In critically ill neonates, peripheral perfusion and oxygenation assessment may provide indirect information on the circulatory failure of vital organs during circulatory shock. The development of pulse oximetry has recently made it possible to calculate the perfusion index (PI), obtained from the ratio between the pulsatile and nonpulsatile signals of absorbed light. The main goals of this study were: (1) to study foot PI; and (2) to evaluate the relationship between foot PI, obtained continuously by pulse oximetry, and a number of variables, i.e. blood flow (BF), oxygen delivery (DO2), oxygen consumption (VO2), and fractional oxygen extraction (FOE), measured indirectly by near-infrared spectroscopy (NIRS) on the calf in 43 healthy term neonates (weight 3474.6±466.9 g; gestational age 39.1±1.4 weeks).
Calf BF, DO2 and VO2 were assessed by NIRS on short-lived venous and arterial occlusion maneuvers. PI was measured on the contralateral foot.
Foot PI was 1.26±0.39. There was a positive correlation between foot PI and both calf BF (r=0.32, p=0.03) and DO2 (r=0.32, p=0.03), but no correlation was found between foot PI and calf FOE and between foot PI and VO2.
In the neonatal intensive care unit, continuously measuring foot PI by pulse oximetry seems clinically more feasible for peripheral perfusion monitoring than spot measurements of the calf BF and/or VO2 by indirect NIRS.
Assessing and maintaining the adequacy of tissue oxygen supply should be considered a primary objective in neonatal/pediatric/adult intensive care. The peripheral perfusion of critical care patients is related to the redistribution of marginal cardiac output and oxygen supply to the brain, heart, and adrenal glands. Under stress-free conditions, newborn skin perfusion is high by comparison with oxygen demand.1
In terms of oxygen availability, transcutaneous PaO2 and pulse oximetry have become invaluable for monitoring critical newborn infants. The development of pulse oximetry has recently made it possible to calculate the perfusion index (PI),2 obtained from the ratio between the pulsatile signal of light absorbed by the pulsating arterial inflow and the nonpulsatile signal (light absorbed by the skin, other tissues and venous or nonpulsatile blood).3 The tissue PI varies with the quantity of red blood cells in the skin microvasculature and it is a reliable indicator of changes in skin blood flow (BF) in humans and animals.4 The amplitude waveform of the pulse oximeter can be used as a noninvasive measurement of volume status in critically ill adult patients.5
It has recently been reported that a foot skin PI value ≤1.24 is an accurate predictor of illness severity in neonates,3 and a PI value ≤1.4 indicates hypoperfusion in adults.6 Near-infrared spectroscopy (NIRS) has been used largely in neonates for the noninvasive assessment of brain tissue oxygenation7 and, less often, to evaluate limb oxygenation.8, 9, 10, 11, 12 Considering that: (1) the spot measurement of BF and VO2 by NIRS takes at least 20 minutes; and (2) the accuracy of NIRS changes is limited because path length and its changes are not known,7 the foot PI measured by pulse oximetry seems to be more feasible than indirect NIRS methods for monitoring peripheral perfusion in the neonatal intensive care unit. We hypothesized that foot PI could also adequately evaluate limb perfusion.
The main goals of this study were: (1) to study foot PI and a number of variables (BF, DO2, VO2, FOE) measured indirectly by NIRS on the calf; and (2) to evaluate the relationship between the foot PI obtained continuously by pulse oximetry and NIRS variables in a group of healthy term neonates.
MATERIALS AND METHODS
The observational study involved 43 healthy newborn infants aged from 1 to 5 days, born since 2003 and admitted to the Neonatal Nursery at the Pediatric Department of Padua University at birth. Ethical approval for the study was obtained from the local ethics committee and informed consent was obtained from the parents.
The mean gestational age was 39.1±1.4 weeks (35 and 42 weeks) and the averaged birth weight was 3474.6±466.9 g (2280 and 4560 g).
