 8

9 6/17/16 NIRS in HLHS - Results NIRS in HLHS - Conclusions Control (n=47) NIRS (n=45) P value Intubated n (%) 39 (83) 27 (6).14 Inspired nitrogen, n (%) 33 (7) 7 (16).1 Duration nitrogen use, hours Arterial saturation (%) Cerebral saturation (%) Somatic saturation (%) Routine use of pre-op NIRS-monitoring resulted in: Reduced mechanical ventilation (higher SaO 2, lower PaCO 2) Reduced use of inspired gases No impact on mortality or length of hospital stay Higher systemic saturation in NIRS monitored group was not associated with hypotension, acidosis, or worsened renal function. Johnson BA et al., Ann Thorac Surg (29) Conclusions Johnson BA et al., Ann Thorac Surg (29) Thank you! NIRS monitoring is a useful non-invasive measure of end organ oxygenation and perfusion providing critical information about the balance of oxygen delivery and consumption. While there are wide ranges on normal values in newborns depending on the device and sensor used, trend monitoring provides critical information in specific clinical scenarios. Further studies are needed to determine if NIRS can improve important patient outcomes in newborn infants. 9