Fiber-optic blood gas sensors have been under development for nearly twenty years. This paper reviews some recent developments in optical blood gas measurement and discusses several clinical applications of this technology.

Multiparameter sensors are fabricated by combining the optical pathways of imaging fibers with the spatial discrimination of CCD video cameras. Distinct sensing regions are placed on the distal face of an imaging fiber through site selective illumination of discrete regions of the fiber. Photopolymerization and detection systems are described.

A fiber optic-based biosensor which integrates a novel array of optical and electrical components, together with long fused silica fibers and proteins for detection of analyte in solution has been put into limited production. The optical fiber core near the distal end is tapered and coated with either antibodies or DNA binding proteins. Assays are performed by flowing a solution containing the fluorescently tagged ligand molecules over the coated fiber. Within seconds, analyte recognition occurs and a fluorescence signal is transmitted back up the fiber. Applications for the biosensor include clinical diagnostics, pollution control, and environmental monitoring. Ten of the laboratory devices described in this article have been constructed and are being used for assay development. In addition, the prototype of a truly portable device has recently been built and is now being tested.

The recent use of fiber optic waveguides for biochemical analyses is mainly based on light intensity changes. Waveguide ellipsometry sensors (WEFE) are proposed for biosensor applications. The WEFE is based on the principle of ellipsometry. A laser beam propagates in the waveguide with two fundamental modes: transverse electric field, parallel to and perpendicular to the sensing surface. The evanescent waves interact with the absorbed or conjugated layer of biomolecules, producing changes in the relative phase, (Delta) , and amplitude, (Psi) , upon interaction. The ellipsometric parameters change with biomolecule adsorption or conjugation on the sensing surface. From the ellipsometric parameter changes, the surface concentration, and thus the solution concentration, of the biomolecules can be estimated. Expected advantages of the WEFE are: (1) the WEFE measurement is independent of light intensity; (2) the WEFE measurement is independent of the waveguide coupler quality; (3) (Delta) is very sensitive to the thickness of the adsorbed layer, the WEFE is much more sensitive than sensors based on intensity; (4) optic fibers are commercialized, the WEFE will be easy to fabricate and use commercially. In this paper -- a comparison of WEFE within other biosensors -- the principle of WEFE including light propagation in waveguide and ellipsometry parameter analysis, the WEFE arrangement and instrumentation, and preliminary experimental results are given.

A fiber optic sensor for general anesthetics based on the phase transition of immobilized phospholipid vesicles is under development. Current work centers on evaluating the sensor response to different anesthetics and instrumentation design. The fluorescence of laurdan- doped liposomes is found to respond linearly to the infusible anesthetics thiopental sodium and Propofol. Preliminary experiments have been performed to determine sources of noise in the optical and electronic components of the sensor as it is now configured. One potential noise source is the liposome sample at the fiber tip; photobleaching and thermal fluctuations due to heating by the illuminating 360 nm radiation can affect measurement of the anesthetic level. Heating of the sample is a factor at high illumination levels, but photobleaching, which reduces the signal intensity, does not alter the intensity ratio upon which the anesthetic concentration measurement is based. Optical microscopy of fiber tips embedded in liposomes allows direct observation of the light intensity near the tip of the fiber despite the extreme turbidity of the suspension. Light intensity drops to less than 10% of its maximum intensity at the fiber tip within 300 micrometers . Further use of this technique should allow monitoring the effects of photobleaching on the spatial distribution of the liposomes responsible for the measured optical signal.

The dependence of the losses induced by ionizing radiation on the ambient conditions is studied for lead glass fibers in the parameter range of relevance for medical technology. On the basis of these results, a concept is developed for a fiber-optic dosimeter system and initial measuring results are communicated using an experimental arrangement of this type of dosimeter system under application-oriented conditions. Some applications of this system in radiotherapy are shown which are not feasible with conventional dosimeters at all, or only in exceptional cases, and which can contribute to safe handling and more effective use in radiotherapy.

