| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
|
Interact CardioVasc Thorac Surg 2007;6:265-269. doi:10.1510/icvts.2006.149658 © 2007 European Association of Cardio-Thoracic Surgery
Correlates of thenar near-infrared spectroscopy-derived tissue O2 saturation after cardiac surgeryDepartment of Intensive Care, Institute for Cardiovascular Research, VU University Medical Center, De Boelelaan 1117, 1081 HV Amsterdam, The Netherlands Received 4 December 2006; received in revised form 17 January 2007; accepted 22 January 2007
*Corresponding author. Tel.: +31-20-4444178; fax: +31-20-4442392.
We studied the significance of near-infrared spectroscopy (NIRS), for measuring tissue oxygenation (StO2) and perfusion adequacy, and thus for haemodynamic monitoring of patients after cardiac surgery. We compared NIRS-derived StO2 of the thenar muscle to haemodynamic variables, oxygenation indices, temperature, lactate levels and urinary output, in 23 patients in the course of time after cardiac surgery and admission into the intensive care unit. Clinical variables, global haemodynamics and NIRS% total haemoglobin (%HT) and StO2 in the thenar for up to 18–22 h after admission were measured. The StO2 declined concomitantly with a rise in the body-finger temperature difference. Cardiac output did not change but mean arterial pressure rose, concomitantly with tapering doses of nitroglycerine, indicative of an increase in vascular tone during recovery from surgery. From all variables, changes in body-finger temperature difference best correlated to changes in StO2 (rs=–0.48, P<0.001). As judged from clinical and haemodynamic correlates, thenar NIRS StO2 is a non-invasive measure of peripheral rather than global perfusion adequacy, after cardiac surgery. This may help to define the role of thenar NIRS monitoring after cardiac surgery in future studies.
Key Words: Tissue oxygenation; Vasoconstriction; Cardiac surgery; Tissue O2 saturation; Near-infrared spectroscopy
After cardiac surgery, patients are admitted to the intensive care unit (ICU) for monitoring of haemodynamics and tissue oxygenation, among others, often with the help of a pulmonary artery (PA) catheter, the use of which is increasingly criticised, even in cardiac (surgery) patients [1, 2]. Near-infrared spectroscopy (NIRS) is a non-invasive monitoring technique, by which, using light of different wavelengths, tissue haemoglobin oxygen saturation (StO2) as a measure of tissue oxygenation and perfusion adequacy can be assessed continuously [1, 3–8]. During cardiac surgery, the technique has primarily been used to assess cerebral oxygenation by monitoring StO2, variably related to jugular or central venous SO2 [9, 10]. During cardiac surgery, NIRS-derived hypothenar tissue pH and PO2 has been validated against invasive measurements [11]. In other conditions such as limb ischemia following vascular disease or compartment syndrome, but also during haemorrhagic shock and resuscitation and cardiac failure in patients, monitoring StO2 in splanchnic and peripheral tissues, such as liver, leg/arm muscles and hand (thenar) by NIRS has been observed to detect tissue hypoperfusion and hypooxygenation better and earlier than invasive and global perfusion and oxygenation variables [3–8]. Indeed, NIRS-derived StO2 may correlate to arterial or mixed venous SO2, reflect peripheral oxygen extraction, or may track oxygen delivery and cardiac output changes, as variably reported in the literature, depending on models, patients and sites of NIRS monitoring [3–5, 7, 8, 12, 13]. Our goal of the current study was to study the significance of StO2 monitoring by NIRS in post-cardiac surgery patients. We, therefore, compared the tissue oxygenation of the thenar muscle measured by NIRS to invasively measured haemodynamic parameters such as cardiac output, mixed venous haemoglobin saturation (SmvO2) and indicators of peripheral perfusion adequacy such as diuresis, body-finger temperature gradient and blood lactate levels [1].