The neonates were studied while sleeping or awake, dressed, lying supine in their beds. If necessary, the child was calmed with a glucose-coated pacifier. Axillary temperatures ranged between 36.5 and 37°C. Eligibility criterion was the absence of disease in normal full-term babies. None of the infants were receiving medication or supplemental oxygen.
The principle behind pulse oximetry relies on the quantity of infrared (940 nm) light transmitted by a luminous source to a photodetector through a translucent and perfused site such as the neonate's hand, finger or foot. The PI was expressed as a ratio of absorbed arterial inflow light (AC) divided by the nonpulsatile absorbed light (venous and nonpulsatile blood/tissue) (DC), PI=AC/DC. This value and arterial saturation (SaO2) were obtained using a Masimo SET radical pulse oximeter (Masimo Corp., Irvine, CA, USA) with a sensor placed on the foot contralateral to the side used for the NIRS measurements. Foot PI measurements were collected just before the venous occlusion, averaging 30 seconds data from the neonate's leg when it was not moving.
NIRS measurements were obtained using the NIRO-300 oximeter (Hamamatsu Photonics, Hamamatsu City, Japan). The design and features of this device are described elsewhere.13 The optical probe consisted of one emitter and one detector (comprising three separate sensors) placed 3.5 cm apart. The two optodes in the optical probe (the distal end of the emitter and the detector) were kept at a constant distance and geometry by a black rubber shell firmly attached to the skin of the main part of the calf (2 cm above the outer malleolus of tibia) with double-sided adhesive tape. Dark felt was used to cover the neonate's leg to prevent ambient light from reaching the optodes. Based on the spatially resolved spectroscopy approach, the NIRO-300 provides both relative concentration changes (expressed in ΔμM) of oxyhemoglobin (O2Hb), deoxyhemoglobin (HHb), and the derived total hemoglobin content (tHb=O2Hb+HHb). O2Hb and HHb concentration changes are calculated using the central sensor alone as the detector, and an algorithm incorporating the modified Beer–Lambert law13 with a differential path length factor of 4.3 measured on adult muscle.14 No differential path length data are available for the neonatal calf. The procedure involved minimal disturbance to the neonate. In the case of venous occlusion, the child's leg was raised 5 cm above heart level with a foam rubber wedge-shaped cushion to facilitate venous drainage after the occlusion.
A disposable noninvasive blood pressure cuff for neonates (No. 4 or 5, depending on the diameter of the thigh) (Agilent Technologies GmbH, Boeblingen, Germany) was placed on the proximal part of the thigh. A pressure-inducing device (DP 903, Tenno Laboratories, Trento, Italy) was used to ensure a standardized, prompt cuff inflation. This device enables a sudden pressure increase to be produced by a suitably adapted machine consisting of a diaphragm compressor with a piston inside a compression chamber that can hold up to a maximal air pressure of 150 mmHg with a rate of inflation and deflation of about 0.2 seconds. The device is connected to the cuff by means of a soft plastic hose (diameter 0.5 cm).
After positioning and fixing the optical probe, the NIRO 300 was initialized and the measurements started. The NIRO 300 data were collected at a rate of 6 Hz and transferred on-line to a computer. The occlusions were considered unsatisfactory if artifact movements were recorded, in which case initialization and occlusions were repeated in stable conditions.
NIRS and Venous Occlusion to Calculate Blood Flow
Automatic rapid cuff inflation (30 to 40 mmHg for 10 seconds) on the thigh proved essential to obtain a venous occlusion, adapting the method already applied to adults.15
During venous occlusion, hemoglobin flow (Hb flow) equates to the rate of tissue Hb accumulation during the initial part of the occlusion. The Hb flow was calculated, according to Wardle and Weindling,8 from the slope of tHb (Hb flow=∫ΔtHb dt) during the first 3 seconds of the occlusion. Calf BF was calculated by dividing Hb flow by Hb concentration (μM) established from a heel-prick sample.
