Arterial oxygen saturation (SaO₂) and partial arterial pressure of carbon dioxide (PaCO₂) are two of the most important respiratory parameters in the evaluation of critically ill neonates. SaO₂ is commonly estimated by pulse oximetry (SpO₂), a non invasive technique widely used in neonatal intensive care units. Partial pressure of carbon dioxide (PCO₂) can be assessed by sampling blood (PCO_2BGA) or by noninvasive techniques such as transcutaneous monitoring (PtcCO₂), which reduces the need for frequent blood drawing and painful punctures and provides a continuous estimation of PCO₂.

As already described elsewhere [1], the TOSCA monitor combines a pulse oximeter of new technology (Masimo SET) with the Stow-Severinghaus electrode for PtcCO₂ measurement in a single sensor, which is also equipped with a heating element to increase the local perfusion Figure 1(a). The sensor is attached to the ear by a clip having an adhesive holder and a reflective element placed onto the inner surface.

Like other similar devices, the TOSCA pulse oximeter determines SaO₂ by measuring the difference in absorption of selected light wavelengths by oxyhemoglobin and deoxyhemoglobin. It utilizes two light-emitting diodes (LEDs): one LED emits red light (LED-R; 658 nm wavelength) selectively adsorbed by deoxyhemoglobin, while the other one emits infrared light (LED-IR; 880 nm wavelength) selectively absorbed by oxyhemoglobin. The sensor also contains a photodiode (photodetector), an electronic device which converts light into an electronic signal proportional to the incident light intensity and which sends it to the TOSCA monitor for calculation. Unlike other pulse oximeters, either the LEDs or the photodiode are positioned on the same surface, since the light emitted is reflected by the clip surface opposite to the sensor place. The TOSCA pulse oximeter is equipped with Masimo SET technology and uses a set of algorithms designed to improve SpO₂ monitoring during motion and low perfusion.

Transcutaneous measurement of CO₂ is based on the observation that this gas has a high tissue solubility and diffusion through the skin, and that the application of local heat dilates blood vessels and enhances skin permeability. The PCO₂ component of the TOSCA consists of a Stow-Severinghaus type electrode, a potentiometric sensor combining a silver/silver chloride reference electrode, and a glass pH electrode. Over the sensor surface there is an electrolyte solution, provided within a thin hydrophilic spacer and coupled to the skin via a highly gas permeable hydrophobic membrane. PCO₂ determines a pH change of this electrolyte solution by diffusing through the skin to the sensor. The equation below illustrates the reaction occurring between carbon dioxide and water: (1) CO 2 + H 2 O ↔ H 2 CO 3 ↔ H + + HCO 3 −

Generation of hydrogen ions at the glass electrode is directly proportional to the amount of carbon dioxide which is present. An increase in the hydrogen ion concentration leads to a fall in pH and this change is proportional to the logarithm of the PCO₂ change.

The heating element, increasing local perfusion, induces the arterialization of the capillary blood flow. However, due to the elevated temperature of the sensor, the transcutaneous PCO₂ results higher than the arterial value mostly due to two factors [2]: first, the elevated temperature increases local blood and tissue PCO₂ (“anaerobic factor”); second, the epidermal cells produce CO₂, which contributes to the capillary CO₂ concentration by a constant amount (“metabolic constant”). As a consequence, in order to provide a monitor readout corresponding as close as possible to PaCO₂, the device applies an algorithm correction of the transcutaneous values as already specifically described by Severinghaus [3].

Since the monitor automatically performs calibration, the sensor is steadily ready to use. The small-sized sensor is easily attached to the ear lobe with a low pressure clip, and has been tested in adults [4–6], children [7], and neonates with birth weight (BW) >1,500 g and gestational age (GA) >28 weeks [8,9]. TOSCA monitor resulted reliable and applicable also in very low birth weight infants when the sensor was placed at the ear pinna, rather than at the earlobe, because of its small size. Figure 1(b) [10].