According to the rules of the institutional committee on ethics, informed consent was waived because of the non-invasive nature of NIRS adjunctive to routine postoperative haemodynamic monitoring. The subjects included in this study were consecutively admitted into the ICU after having had cardiac surgery and cardiopulmonary bypass (CPB). The patients have a pulmonary artery catheter as part of normal monitoring. When measurements of both the thenar eminences were impossible, because of amputations, fractures or haematomas, the patient was excluded. Patients underwent cardiac surgery utilising CPB, as described previously [14]. In brief, anaesthesia was induced on the day of surgery by sufentanil 3 µg·kg–1 i.v., pancuronium 0.1 mg·kg–1 i.v. and midazolam 0.1 mg·kg–1 i.v. and maintained by a continuous infusion of propofol (5–15 ml·h–1, 20 mg·ml–1) with supplemental bolus doses of sufentanil. Radial artery and PA catheters were inserted. After tracheal intubation, the lungs were volume-controlled ventilated with a tidal volume of 8–10 ml·kg–1 resulting in an end-tidal CO2 concentration between 4 and 5%, using an O2-air mixture with an inspiratory O2 concentration of 40%. A positive end-expiratory pressure (PEEP) of 5 cm H2O was applied. Patients received 50–100 mg dexamethasone at induction. The CPB (Sockert-Sorin S3, Sorin Biomedica, Mirandola, Modena, Italy) was primed by 300 ml of Ringers lactate, 1000 ml of gelatin 4%, 100 ml of 20% mannitol, 50 ml of 8.4% sodium bicarbonate, 200 ml of aprotinin and 5000 IU of heparin. After systemic heparinization (300 IU·kg–1), extracorporeal blood flow was started, provided that the activated clotting time was more than 480 s. Non-pulsatile flow rate was maintained at 2–3 l·min–1·m–2, depending on the acid-base balance, pre-operative cardiac output and lactate concentrations. Cardiac surgery patients were cooled to 32 °C nasopharyngeal temperature. Mean arterial pressure (MAP) was maintained at 50–80 mmHg during CPB and if the MAP declined to less than 50 mmHg, the blood flow rate was increased and/or vasoactive drugs were given. After aortic cross-clamping, patients received crystalloid cardioplegia for myocardial protection (at 4 °C). Patients were weaned from the CPB, using inotropic support if necessary, and heparin was neutralized using an equivalent dose of protamine sulphate. Autologous blood and residual volume from the extracorporeal circuit were infused as first-choice fluid administration. Guided by low systemic and filling pressures, saline or colloids were infused additionally. If the haemoglobin concentration was less than 6 mmol·l–1, packed red blood cell concentrates were infused. Pre-operative use of aspirin, excessive bleeding and a long pump time prompted for administration of donor platelets. Mean arterial blood pressure (MABP) was obtained by electronic averaging (Tramscope®, Marquette/General Electric Medical Systems, WI, USA). The central venous pressure (CVP) was measured at end-expiration via the port of the PA catheter. Pressures were obtained with patients in supine position after zeroing to atmospheric level and at the level of the right atrium. Cardiac output was obtained by averaging triplicate measurements after injecting 10 ml room temperature glucose 5% boluses, regardless of the respiratory cycle. After indexing to body surface area calculated from height and weight, cardiac index (CI) was derived. Arterial and mixed venous blood was slowly aspirated from arterial and PA catheters, respectively, for measurement of haemoglobin/haematocrit (Sysmex SE-9000, Sysmex Corporation, Kobe, Japan), lactate levels and partial pressures of O2 and of O2 saturations in arterial (SaO2) and mixed venous (SmvO2) blood, corrected for body temperature (Rapidlab 865, Bayer Diagnostics, Tarrytown, NY, USA). Body temperature was measured with the help of the PA catheter and a probe served to obtain index finger skin temperature at the side of the NIRS probe, in order to calculate the body-finger temperature gradient as an index of peripheral perfusion adequacy [1]. For measuring tissue oxygenation, InSpectra's Tissue Spectrometer model 325 (Hutchinson Technology Inc., Arnhem, Netherlands) was used, which generates light of four different wavelengths between 680 and 800 nm [6]. Using a modified