CORRELATION BETWEEN MEASUREMENT OF ARTERIAL SA TURA TION BY PULSE OXIMETRY AND BY HEMOXYMETER IN CHILDREN WITH CONGENITAL HEART DISEASE

Transcription

1 16 CORRELATION BETWEEN MEASUREMENT OF ARTERIAL SA TURA TION BY PULSE OXIMETRY AND BY HEMOXYMETER IN CHILDREN WITH CONGENITAL HEART DISEASE OMAR GALAL, MO, PhO; NEIL WILSON, MO Pulse oximetry is a noninvasive method of assessing hemoglobin arterial saturation, heart rate, and pulse amplitude which, for the past few years, has been used as an alternative to arterial blood gas analysis. There are few reports describing its use in the management of children with congenital heart disease, and this was the aim of our study. Sixty-one oximetry studies were done in 55 children who underwent cardiac catheterization. All were studied prospectively. There were 33 males and 22 females, with a mean age of 6.4 years (range, 0.5 to 29 years). During the invasive procedure, arterial saturation (SAT) and saturation by arterial blood gases (ABG) were measured. using the OSM 2 hemoxymeter and an ABL 300 instrument, respectively. Hemoglobin saturation (OXY) was simultaneously assessed non invasively by a Nellcor N-l 00 E pulse oximeter. Statistical analysis of the data yielded an excellent correlation: OXY/SATr = 0.835; P< O.OOI;ABG/SATr = 0.850; P< In children with saturations lower than 85% (n = 16), the correlation was not significant (r = 0.419). The correlation for patients with oxygen saturation higher than 85% (n = 45) was significant: r = 0.669, P < In children less than 4 years of age, the measurement was more accurate (n = 32, r = 0.850, P < 0.001) than in children older than 4 years (n = 29, r = 0.757, P < 0.001). There was a good correlation betwt;en OXY and SAT in children with a hematocrit less than 45% (n = 36, r = 0.855, P < 0.001) when compared with the correlation of those with a hematocrit higherthan 45% (n = 25, r = 0.782, P < 0.001). We conclude that pulse oximetry is a reliable method for measuring hemoglobin saturation in children with congenital heart disease but should be used more selectively in children with low oxygen saturation. THE NONINVASIVE MEASUREMENT of oxygen saturation by pulse oximetry is attractive because it does not disturb the patient, is painless and easy to use, and does not require calibration. It can also be used for long-term monitoring as well as single measurements, the results can be read directly without the need for laboratory processing. The method of pulse oximetry was initially described in In the 1980s pulse oximetry was shown to have a wide range of clinical applications that include patient care in the perioperative and postoperative periods, in patient transport, From the Department of Cardiovascular Diseases, King Faisal Specialist Hospital and Research Centre, Riyadh, Saudi Arabia. Address reprint requests and correspondence to Dr. Galal: Section of Pediatric Cardiology, Department of Cardiovascular Diseases, King Faisal Specialist Hospital and Research Centre, P.O. Box 3354, Riyadh 11211, Saudi Arabia. and care in patients suffering from respiratory failure.2-4 However, there are few reports on the use of this method in children with congenital heart disease.5.6 Therefore, the aim of this study was to evaluate the accuracy of pulse oximetry and to determine its reliability in detecting oxygen des aturation in pediatric patients with congenital heart disease. Patients and Methods Sixty-one measurements were performed in 55 patients (6 patients were studied twice). There were 22 girls and 33 boys with a median age of 4 years (range, 2 days to 29 years). All patients underwent cardiac catheterization as part of a diagnostic or therapeutic investigation. The mean heart rate (:t SO) of the children was 110 :t 22 (71 Journal of the Saudi Heart Association. Vol. 4. No. I. 1992

