Historically, the basic monitor for the cardiorespiratory systems has been the stethoscope, either chest wall or esophageal. This is less frequently used today since the introduction of arterial oxygen saturation and end-tidal CO₂ (EtCO₂) monitoring but may still be very useful to detect specific conditions and should be applied whenever feasible. A stethoscope is particularly useful as an immediate warning in cases when surgical manipulations may suddenly kink major airways, as in operation for tracheoesophageal fistula. A precordial stethoscope applied to the left chest may also provide early evidence of tracheal tube that migrates endobronchially while positioning the infant. It may also be useful to monitor heart sounds during surgery and might provide other information, e.g., a clue to successful or unsuccessful closure of a persistent ductus arteriosus during thoracoscopic surgery [1]. It is also quite indispensible should other monitors fail. The esophageal stethoscope is a relatively benign instrument, although minor esophageal injuries and complications have been reported [2]. It may cause airway obstruction in infants with vascular ring, even with an endotracheal tube in place. The use of an esophageal stethoscope might result in the esophagus being confused with the trachea during neck operations, and its use should always be communicated to the surgeon [3]. Esophageal stethoscopes often include thermistors to monitor temperature; positioning the stethoscope where the heart sounds are loudest ensures a retrocardiac position of the bulb and central temperature monitoring.

Pulse oximetry was introduced in the 1980s and rapidly became an indispensible aid to the management of the neonate in the perioperative period. The pulse oximeter probe consists of two light emitting diodes (LEDs) producing respectively red and infrared light with a semiconductor as a detector. The cyclical changes in the absorption of these two wavelengths during arterial pulsations are recorded, and their ratio is derived using an internal algorithm. Optimally the LEDs and the detector should be placed exactly opposite each other with 5–10 mm of intervening tissue. Low readings may occur if the components of the probe are not exactly aligned [4]. In the neonate, the probe is commonly placed across the palm of the hand or foot but also may be applied to the ear lobe, cheek, or tongue [5]. Pulse oximeters vary in their performance but most claim an accuracy of ±2–4 % or less with saturations above 80 %. At low saturations, pulse oximeters are much less accurate. Hence during profound desaturations, they cannot be relied upon. It is also important to realize that the top portion of the oxyhemoglobin desaturation curve is flat; hence at high saturations relatively large changes in arterial pO₂ occur with only minor changes in saturation. Thus, there has been continuing difficulty in defining a “safe” target saturation for the preterm infant in order to avoid the consequences of hyperoxia [6–8]. Different models of pulse oximeter vary in their response time to physiological changes in the patient [9]. In general, those more recent models with “signal extraction technology (SET)” have faster response times.

The pulse oximeter is extremely useful in neonates because desaturation occurs most frequently in this age group [10], but the oximeter is subject to interference from several factors. Motion artifact is usually not a problem during anesthesia but may be significant during the induction and recovery phases. When this may be a problem, the use of models with SET technology (e.g., Masimo) is preferred [9]. A strong external light source may affect the oximeter’s performance. The probe should be covered to exclude extraneous light and protected by a rigid frame to prevent pressure on the sensor. In low perfusion states the signals may not be adequate for interpretation, and no result will be displayed; again later model machines with SET may perform better under these conditions. The performance of the pulse oximeter is not affected by changes in the hematocrit, anemia, or bilirubinemia. However in bronze baby syndrome which may occur after phototherapy, SpO₂ readings become unreliable [11]. Dark skin pigmentation also may cause falsely low readings especially at lower levels of saturation (<80 %) [5]. False high readings may occur in the presence of carboxyhemoglobin. Significant levels of methemoglobin bias the reading toward 85 % [5].

In neonatal anesthesia practice it is often advantageous to place two separate pulse oximeter probes on the patient. In some cases it may be desired to monitor both pre-ductal (R arm or head) and post-ductal saturation, and in others the second probe may act simply as a back-up reference [12].