Accurate measurement of ionizing radiation dose during radiotherapy planning and treatment is critically important to avoid damage to surrounding tissue. In order to improve the effectiveness of radiotherapy, i.e., maximizing the dose delivered to the cancerous region without damaging surrounding tissue, a real-time in-vivo dosimetry technique would be desirable for directly measuring the delivered dose. The primary method presently known to be suitable for direct dose measurements in vivo is by the use of thermoluminescent dosimeters (TLDs). Unfortunately, TLDs only provide integrated dose information some time after the patient has been irradiated. Therefore, the radiotherapist cannot adjust the exposure in real-time to ensure that the proper dose is delivered to the desired region. Additional limitations of TLDs include their poor dose reproducibility, limited dynamic range and sensitivity, and nonlinear response in certain cases. In order to overcome certain limitations of existing dosimetry systems, a new fiber optic radiation dosimeter has been developed using a patented electron trapping material. This system uses a SrS:Ce,Sm photostimulable storage phosphor which can be exposed to ionizing radiation to produce a population of trapped electrons. The SrS:Ce,Sm material can then be stimulated with 1 micrometers near-IR to produce a visible luminescence directly proportional to the radiation exposure. These processes are optically induced electron transitions and do not involve heating of the SrS:Ce,Sm phosphor, so that measurements are performed at ambient temperatures. The system is further evaluated for the feasibility of fabricating miniature fiber optic probes for real time in-vivo dosimetry.

This paper demonstrates the application of a lensless fiber optic spectrometer (sensor) to study the onset of cataracts. This new miniaturized and rugged fiber optic probe is based upon dynamic light scattering (DLS) principles. It has no moving parts, no apertures, and requires no optical alignment. It is flexible and easy to use. Results are presented for cold-induced cataract in excised bovine eye lenses, and aging effects in excised human eye lenses. The device can be easily incorporated into a slit-lamp apparatus (ophthalmoscope) for complete eye diagnostics.

Infrared (IR) fiber optic radiometry of thermal surfaces offers several advantages over refractive optics radiometry. It does not need a direct line of sight to the measured thermal surface and combines high capability of monitoring small areas with high efficiency. These advantages of IR fibers are important in the control of nonuniform temperature distributions, in which the temperature of closely situated points differs considerably and a high spatial resolution is necessary. The theoretical and experimental transforming functions of the sensor during scanning of an area with a nonuniform temperature distribution were obtained and their dependence on the spacial location of the fiber and type of temperature distribution were analyzed. Parameters such as accuracy and precision were determined. The results suggest that IR fiber radiometric thermometry may be useful in medical applications such as laser surgery, hyperthermia, and hypothermia.

This paper presents preliminary sensor designs and reports initial test results for the implementation of real-time simultaneous oxygen saturation and blood pressure measurement. Closed-lumen catheter-type in-vivo sensing schemes incorporate a proven, well-established distal-tipped fiberoptic pressure sensor. Spectroscopic oxygen saturation determination is multiplexed with pressure sensing in a shared transceiver approach, and oxygen saturation sensitivity may be enhanced via synchronous pressure detection. Operating principles for pressure and oxygen saturation measurement are described, and various catheter side-port oxygen saturation sensor designs are explored. Experimental methods and results are presented, and sensor sensitivity is examined.

The problem of entero-gastric reflux is clinically relevant and none of the presently available techniques for its detection is satisfactory. Therefore, a portable fiber optic system for ambulatory reflux assessment has been conceived and developed. Two light emitting diodes (for signal and reference) and a suitable electronic circuit for signal processing are associated with a plastic fiber optic bundle. The working principle of the system is based on the characteristic absorption of bilirubin and bile around 450 nm. In-vitro results have shown good accuracy, linearity, and stability in the measurements of the fiber optic system. In-vivo results are promising.

The authors designed a number of sensors for different physical value measurements, based on a laser autogenerator with fiber-optical line of delay. Some influence upon the sensitive element joined into the gap of FOLD results in corresponding change of the generating frequency. This autogenerator property is used for measurements of such values as: displacement, vibrations, temperature, pressure, and refractive index. Contact and some remote sensors are possible.