SvO2 was calculated during the first 3 seconds of venous occlusion according to Yoxall and Weindling:9 SvO2=Δ[O2Hb]/Δ[tHb], that is, the change in O2Hb concentration divided by the change in tHb during the interval used to calculate the blood flows. Initial occlusion artifacts caused by the abrupt inflation of the blood pressure cuff were rare.16 DO2 and VO2 were calculated as follows: DO2=BF × Hb × 4 × SaO2 and VO2=BF × Hb × 4 × (SaO2−SvO2), where 4 corresponds to the molar ratio due to 4 moles of oxygen being carried by each mole of Hb. FOE was calculated as the ratio VO2/DO2=(SaO2−SvO2)/SaO2.
NIRS and Arterial Occlusion to Calculate VO2
The cuff was inflated rapidly on the thigh to 100 mmHg for 30 to 40 seconds to obtain an arterial occlusion; VO2 was calculated from the rate at which O2Hb changes to HHb during the first 3 seconds=∫ΔO2Hb dt.10, 17 The interval between occlusions was set at 5 minutes to ensure a sufficient recovery period.
A typical time course of the O2Hb and HHb changes of a newborn calf during three venous occlusions followed by three arterial occlusions is shown in Figure 1. Venous occlusion caused an increase in O2Hb and HHb; arterial occlusion caused a decrease in O2Hb and an increase in HHb.
Few measurements were rejected due to movement artifacts or for other technical reasons (rejections ranged from 0 to 3, on both arterial and venous occlusions).
μM O2/100 ml/minute is comparable with ml O2/kg/minute,19 assuming that a mole of a gas, at body temperature and sea level, corresponds to a 25.4 l (Avogadro's law); and 1 kg corresponds to 1000 ml, assuming a muscle density of 1.06 g/ml.20
Data Analysis and Statistics
All parameters were checked for normal distribution using the Kolmogorov–Smirnov test procedure. Data are presented as mean±SD. Least square regression analysis was used to study the relationship between foot PI and calf BF, DO2, FOE, and VO2. Correlation analysis was performed using the Pearson's correlation coefficient, given the normal distribution of our data. A p-value <0.05 was considered statistically significant. The statistical analysis was performed using the Epi Info program21 and Excel for Windows XP.
The O2Hb and HHb data points, used to calculate BF and/or VO2, had a regression coefficient (r) higher than 0.8. The reported data for each subject are the average of only the valid occlusions.
Clinical details, calf flow/oxygenation parameters, and foot PI are summarized in Table 1. A positive correlation was found between foot PI and calf BF (r=0.32, p=0.03) (Figure 2), and between foot PI and DO2 (r=0.32, p=0.03).
No correlation was found between foot PI and calf FOE, or between foot PI and VO2 calculated by arterial or venous occlusion. The difference (SaO2−SvO2) correlated with VO2 measured by venous occlusion (r=0.36, p=0.01), but not with VO2 measured by arterial occlusion.
A strong correlation emerged between calf DO2 and calf VO2 calculated by the venous occlusion method (r=0.88, p=6.72E−15) (Figure 3), but not with VO2 by arterial occlusion.
