Near Infrared Spectroscopy: Toy or Tool?


A short history of oximetry During the second half of the 19th century it was discovered that blood cells contain a colouring substance whose spectrum is influenced by various blood gases [Hoppe, 1862]. Stokes discovered that this coloured substance was the oxygen carrier of the blood. Felix Hoppe-Seyler, a German chemist, named the substance haemoglobin. He found that when shaking a solution of haemoglobin with air this resulted in a spectral change. He called the component responsible for this change “oxyhaemoglobin”. In 1876 Karl von Vierordt published the first study in which spectroscopy was used to study haemoglobin in tissue. He found a transition from oxyhaemoglobin to deoxyhaemoglobin in the tissue of the finger after the arm was fully occluded, a technique still practised today to study oxygenation in muscles (refer to Figure 1.2 at page 16). At the begining of the 20th century oximetry developed rapidly in Germany with scientists like Nicolai [1932], Kramer [1934], Matthes [1942] and Gross. Last two scientists mentioned were the first to use a wavelength region in the near infrared where absorption was independent of the oxygenation status of blood and which could therefore compensate for changes in blood volume, light scattering and tissue thickness [Matthes et al. 1939]. Their instrumentation however was still very bulky and not easy to use. The first portable instrument measuring haemoglobin saturation in tissue accurately and automatically was built and described by Millikan [1942]. He named the instrument an “oximeter”. The sensor weighed 30 grams, used two wavelengths and could be slipped over the subject’s ear. To obtain arterial saturation the ear was moderately heated, a method also used in later versions of oximeters. Millikan’s instrument had an accuracy of 3 to 5% at the higher end of the scale ( — 98% saturation) and an accuracy of 8% at the lower end of the scale ($∼$50% saturation). The oximeter of Millikan was improved by Wood and Geraci [1949]. Further developments on reflection oximetry were done by Brinkman and Zijlstra. They developed the “Cyclops”, a device measuring oxygen saturation on the forehead of a subject [Brinkman et al. 1950]. In the following years, oximetry was used more and more in clinical situations. The title of Zijlstra’s thesis [1951]: “Fundamentals and applications of clinical oximetry” indicated that they were far ahead of their time. In the seventies two important findings contributed to the development and clinical applicability of oximetry. The first was by Aoyagi, who found that the variations in arterial blood volume could be used to obtain a signal dependent only on arterial blood changes. With this knowledge the arterial oxygen saturation of the blood could be measured [Aoyagi et al. 1974, Nakajima et al. 1975]. This technique 10 Introduction was called pulse oximetry and has been developed into a reliable and widely used method [Kelleher 1989]. The second finding was by Jóbsis [1977] and was the beginning of near infrared spectroscopy (NIRS)