In recent years, progress has been made in understanding the theory of laser absorption and emission. In addition, advancements of electronic components and fiber optics permitted construction of useful medical instruments. Such a fiber optic sensor (FOS) has been developed at the authors' institution. The technique is a method of major importance in the areas of geochemistry, biomedical engineering, and blood gas analysis. According to the type of variable to be measured, two wide classes can be defined; Biomedical (for oxygen saturation, pH, pO₂, and pCO₂ measurement) and physical (for pressure, temperature, and blood velocity). In general, biomedical sensors use an appropriate reversible reagent fixed at a fiber end, allowing spectrophotometric or fluorimetric analysis by modulation. Sensors for measuring blood velocity and flow are mainly applied in the cardiovascular clinic and in pathophysiology. Velocity in vessels can be measured using the laser Doppler method based on the frequency shift of light scattered by the moving red cells (erythrocytes, diameter 7/micrometers ). The Doppler-shifted backscattered light is partially collected by the same fiber tip end transmitted back to its entrance. There was shown to be a highly sensitive and linear relationship between the results obtained with this method and the blood gas analyzer.

Chemical sensing with fluorescent probes is usually accomplished by measuring the fluorescence intensity or using wavelength-ratiometric probes. Measurement of fluorescence decay times, rather than intensities, is advantageous because the decay times are largely independent of intensity changes due to light losses, photobleaching, or probe washout. Present technology allows measurement of nanosecond decay times by the phase-modulation method with inexpensive and robust instrumentation. This report describes the use of phase- modulation fluorometry for lifetime-based sensing of O₂, pH, Ca²⁺ and K⁺.

Our group has previously described large chelation-enhanced fluorescence (CHEF) effects upon the binding of metal ions, phosphates, and carboxylates to conjugate probes, providing large, readily measurable signals to these molecular recognition events. In understanding the structural requirements for CHEF, it is now possible to use the vast body of information on selective binding by azacrowns and cryptands in the synthesis of selective fluorescence probes. For example, a conjugate probe that allows for the selective, simultaneous assay of Zn(II) and Cd(II) ions has been synthesized and is described. In the homologous series of anthrylazamacrocycles that demonstrate chelation-enhanced fluorescence (CHEF) upon Zn(II) or Cd(II) binding in water, the pentacyclen derivative uniquely complexes Cd(II) with perturbation of the emission spectrum. The binding of anions such as phosphate and citrate give rise to fluorescence enhancements as large as six-fold; an observed pH dependence on the magnitude of fluorescence enhancements upon phosphate binding points to intracomplex protonation of the benzylic nitrogen by the HPO₄^2- ion as the origin of this CHEF effect. Anthrylpolyamine conjugate probes yield large (up to 80-fold) changes in fluorescence upon binding to biological polyanions (e.g., DNA, heparin, and polyglutamate) at 1 M concentrations. These fluorescence changes have been used as the basis for a fluorometric assay of heparinase activity; the enzymatic hydrolysis of ATP can also be monitored conveniently using anthrylpolyamine fluoroionophores.

The pH of interstitial fluid of malignant tumors tends to be lower than that of normal tissue and depressed by glucose administration. This study aimed to evaluate the effectiveness of dual-wavelength ratio fluorometry using a pH-dependent indicator (5,6-carboxyfluorescein: 5,6-CF) for the characterization of normal and tumoral areas in vivo. 5,6-CF has two main characteristics: it has two wavelengths of maximum absorbance (465 and 490 nm) and its fluorescence emission (maximum at 515 nm) increases as a function of pH in the physiological 6 - 7.4 pH range. The experimental study was performed on 31 CDF mice bearing lymphoid leukaemia P388 grafted subcutaneously. The tissular pH values were evaluated from the ratio of the fluorescence intensities (I490/I465) on the basis of a calibration curve linking pH measurements performed intratissularly with a microelectrode and fluorescence intensities ratio values. The fluorescence intensity reached its maximum value at 60 min after 5,6-CF and glucose administration, followed by a plateau (90 min). The ratios remain constant at 1.79 +/- 0.06 for normal tissue and 1.61 +/- 0.07 (without glucose administration) for tumoral tissue. The tumoral tissue ratios decrease down to 1.35 +/- 0.04 after 6 g/kg glucose administration. These results were correlated to the pH measurements in accordance to the calibration curve. This study validates the relevance of dual-wavelength fluorometry using a pH-dependent indicator to characterize in-vivo normal and tumoral tissues after glucose administration.