Complications have been reported from the application of oximeter sensors. When applying sensors circumferentially to the finger, caution must be exercised to avoid too tight an application, which might cause injury [13]. When a reusable clip-type sensor is used on the ear, care should be taken to ensure that the clip does not exert excessive pressure [14].

The basic noninvasive equipment to measure blood pressure is the blood pressure cuff. It is suggested that the width of the cuff used should be 0.44–0.55 × the midpoint circumference of the limb utilized; thus the optimal width in the full-term neonate is approximately 1 in. [15]. In the operating room, the most accurate determination of the systolic pressure can be obtained by placing a Doppler flow probe on an artery distal to the cuff [15]. Measurements taken with an automated oscillometric devices tend to overestimate systolic and mean blood pressures, especially when the neonate is hypotensive [16]. More recent evidence suggests that the greatest discrepancy in noninvasive blood pressures between upper and lower extremities occurred in the smallest infants (<1,000 g) [17]. These devices should not be relied upon in sick infants or those requiring extensive surgery. In healthy infants, blood pressure measurements taken with a cuff applied to the upper and lower extremities are normally similar.

Direct measurement of blood pressure using an intra-arterial line is often required in all critically ill neonates and those requiring major surgery. Arterial lines may be inserted at various sites, and each has advantages and potential disadvantages.

The umbilical artery is relatively easy to access in the immediate neonatal period and has been widely used in neonatal intensive care units. However, serious thromboembolic complications may follow and involve intra-abdominal organs, the lower limbs, and even the spinal cord [18]. Caution should be exercised in the choice of catheter material and design and the fluids administered. Silicon rubber catheters with an end hole are the preferred type of catheters. Hypertonic and alkaline solutions should be avoided. The use of heparin in the infusate may decrease the incidence of line occlusion but does not reduce the incidence of thromboembolism [19]. When managing an infant with an umbilical artery catheter, the anesthesiologist should exercise caution when withdrawing blood samples or flushing the line. Sampling and reinfusing rates should not exceed 1 ml in 30 s in the preterm infant. Rates in excess of this may significantly and adversely affect cerebral blood flow and oxygenation [20].

Radial artery lines are more commonly used for intraoperative monitoring by anesthesiologists. Various methods have been recommended to improve the success of transcutaneous insertion of a catheter in neonates. Smooth insertion of the catheter into small arteries is more likely to be attained if the bevel of the needle is rotated inferiorly once a flashback of blood is observed. In some cases the use of a fine guidewire may be useful to thread an obstructed catheter into the artery. If the artery cannot be palpated readily, a Doppler flowmeter probe or transillumination of the infant’s wrist may facilitate its location [21]. In cases of difficulty, it may be necessary to resort to a cutdown for cannulation of the vessel. Allen’s test of the adequacy of collateral flow to the hand is difficult to perform in small infants and is unreliable even in adults; hence it is not routinely performed in many centers. Once a radial artery access has been successfully established, the limb should be immobilized on a splint and a secure continuous flush system attached. Normal saline is preferred to glucose-containing solutions for all monitoring lines [22]. Flushing of radial arterial lines should be limited to small volumes and slow rates of injection as retrograde flow into the cerebral circulation may occur with as little as 0.5 ml flush solution [23]. Blood, which has been withdrawn while taking a laboratory sample, should be re-injected into a venous access site rather than an arterial access.

Serious complications with radial artery lines are relatively rare in neonates although instances of ischemic damage to the hand have been described [23–25]. Any evidence of impaired circulation or skin changes distal to the catheter is an indication for its immediate removal. There is no evidence that a cutdown approach to cannulation is accompanied by increased risk of complications [26].

