In this case the coefficients of R in Equation (3) are constants and can be determined through calibration, as described above [5,11]. If the parameter l2/l1 changes between different subjects, in particular between the healthy subjects on whose fingers the calibration was performed and the patients on whose fingers the clinical examination is carried out, inaccuracy in SpO2 measurement is to be expected.1.2. Accuracy of Pulse OximetryMost manufacturers of pulse oximeters claim an accuracy of 2% [13,14], which is the standard deviation (SD) of the differences between SpO2 and SaO2 measured on extracted arterial blood samples by a co-oximeter. An SD of 2% means that an error of 4% or more (2SD) is expected in 5% of the examinations, assuming that the differences between SpO2 and SaO2 are normally distributed.

Since the full range of SaO2 in adults, even including most of the sick patients is 80%�C100%, an error of 3%�C4% may be significant.Despite its potential error of 3%�C4%, pulse oximetry is a clinically valuable tool for monitoring patients during surgery or rehabilitation, ensuring intraoperative and post-operative patient safety. Pulse oximetry can detect a sudden decrease of SpO2 by 3%�C4%, enabling the early detection of acute deterioration of the respiratory function of the patient or technical failure of mechanical ventilatory support and is, therefore, considered a mandatory tool for most patients in operating rooms and intensive care units.An inaccuracy of 3%�C4% in oxygen saturation measurement is, however, too high to allow such pulse oximeters to be used in routine management of various patient populations, in particular critically ill patients.

Because of the Brefeldin_A low accuracy of SpO2 measurements, some researchers suggested that SpO2 levels as high as 94% or 96% should be maintained, in order to ensure SaO2 value of 90% during mechanical ventilation and oxygen support [1,15,16]. In a study on critically ill patients, relatively low correlation was found between spontaneous changes in SpO2 and in SaO2 (r = 0.6, r2 = 0.37) [17], and the authors inferred that changes in SpO2 do not reliably predict equivalent changes in SaO2 in the critically ill. In a study on patients with severe sepsis and septic shock in emergency department, the mean difference between SpO2 and SaO2 was 2.75% and the standard deviation 3.1% [18].

The low accuracy of pulse oximetry can be attributed, at least partly, to the calibration process currently used for pulse oximeters, which is based on examinations on healthy volunteers and is not necessarily applicable to sick patients, in particular neonates [14]. Since the calibration process is based on SaO2 measurements by co-oximeter, the inaccuracy in the latter also contributes to the error in SpO2 measurement [13].