Lambert-Beer law, changes in oxy- and deoxyhaemoglobin concentrations were determined at a sample rate of 3.5 s. From relative concentrations, the tissue haemoglobin O2 saturation (StO2) can be calculated, which is the ratio of oxygenated to total haemoglobin, and approximately halfway between arterial and venous O2 saturation, as validated in vitro and in vivo [6]. Furthermore, the spectrometer also measures the tissue concentration of total haemoglobin expressed as %HT. Using a 25-mm probe, approximately 95% of the signal is recovered from tissues zero to 23 mm under the skin. The need for a 10-min warming up period after an 8-min calibration procedure was always acknowledged. Thereafter, the signal remains within 5%, in steady-state conditions [6]. After surgery, the patients were admitted to the ICU, and connected to the ventilator (Servoi, Maquette, Sweden) and volume-controlled ventilation was started with similar settings as during surgery. Demographics and co-morbidities were noted as well as the CPB and aortic clamping times. The NIRS probe was placed in a polyethylene cover (Optoshield®, Hutchinson Technology Inc.) and placed on intact skin over the thenar eminence, using extra bandages for good skin contact, of the right or left hand, whichever was best accessible. For each set of measurements, StO2 (normal about 87%) and %HT (normal about 0.20) [5, 6] were measured twice for 4 min, after a 1-min adjusting period. Results from both periods were averaged. Between the two periods, haemodynamic parameters, body and finger temperatures, O2 saturation by pulse oximetry (SpO2, Nellcore Optismart, Tyco Health care, Netherlands) and doses of vasoactive drugs administered were recorded. Cardiac output was measured and diuresis was recorded every hour. Blood was taken from arterial and PA catheters, in 2-ml heparinised syringes and placed on ice. The first set of measurements was obtained about 30 min after the patients arrival in the IC. The second was obtained 2 h after arrival, the third one 4 h after, and the fourth and last set was obtained the following morning, 18–22 h after admission, leaving the Optoshield® probe in place. Patients were taken care of by attending physicians not aware of the monitoring and deciding on doses of sedative and vasoactive drugs, transfusion of red blood cell concentrates, ventilatory settings and timing of extubation. The duration of the mechanical ventilation (starting from ICU admission onwards) was recorded in hours (until t=48 h). Non-parametric tests were used. The Spearman rank correlation was used for relations, and the Friedman analysis of variance for repeated measures to evaluate changes in time. Data are summarised as median and range. A P-value <0.05 was considered statistically significant.
A total of 23 patients were included in the study, and characteristics are given in Table 1, showing that removal of catheters or discharge from the ICU precluded measurements after ICU admission in a few patients. All patients survived to hospital discharge and only three patients stayed for more than one day in the IC, before discharge to specialised cardiac surgery wards.
Table 2 describes the postoperative course of global, peripheral and NIRS variables. It is shown that NIRS-derived %HT and StO2 declined as a function of time, concomitantly with a fall in diuresis and rise in body-finger temperature gradient despite a rise in body and finger temperature. The CI did not change but MABP rose, concomitantly with tapering doses of nitroglycerine and rises in CVP. SmvO2, PaO2 and SpO2 fell while SaO2 was unchanged. Blood Hb/Hct rose.
3.1. Correlations For pooled data, the SmvO2, NIRS-derived StO2 and CI interrelated (minimum rs=0.33, P=0.002), but not for changes in these variables in the course of time, except for a direct relation between changes in CI and SmvO2 (P=0.003). For pooled data, the NIRS-derived StO2 related to urinary production (rs=0.32, P=0.003) and tended to relate to body-finger temperature differences (rs=–0.21, P=0.057), while changes in time in body-finger temperature difference best correlated to changes in NIRS-derived StO2: rs=–0.48, P<0.001 (Fig. 1) and not to changes in %HT. Otherwise, StO2 highy correlated to NIRS-derived %HT, for pooled data and changes in time (rs=0.77, P<0.001).