This study tested the relationship between foot PI obtained continuously by pulse oximetry and a number of variables (BF, DO2, VO2, FOE) measured indirectly by NIRS on the calf of a group of healthy term neonates. Despite the difference in the measuring points and penetration depth between the pulse oximetry and NIRS measurements, calf BF significantly correlated with foot PI, suggesting that the latter can be used to evaluate peripheral perfusion noninvasively and continuously. To the best of our knowledge, this is the first time that foot PI, measured by pulse oximetry, is compared with parameters relating to the peripheral perfusion and oxygen consumption measurable by NIRS.10, 11, 12, 15, 16, 18, 22
Only recent developments of pulse oximetry have enabled foot PI to be calculated.2, 3 Although the mechanisms underlying PI changes remain to be elucidated, local skin vasoconstriction appears to be the most likely reason for significantly lower PI readings in very sick patients.23 On the other hand, very little data are available as yet (obtained by methods of varying complexity, including skin colorimetry) on microcirculatory skin perfusion in neonates.1, 23, 24 Since pulse oximetry requires relatively low-cost, standard equipment in the neonatal intensive care unit, the PI is easy to monitor on the newborn's foot or hand, whereas NIRS measurements take at least 20 minutes, the equipment is relatively expensive, and the accuracy of NIRS measurements is limited. In fact, the commercially available NIRS instrumentation does not measure the path length needed to express tissue O2Hb and HHb changes in absolute units (μM).7 A constant differential path length factor was used in the present study, and this might explain the moderate correlation between PI and the variables measured by NIRS. Given the lack of other noninvasive methods, NIRS has been the only bedside clinical tool available in the last few years for evaluating muscle circulatory status in healthy and sick neonates and many significant clinical studies have been published.8, 9, 10, 11, 12, 16, 18, 22, 25
NIRS was combined in our study with arterial/venous occlusion maneuvers to measure local muscle BF, VO2, and SvO2.9, 10, 22 Using this NIRS approach, FOE proved to be a better indicator of the need for transfusion in anemia than hemoglobin concentration.18, 25 The relationship between cerebral and peripheral FOE was lost in hypotensive preterm infants, but not in anemic babies.22
Our healthy term neonates had a higher mean FOE (0.44) than healthy preterm newborns (0.3 to 0.35)18, 22 or preterm anemic infants (0.33),18 so it was higher than the median value of 0.43 reported for symptomatic preterm newborn infants with anemia. We might speculate a higher VO2 in relation to DO2, or FOE, in our resting neonates than in normal or anemic preterm infants, associated in our series with a limb activity prior to the assessment, or with a better muscle tone.
Bone BF and skin BF may lead to an underestimation of NIRS-measured BF in the legs of our newborn infants, and so may the thickness of subcutaneous fat. Bay-Hansen et al.16 hypothesized that neonatal peripheral BF should not differ from young adults, and our results also suggest that healthy neonates have a lower flow rate than severely ill newborn.16, 18
Like De Blasi et al.,15 we consider muscle BF and VO2 as depending mainly on muscle activity. We nonetheless suggest that forearm BF may also be influenced by changes in the outside environment, as well as by intrinsic hemodynamic postnatal adaptation, that is patent ductus arteriosus. This would explain not only a BF change due to the arterial inflow, but also a BF redistribution from the capillaries to the skeletal muscles26 according to the energy demands. Neonatal calf and forearm VO2 were measured by NIRS on both arterial and venous occlusion, but Hassan et al.10 demonstrated that arterial occlusion produces more consistent results; they used arterial occlusion to investigate the effects of a change in global metabolic rate on peripheral oxygen consumption,12 and the effect of limb cooling on peripheral and global oxygen consumption.11
As for VO2 measured by arterial occlusion, Hassan et al. gave us the only comparable measurements in healthy newborn infants.10 We agree with his results because this method gave us a lower coefficient of variation than venous occlusion, as well as a lower VO2 value (0.3 μM/100 ml/minute) than the 3.2 μM/100 ml/minute obtained using venous occlusion. Hassan et al.11 reported 10.4 μM/100 ml/minute for arterial occlusion and 14.8 μM/100 ml/minute for venous occlusion. Both values are higher than our resting values in both sets of measurements. On venous occlusion, the VO2 in healthy young adults19 coincided with a value (1.1 ml/kg/minute translates to 4.3 μM/100 ml/minute) comparable with the one recorded in our neonates. Bay-Hansen et al.16 found a higher VO2 on venous occlusion in his sick neonates than those recorded in our healthy term neonates. Our low VO2 values may be partially justified by limb cooling, due to the forearm being exposed to room air during the study period. According to Hassan et al.,11 in fact, mild limb cooling (just over 2°C) due to room air exposure for 15 to 20 minutes seems to reduce arterial VO2 by 19.6%.