Since the use of pulse oximetry immediately after birth is indicated to titrate oxygen supplementation, as recommended by the Guidelines for Neonatal Resuscitation [11], the present study was designed to test the usability and reliability of TOSCA in the delivery room (DR) among term, late-preterm, and preterm neonates.

Of the 30 neonates born during period A (GA: 37.9 ± 1 weeks; BW: 3138 ± 459 g, 13 males and 17 females), none needed respiratory support at birth. No differences between SpO_2TOSCA and SpO_2MASIMO were detected at all time points (Figure 2(top)). For both monitors, data were available 60 s after sensor placement. The clipping of the TOSCA sensor was easy. Sensor repositioning was necessary in 16/30 registrations (53%) with TOSCA and in 5/30 registrations (16.6%) with MASIMO (p = 0.003). After sensor repositioning, time required to obtain a stable reading was shorter for TOSCA than for MASIMO (about 10 vs. 15 s). PtcCO_2TOSCA progressively increased during the whole monitoring (Figure 2(bottom)).

PCO_2BGA at 10 min was 61.5 ± 10.8 mmHg, significantly higher than PtcCO_2TOSCA (48.6 ± 11.7 mmHg; p < 0.005). At 10 min, the bias (precision) between PtcCO_2TOSCA and PCO_2BGA values was 12.9 (26.4). No signs of skin erythema or burn at the ear pinna were found at the end of monitoring.

None of the 10 neonates (GA: 36.1 ± 1.4 weeks; BW: 2743 ± 714 g; five males and five females) born during period B needed respiratory support. As shown in Figure 3, during the first 5 min PtcCO_2TOSCA values resulted significantly lower than PtcCO_2TINA values, while at 10 min similar values were recorded (PtcCO_2TOSCA 52 ± 17.6 mmHg vs. PtcCO_2TINA 58.6 ± 20.2 mmHg; p = ns).

PCO_2BGA at 10 min was 59.4 ± 17.3 mmHg; the bias (precision) between PtcCO_2TOSCA and PCO_2BGA values was 7.42 (39.8), between PCO_2BGA and PtcCO_2TINA 0.82 (31.9), and between PtcCO_2TOSCA and PtcCO_2TINA 6.6 (43.4). Sensor repositioning was required 3 times (30%) for TOSCA.

Fifteen preterm neonates (GA: 31.6 ± 2.0 weeks; BW: 1622 ± 569 grams; nine males and six females) born during period C were monitored for 40 min with TOSCA and MASIMO. Two neonates required resuscitation and were transferred to NICU mechanically ventilated; six neonates were assisted with nasal continuous positive airway pressure. Sensor repositioning was required six times (40%) for TOSCA and never for MASIMO. SpO_2TOSCA and SpO_2MASIMO values were similar during the whole study period (Figure 4(top)). PtcCO_2TOSCA values progressively increased until 10 min, then became stable (Figure 4(bottom)).

PCO_2BGA at 10 min was 63.1 ± 20.3 mmHg and the bias (precision) between PtcCO_2TOSCA and PCO_2BGA was 11.1 (43.4).

The Guidelines on Neonatal Resuscitation recommend the use of pulse oximetry to monitor neonates' oxygenation status in the DR [11], since skin color is only partially reliable [17]. Moreover, awareness about neonatal oxygenation status is helpful to titrate oxygen supplementation, if requested.

In case of frequent motion artifacts or low perfusion, often occurring in DR, pulse oximeters with MASIMO Signal Extraction Technology are indicated [18].

TOSCA monitor records SpO₂ with MASIMO technology and simultaneously detects PtcCO₂, allowing a combined assessment of both parameters. The present study investigated for the first time the possible use of TOSCA monitor in neonates immediately after birth, a period in which the use of excessive pressure or volume (overstretching) and oxidative aggression (reactive oxygen and nitrogen species) may cause acute damage to the lung and other organs, in particular in preterm babies. The simultaneous measurement of SpO₂ and PtcCO₂ could be helpful both to verify the physiologic changes occurring during transition from the intra- to the extrauterine life and to determine the appropriate grade of respiratory support when resuscitation is required.