2 ARTERIAL SATURATION 17 to 190) beats per minute, and the mean hematocrit value (i: SD) was 43 i: 8.8% (26% to 64%). The patients were supine throughout cardiac catheterization. The electrode of the pulse oximeter was attached to the right or left index finger of the patient. Hypotonic patients were excluded.7,s Care was taken to prevent direct light from falling on the electrode9 and to keep the patient warm during the measurement. During cardiac catheterization, simultaneous measurements of oxygen saturation from a pulse oximeter was compared with oxygen saturation measured in vitro on obtained heparinized arterial blood samples. The age, weight, height, heart rate, and hematocrit were documented for each patient. Fetal hemoglobin concentration was not measured because most patients were over six months of age. The apparatus used for the study was a pulse oximeter (Nellcor model N-lOO, Drager, Lubeck). To determine the arterial saturation of the blood sample, a hemoxymeter (OSM 2, Radiometer, Copenhagen) was used. Blood gas analysis was performed with an ABL 300 instrument (Radiometer, Copenhagen). Results Simultaneous data were obtained from 61 studies in the 55 patients. The correlation between arterial saturation measured by pulse oximetry (OXY) with the hemoxymeter (SAT) and by the blood gas analyzer (ABG) showed the following results. There was an excellent correlation between the arterial saturation measured by OXY when compared with the arterial saturation measured by SAT: OXY/SATn = 61, r = 0.835, P <0.001 (Figure 1A). Furthermore, the arterial saturation measured by the arterial blood gas monitor correlated well with SAT: ABG/SATn = 61, r = 0.850, P < (Figure 1B). Principle of Pulse Oximeter Pulse oximetry uses the technique of absorption spectroscopy. The pulse oximeter estimates arterial hemoglobin saturation by measuring the light absorbency of pulsating vascular tissue at two wavelengths of 660 and 920 nm. The relationship between measured light absorbency and saturation was developed empirically and is built into the oximeter software. Figure la. Relationship between the arterial saturation measured by pulse oximeter and by hemoxymeter (SAT). OSM 2 Hemoxymeter The OSM 2 hemoxymeter is a photometer that automatically measures hemoglobin saturation. This photometer uses two wavelengths of 505 and 600 nm. It differentiates between the wavelengths of oxyhemoglobin and desoxyhemoglobin and is the standard photometer for measuring oxygen saturation of blood samples taken in the catheter laboratory. Statistics Data were compared using linear regression analysis and calculation of correlation coefficients. The level of statistical significance was set at P < Figure lb. Relationship between the arterial saturation measured by blood gas analyzer and by hemoxymeter (SAT). Journal of the Saudi Heart Association, Vol. 4, No. I, \992

3 18 GALAL AND WILSON In children with saturations lower than 85% (n = 16), the correlation was not significant (r = 0.419) (Figure 2A). The correlation for patients with oxygen saturation higher than 85% (n = 45) was significant (r = 0.669, P < 0.001)(Figure 2B). In children less than 4 years of age, the measurement was more accurate (n = 32, r = 0.850, P < 0.OO1)(Figure 3A) than in children older than 4years(n = 29, r = 0.757, P<0.001)(Figure3B). There was a better correlation between OXY and SAT in children with a hematocrit less than 45% (n = 36, r = 0.855, P < 0.001)(Figure 4A), when compared with the correlation of those with a hematocrit higher than 45% (n = 25, r = 0.782, P < 0.001)(Figure 4B). < 4 Years Figure 3A. Relationship between the arterial saturation measured by pulse oximeter and by hemoxymeter for patients < four years old. 100 > 4 Years Figure 2A. Relationship between the arterial saturation measured by pulse oximeter and by hemoxymeter for patients with arterial saturation < 85%. Figure 38. Relationship between the arterial saturation measured by pulse oximeter and hemoxymeter for patients> four years old Hct < 45% Figure 28. Relationship between the arterial saturation measured by pulse oximeter and by hemoxymeter for patients with arterial saturation> 85%. Figure 4A. Relationship between the arterial saturation measured by pulse oxim~er and by hemoxymeter for hematocrit levels < 45%. Journal of the Saudi Hean Association, Vol. 4, No. t, 1992