The relationships we found between the perfusion index and both BF and DO2 might seem obvious, but this study also demonstrated them quantitatively. On the other hand, we found no correlation between PI and muscle metabolic demand, particularly when either VO2 (calculated using the arterial or venous occlusion method) or FOE are considered. Our neonates were at rest, with neither muscle activity nor sepsis — both of which could increase BF and VO2 in adults.19, 27 The relationship we found between DO2 and VO2 may have been due mainly to the use of BF in the calculation of both the values. Moreover, this does not mean that oxygen utilization is perfusion-limited or supply-dependent, as in septic shock patients.28, 29 In fact, our population had an excessively narrow arterial–venous saturation difference, showing a low muscle oxygen consumption at rest.
In conclusion: (1) our data seem useful with a view to evaluating both peripheral perfusion and oxygen metabolic demand in term healthy neonates. Our mean PI value differed from the one reported by De Felice et al.,3 but our neonates were more homogeneous as a sample and not liable to the hemodynamic changes of preterm infants; (2) considering the limits of quantifying O2Hb and HHb using commercial NIRS tools and the relatively long time it takes to perform BF indirect measurements, PI monitoring using pulse oximetry — as an index of peripheral perfusion — would represent a very simple bedside clinical tool for evaluating the peripheral circulatory status of healthy and sick neonates. PI provides helpful information correlating with BF and DO2 data from NIRS, but not with VO2 or FOE. Monitoring peripheral circulatory status could become a regular part of neonatal intensive care.
Genzel-Boroviczeny O, Strotgen J, Harris AG, Messmer K, Christ F . Orthogonal polarization spectral imaging (OPS): a novel method to measure the microcirculation in term and preterm infants transcutaneously. Pediatr Res 2002;51:386–391.
Goldman JM, Petterson MT, Kopotic RJ, Barker SJ . Masimo signal extraction pulse oximetry. J Clin Monit Comput 2000;16:475–483.
De Felice C, Latini G, Vacca P, Kopotic RJ . The pulse oximeter perfusion index as a predictor for high illness severity in neonates. Eur J Pediatr 2002;161:561–562.
Hales JR, Stephens FR, Fawcett AA, et al. Observations on a new non-invasive monitor of skin blood flow. Clin Exp Pharmacol Physiol 1989;16:403–415.
Partridge BL . Use of pulse oximetry as a noninvasive indicator of intravascular volume status. J Clin Monit 1987;3:263–268.
Lima AP, Beelen P, Bakker J . Use of a peripheral perfusion index derived from the pulse oximetry signal as a noninvasive indicator of perfusion. Crit Care Med 2002;30:1210–1213.
Nicklin SE, Hassan IA, Wickramasinghe YA, Spencer SA . The light still shines, but not that brightly? The current status of perinatal near infrared spectroscopy. Arch Dis Child Fetal Neonatal Ed 2003;88:F263–F268.
Wardle SP, Weindling AM . Peripheral oxygenation in preterm infants. Clin Perinatol 1999;26:947–966.
Yoxall CW, Weindling AM . The measurement of peripheral venous oxyhemoglobin saturation in newborn infants by near infrared spectroscopy with venous occlusion. Pediatr Res 1996;39:1103–1106.
Hassan IA, Spencer SA, Wickramasinghe YA, Palmer KS . Measurement of peripheral oxygen utilisation in neonates using near infrared spectroscopy: comparison between arterial and venous occlusion methods. Early Hum Dev 2000;57:211–224.
Hassan IA, Wickramasinghe YA, Spencer SA . Effect of limb cooling on peripheral and global oxygen consumption in neonates. Arch Dis Child Fetal Neonatal Ed 2003;88:F139–F142.
Hassan IA, Wickramasinghe YA, Spencer SA . Effect of a change in global metabolic rate on peripheral oxygen consumption in neonates. Arch Dis Child Fetal Neonatal Ed 2003;88:F143–F146.