Concerning usability, our results showed that TOSCA sensor can be easily applied, even in the DR immediately after birth, to the ear pinna of both term and preterm neonates. As previously demonstrated [10], sensor application to the ear pinna rather than to the ear lobe did not alter data displaying. The size of the clipped sensor is well suited for the thin, flat, and soft ear pinna of the newborn, while it may result too large compared to the small ear lobe. Nevertheless, dislocations occurred more often using TOSCA than MASIMO. In this study, repositioning of the TOSCA sensor was more frequently required among term than among late-preterm or preterm neonates, presumably because of their greater mobility. However, the viability of TOSCA monitoring was not impaired thanks both to the shorter time required for sensor clipping and to its faster reading, due to the high sensor's stability [1,2].

No burns underneath the TOSCA sensor were identified in any of the infants, including premature neonates monitored for longer period (40 min) in period C. This was presumably due to a lower electrode temperature after 20 min of monitoring in the QUICKSTART mode. In fact, this modality allows a long term (up to 12 h) application of the sensor without adverse events, although measurements of PtCO₂ may result less reliable [9]. This could be due to the lower sensor temperature which is not able to provide an adequate grade of arterialization in order to compensate the greater CO₂ local production induced by temperature itself. However, this occurrence has only limited implications in our study, since we performed a “long time evaluation” of the PtcCO_2TOSCA exclusively in Period C, throughout a monitoring period which is much shorter than the one described by other authors [9].

Concerning the reliability of SpO_2TOSCA, the recorded values were comparable to those detected by SpO_2MASIMO. As formerly demonstrated [10], the different placement site of both sensors, always preductal, did not affect SpO₂ detection.

Regarding PtcCO_2TOSCA measurement in DR, the estimation accuracy of TOSCA resulted generally acceptable compared to blood samples collected at 10 min of use. Although this single BGA, representing a punctual evaluation, may be a limit in the estimation of PtcCO₂ values, a higher number of blood collection was avoided to prevent painful punctures. However, the agreement between PtcCO_2TOSCA and PCO_2BGA was more accentuated when blood samples were taken by the umbilical line (period B and C) compared to the capillary sampling (period A). A possible explanation of these results may be that capillary PCO_2BGA is comparable to arterial PCO_2BGA only after the first hours of life [19]. Nevertheless, after 10 min of registration, the TOSCA monitor reveals less accurate PtcCO₂ values than the TINA one compared to the umbilical PCO_2BGA (period B), confirming that the measurement of PtCO₂ with TOSCA was less precise than the conventional monitoring [10].

The results achieved throughout period B demonstrated that during the first minutes of registration the PtcCO_2TOSCA values were lower than those recorded by TINA and became comparable only after 5 min. This is in agreement with previously reported data achieved among preterm neonates admitted to NICU, and was probably dependent on the placement site of the TOSCA sensor [8]. The thin dermal layer and the poor capillary bed of the ear pinna may induce a lower grade of arterialization than the one obtained in other sites [10].

A similar slowly increasing trend of the PtcCO_2TOSCA was also detected among preterm infants monitored after birth and during transport to the NICU (period C), and was followed by value stabilization after 10 min of monitoring.

We can assert that TOSCA was easily and safely usable in DR both in term and preterm neonates, allowing a reliable and noninvasive estimation of SpO₂. Considering that at least 10 min were required to obtain reliable and stable values of PtcCO_2TOSCA, TOSCA reliability results limited during the first minutes of life, when resuscitation procedures may be necessary. However, even if the optimal level of PCO₂ during neonatal resuscitation is not known yet [20], PtcCO₂ measurement with TOSCA could be useful as trend monitoring after neonatal stabilization and during transport to the NICU to prevent lung injury, since its beginning may occur soon after birth among neonates requiring positive-pressure ventilation. We believe that our study may represent a stimulating starting point for further investigations focusing on the development of respiratory function monitoring devices to be used in DR [21]. In the near future, an efficient technology may enable the individualization of the specific ventilatory and oxygenation requirements, thus minimizing physical and oxygen-derived damage [22].