4 ARTERIAL SA TURA non 19 Figure 4B. Relationship between the arterial saturation measured by pulse oximeter and by hemoxymeter for hematocrit levels> 45%. Discussion In the last ten years, pulse oximetry has been recognized to possess a wide range of clinical applications in the operating room, during patient transport, in postoperative care, and care in patients with respiratory failure.2-4 Few reports are available on the use of this method in children with congenital heart disease.5,6 An excellent correlation has been reported between oxygen saturation measured by pulse oximetry and by the standard techniques.8,10 Our data support reported findings, but many of these studies were concerned more with patients who were fully saturated.8 Some authors suggested that the method has limitations in patients with low oxygen saturation,5,1i,12 We were therefore interested to see if pulse oximetry would accurately reflect oxygen saturation in cyanosed patients with congenital heart disease. Our patients were divided into two groups. One group had saturations over 85% while the other had saturations less than 85%; measurements of standard techniques were used. Our data confirm the data of others5,11,12 that low saturations are not measured as accurately as high saturations by pulse oximetry. Lower saturations yielded more dispersed data than did higher saturations. Pulse oximetry tends to overestimate arterial saturation at values below 85%. For clinical use these limitations are not as important when values either exceed 90% or are lower than 80%, If a patient with a value far less than 80% shows a high hematocrit, then the decision for further management, whether it is for a systemic-pulmonary shunt or for corrective surgery, is relatively straightforward, We thought that it might be more difficult to measure saturation in children less than 4 years of age, but actually found that the opposite was true. This may be due to the fact that pulse detection is easier in younger children because they have less body mass. On the other hand, very small children are less likely to be cooperative and thus pose problems for pulse oximetry. In this study the patients were sedated when the measurements were made. Because many cyanosed children tend to have high hematocrit values, we evaluated the effect of a high hematocrit on pulse oximetry measurements. There is a good correlation between the pulse oximeter measurement and the arterial saturation obtained by the hemoxymeter in children with a hematocrit less than 45%. However, this correlation was not as good in children with a hematocrit higher than 45%. This seems to indicate that a higher blood viscosity has a deleterious effect on the accuracy of pulse oximetry. Desaturation, high hematocrit, and an age over four years have cumulative negative effects on the accuracy of the pulse oximeter reading in cyanosed children who have high hematocrit levels. The pulse oximeter detects changes in oxygen saturation quickly, accurately, and reliably. It is useful as an acute indicator of hypoxia. 13 One of our patients with pulmonary stenosis experienced sudden hypoxia during cardiac catheterization, following occlusion of the right ventricular outflow tract with a balloon catheter. Pulse oximetry recorded both the fall in saturation during the occlusion and its return to normal when the occlusion was relieved by deflating the balloon. To use pulse oximetry to its full potential, we must be aware of its limitations as well as its advantages. Because of the nature of the hemoglobin saturation dissociation curve, saturation measurements will not be sensitive to changes in Pa02 when the Pa02 is greater than 100 torr. Since the pulse oximeter uses two wavelengths of light, it cannot distinguish more than two hemoglobin species. Thus, carboxyhemoglobin and Journal of the Saudi Heart Association, Vol. 4, No. I, 1992

5 20 GALAL AND WILSON methemoglobin will cause errors in oxygen saturation readings if present in large amounts For example, smokers have high levels of carboxyhemoglobin and this might affect the results measured by pulse oximetry. We conclude from our results that a noninvasive oxygen saturation measurement by pulse oximetry is accurate for saturations above 85%. Saturations below 85% are less reliably estimated by this method, though for clinical use the degree of accuracy is acceptable. We believe that pulse oximetry should be useful in the follow-up of patients seen in a pediatric cardiology clinic. References 1. Severinghaus JW, Honda Y. History of blood gas analysis. VII. Pulse oximetry. J Clin Monit 1987:3: Cecil WT, Petterson MT, Lamoonpun S. et al. Clinical evaluation of biox MIA ear oximeter in the critical care environment. Respir Care 1985 ;30: Brodsky JB, Shulman MS, Swan M, et al. Pulse oximetry during one-lung ventilation. Anesthesiology 1985;63: Motoyama EK, Glazener CH. Hypoxemia after general anesthesia in children. Anesth Analg 1986;65: Gillor A. Schickendantz S, Hein~r K, et al. Pulsoxymet rie in der padiatrischen Kardiologie. Mschr Kinderheilk 1988:136: Lynn AM, Bosenburg A. Pulse oximetry during cardiac catheterization in children with congenital heart disease. J Clin Monit 1986;2(4): Solmano AJ. Smyth JA, Mann TK, et al. Pulse oximetry: advantages in infants with bronchopulmonary dysplasia. Pediatrics 1986;78: Yelderrnan M, New W. Evaluation of pulse oximetry. Anesthesiology 1983;59: Brooks TD, Paulus DA, Winkle WE. Infrared heat lamps interfere with pulse oximeters. Anesthesiology 1984;61: Marian F, Spiss CK, Hiesmayr M, et al. Uberwachung der fiberoptischen Intubation mittels nicht invasiver Pul soxymetrie. Anaesthesist 1985;34: Thys DM, Cohen E, Girard D, et al. The pulse oximeter: a noninvasive monitor of oxygenation during thoracic surgery. Thorac Cardiovasc Surg 1986;34: Fanconi S. Reliability of pulse oximcjtry in hypoxic infants. J Pediatr 1988; 112: Tyler IL. Tantisira B, Winter PM, et al. Continuous monitoring of arterial oxygen saturation with pulse oximetry during transfer to the recovery room. Anesth Analg 1985;64: Journal oflhe Saudi Heart Association, Vol. 4. No. I. 1992