Suzuki S, Takasaki S, Ozaki T, Kobayashi Y . A tissue oxygenation monitor using NIR spatially resolved spectroscopy. Proc SPIE 1999;3597:582–592.
Duncan A, Meek JH, Clemence M, et al. Optical pathlength measurements on adult head, calf and forearm and the head of the newborn infant using phase resolved optical spectroscopy. Phys Med Biol 1995;40:295–304.
De Blasi RA, Ferrari M, Natali A, Conti G, Mega A, Gasparetto A . Noninvasive measurement of forearm blood flow and oxygen consumption by near-infrared spectroscopy. J Appl Physiol 1994;76:1388–1393.
Bay-Hansen R, Elfving B, Greisen G . Use of near infrared spectroscopy for estimation of peripheral venous saturation in newborns: comparison with co-oximetry of central venous blood. Biol Neonate 2002;82:1–8.
De Blasi RA, Cope M, Elwell C, Safoue F, Ferrari M . Noninvasive measurement of human forearm oxygen consumption by near infrared spectroscopy. Eur J Appl Physiol Occup Physiol 1993;67:20–25.
Wardle SP, Yoxall W, Crawley E, Weindling AM . Peripheral oxygenation and anemia in preterm babies. Pediatr Res 1998;44:125–131.
Girardis M, Rinaldi L, Busani S, Flore I, Mauro S, Pasetto A . Muscle perfusion and oxygen consumption by near-infrared spectroscopy in septic-shock and non-septic-shock patients. Intensive Care Med 2003;29:1173–1176.
Winter DA . A new definition of mechanical work done in human movement. J Appl Physiol 1979;46:79–83.
Harbage B, Dean AG . Distribution of Epi Info software: an evaluation using the internet. Am J Prev Med 1999;16:314–317.
Wardle SP, Yoxall CW, Weindling M . Peripheral oxygenation in hypotensive preterm babies. Pediatr Res 1999;45:343–349.
De Felice C, Flori ML, Pellegrino M, et al. Predictive value of skin color for illness severity in the high-risk newborn. Pediatr Res 2002;51:100–105.
Beinder E, Trojan A, Bucher HU, Huch A, Huch R . Control of skin blood flow in pre- and full-term infants. Biol Neonate 1994;65:7–15.
Wardle SP, Weindling AM . Peripheral fractional oxygen extraction and other measures of tissue oxygenation to guide blood transfusions in preterm infants. Semin Perinatol 2001;25:60–64.
Slaaf DW, Oude Egbrink MG . Capillaries and flow redistribution play an important role in muscle blood flow reserve capacity. J Mal Vasc 2002;27:63–67.
Van Beekvelt MC, Colier WN, Wevers RA, Van Engelen BG . Performance of near-infrared spectroscopy in measuring local O2 consumption and blood flow in skeletal muscle. J Appl Physiol 2001;90:511–519.
Astiz ME, Rackow EC, Falk JL, Kaufman BS, Weil MH . Oxygen delivery and consumption in patients with hyperdynamic septic shock. Crit Care Med 1987;15:26–28.
Fink M . Cytopathic hypoxia in sepsis. Acta Anaesthesiol Scand Suppl 1997;110:87–95.
About this article
Cite this article
Zaramella, P., Freato, F., Quaresima, V. et al. Foot Pulse Oximeter Perfusion Index Correlates with Calf Muscle Perfusion Measured by Near-Infrared Spectroscopy in Healthy Neonates. J Perinatol 25, 417–422 (2005). https://doi.org/10.1038/sj.jp.7211328
The effect of patent ductus arteriosus on pre-ductal and post-ductal perfusion index in preterm neonates
Physiological Measurement (2018)
Archives of Disease in Childhood - Fetal and Neonatal Edition (2018)
Biological Research For Nursing (2017)
Pediatric Research (2017)
The Journal of Pediatrics (2017)