Femoral artery cannulation may be used as an alternative if radial puncture is impossible, and this may better reflect true arterial pressure than does the radial artery in some instances [27]. The risk of infection at the puncture site is not increased with a femoral artery cannulation, although in small infants, the distal circulation should be carefully monitored as perfusion-related complications may occur [28]. Great care should be taken during insertion of femoral lines to avoid needle injury to the hip joint, septic arthritis, and damage to the head of the femur [29]. It is extremely important that the artery be accessed caudal to the level of the inguinal ligament; insertion above the ligament may cause a retroperitoneal hemorrhage. A Seldinger technique with careful aseptic technique is recommended for insertion of an arterial catheter, e.g., 3 Fr, 5 cm polyurethane catheter. The puncture site should be covered with a transparent sterile occlusive dressing and should be regularly inspected. If there is any evidence of impaired circulation in the limb, the catheter should be removed immediately. Sepsis is more likely to occur with femoral (or radial) arterial catheters that are left in place for more than 5 days [28].

The arterial wave form and the actual pressures obtained from arterial catheters in small infants may be affected to a degree by the compliance of the connecting tubing and by a continuous fluid flush. It is also necessary to consider the volume of fluid that is administered in order to continuously flush the tubing. The use of a pressurized bag with a controlled infusion device (e.g., Intraflo or Squeezeflow system) claims to deliver 3 ml/h of flush solution. However, under some circumstances, it may deliver larger volumes of solution, especially when the rapid flush activator is frequently used or malfunctions [30]. This may lead to fluid overload and possible coagulopathy secondary to excessive heparin administration. A cerebral vascular accident may occur when the arterial line is flushed with crystalloid solution continuously, particularly in the radial artery, resulting in an ischemic bolus of crystalloid in the carotid artery. A preferable method to continuously flush arterial lines is to use a syringe pump [30] set to deliver 1 ml/h.

Alternative sites to monitor arterial pressure invasively have been suggested. The axillary artery is an attractive alternative as it has a very good collateral circulation and it can be easily palpated. It has successfully been used without serious complication in critically ill neonates [31]. There are also reports of brachial artery cannulation without serious complication [25], although this must be viewed with some caution as this artery has a less well developed collateral circulation.

Central venous catheters (CVCs) may be inserted into the neonate to monitor central pressure and also to permit the infusion of inotropic and hyperosmolar solutions. Transcutaneous insertion of a CVC via the internal jugular vein in the neonate is greatly facilitated by the use of ultrasound [32]. The vein is readily recognized by the ease with which it collapses with slight pressure of the transducer on the neck. Slight pressure over the liver will increase the lumen of the vein and facilitate puncture. It is not usual to place small infants in Trendelenburg position because the tilt adds little to the venous diameter since they are so short. It is common to place a small roll under the shoulders. The left subclavian vein has also been suggested as a route, again preferably with the use of ultrasound to locate the vein accurately [33]. However, the subclavian vein has a smaller diameter in the neonate [34], and its cannulation is associated with a greater incidence of complications, especially pneumothorax. Selecting the correct depth of insertion may be difficult in the small infant. The tip of the catheter should not be inferior to the junction of the superior vena cava with the (R) atrium and this should be confirmed by radiology. If it is further advanced into the (R) atrium, serious complications may result, including arrhythmias, damage to the tricuspid valve, or even cardiac perforation with tamponade. Full aseptic precautions should be observed during central venous cannulation in neonates. When CVCs are to be used to deliver hyperalimentation solutions, extreme care with asepsis is essential as catheter-related sepsis and endocarditis is a common complication.

When it is impossible to access the superior vena cava, a reliable index of central venous pressure may be obtained by monitoring the pressure in the inferior vena cava (IVC) via the femoral vein. Low IVC pressure tends to be only very slightly greater than central venous pressure [35].

Capnography provides evidence of ventilation and indirect evidence of cardiac output, pulmonary blood flow, and metabolic state; as such it is an invaluable intraoperative monitor. The shape of the capnogram may be useful to validate readings by observation of an “alveolar plateau” on the tracing. Alterations in this shape may alert the anesthesiologist to developing physiological changes, or technical problems with the circuit and airway. EtCO₂ levels also provide an approximation of arterial pCO₂ levels, the accuracy of which is dependent upon many factors. Direct determination of arterial pCO₂ is essential when this level might be crucial. Accuracy in the control of arterial pCO₂ intraoperatively is most desirable as either hypercarbia or hypocarbia may have serious adverse physiological effects.