The aim of this study was to evaluate the physiologic and clinical significance and relevance of thenar NIRS StO2 monitoring after cardiac surgery, by looking at clinical and haemodynamic correlates. The thenar was selected for monitoring by NIRS, since there is little overlying adipose tissue confounding measurements [6] and extremity perfusion is expected to be sensitive to decreases in cardiac output, resulting in peripheral vasoconstriction, during pump failure, cardiogenic or haemorrhagic/hypovolaemic shock and resuscitation [3–5, 7, 8]. The fall in NIRS-derived StO2 and %HT as function of time can be explained by a fall in well oxygenated blood volume in the microvasculature of the thenar eminence or, less likely, by a shift in the contribution to overall StO2 from oxygenated arterial to deoxygenated venous blood, in the course after cardiac surgery. Indeed, the body-finger temperature gradient increased in spite of a rise in body and finger temperature. Increases in the gradient and their relation to falls in NIRS-derived %HT and StO2, in spite of rising systemic Hb/Hct and unchanged SaO2, are most likely caused by a rise in peripheral vascular tone [5]. The latter may also have been responsible for the slight but significant fall in SpO2 in spite of unchanged SaO2. Indeed, NIRS StO2 in the thenar best (and inversely) related to the body-finger temperature difference (Fig. 1), from all variables studied. That changes in the temperature gradient and StO2, better correlated with each other than absolute values, can partly be attributed to interindividual differences in the contribution of non-muscular tissue relatively poor in blood such as adipose tissue to NIRS of the thenar [6]. Since ambient temperature was held constant in our ICU, we may surmise that the rise in global vascular tone rather than external cold was responsible for the rise in peripheral vascular tone. Indeed, the rise in blood pressure with declining nitroglycerine dosages at an unchanged cardiac output, implied a rise in global vascular tone in our patients until the first day following cardiac surgery. This may have contributed to a fall in SmvO2, by increasing O2 extraction. Otherwise, the SmvO2 correlated less to NIRS-derived StO2 than the body-finger temperature difference, thereby arguing again in favor of StO2 as a peripheral rather than a global measure of perfusion adequacy. This contrasts to findings in critically ill children, showing a fair relation between mixed venous and (cerebral or liver) StO2 [12, 13]. Since Hb increased and SaO2 and cardiac output remained unchanged, peripheral O2 delivery must have increased. The fall in SmvO2 may thus have been caused by increased O2 demands by the body, concomitantly with increased body temperature and recovery from anaesthesia and surgery. We cannot judge the cause of diminished urinary production in our patients, unless the initial diuresis was caused by the mannitol priming during CPB. A limitation of this study is the small sample of patients and absence of direct measurements on thenar perfusion to compare with the NIRS data. Moreover, we cannot judge the place of NIRS monitoring after cardiac surgery and particularly, whether this technique could replace or supplement the PA catheter, even though the latter is more likely since cardiac output and StO2 had a dissimilar course in our patients. By reflecting regional rather than global perfusion adequacy, thenar NIRS monitoring could be less helpful than monitoring of cardiac output in detecting postoperative cardiac dysfunction and hypoperfusion. In any case, we are not aware of any other studies of thenar NIRS monitoring after cardiac surgery. Future research should direct at the adjunctive and prognostic value of continuous StO2 monitoring, since peripheral vasoconstriction may precede a fall in arterial blood pressure during a fall in cardiac output and NIRS is a clinically applicable, non-invasive technique, yielding more direct information on peripheral perfusion adequacy than the (albeit also non-invasive) temperature gradient. Conversely, the body-toe temperature gradient may inversely but poorly correlate with cardiac output after cardiac surgery [1, 15], but not in our study, again indicating that peripheral perfusion depends on cardiac output as well as peripheral vascular tone. In conclusion, oxygenation of the thenar of patients after cardiac surgery measured with NIRS may decline with time as a consequence of a rise in peripheral vascular tone, as judged from clinical and haemodynamic correlates. Hence, NIRS-derived tissue oximetry is an index of peripheral rather than global perfusion adequacy, after cardiac surgery. Future research is needed to further define the role of thenar NIRS monitoring.
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| ANN THORAC SURG | ASIAN CARDIOVASC THORAC ANN | EUR J CARDIOTHORAC SURG |
| J THORAC CARDIOVASC SURG | ICVTS | ALL CTSNet JOURNALS |
Top
Abstract
1. Introduction