A fiber optic sensor for halides was constructed based on the collisional quenching of quinolinium chromophores bound to porous glass beads. The halide indicators N-(4- aminobutyl)-6-methoxyquinolinium (ABQ) and N-(7-carboxyheptyl)-6-methoxyquinolinium (CHQ) were covalently attached to surface derivatized glass beads. The labeled beads were fluorescent with excitation and emission maxima at 350 and 450 nm, respectively. Halide sensitivity and kinetics were examined by epifluorescence microscopy of immobilized beads. The CHQ labeled beads had a higher sensitivity to halides than the ABQ beads. CHQ bead fluorescence was quenched at 100 mM halide concentration by 40% for Cl^-, 55% for Br^-, 80% for SCN^-, and 85% for I^-. The 10 - 90% rise time for a step change in chloride concentration was approximately 4 s. Similar results were obtained when the beads were cemented to the tip of a single quartz fiberoptic. Time-resolved microfluorimetry studies showed that nanosecond fluorescence lifetimes decreased with halide concentration.

Optical fiber biosensors based on fluorescence assays have several distinct advantages when measuring biological analytes such as metabolites, cofactors, toxins, etc. Not only are optical signals immune to electronic interferences, but the polychromatic nature of most fluorochemical assays provides more potentially useful data about the system being studied. One of the most common difficulties normally encountered with optical biosensors is the inability to routinely recalibrate the optical and electronic components of the system throughout the life of the sensor. With this in mind, we present an optical fiber assay system for glucose based on a homogeneous singlet/singlet energy transfer assay along with the electronic instrumentation built to support the sensor system. In the sensor probe, glucose concentrations are indirectly measured from the level of fluorescence quenching caused by the homogeneous competition assay between TRITC labeled concanavalin A (receptor) and FITC labeled Dextran (ligand). The FITC signal is used to indicate glucose concentrations and the TRITC signal is used for internal calibration. Data is also presented on a protein derivatization procedure that was used to prevent aggregation of the receptor protein in solution. Also, a molecular model is described for the singlet/singlet energy transfer interactions that can occur in a model system composed of a monovalent ligand (FITC labeled papain) and a monovalent receptor (TRITC labeled concanavalin A).

This paper describes the development of a novel biosensor system for competitive binding immunofluorescence assay. The system is based on a specially designed fluorescence microscope combined with optical fibers. Protein A is covalently immobilized on the activated sensing tip surface of a quartz fiber or on the activated surface of glass cover slips. The surface is then incubated with rabbit IgG to reach a reasonable binding density and homogeneous distribution for adequate detectability. The sensor is exposed to FITC labeled and unlabeled goat anti-rabbit IgG. Fluorescence of sensor bound analyte, excited by an Argon ion laser 476 nm line and emitted from the sensing tip or sensing spot on the sample slips, is coupled back into the fiber and detected by the highly sensitive sensing system. The fluorescence signal is inversely proportional to the concentration of unlabeled anti-rabbit IgG in the sample. Sensor's limitation of detection has been lowered to a concentration level of 10^-10 M and 10 - 20 (mu) l sample volume. Typically, 6 f mol of unlabeled antibody in a 15 (mu) l sample volume and with a 10 - 20 min incubation period, could be measured.

A key factor in the analysis of evanescently coupled optical sensors, such as the planar waveguide immunosensor analyzed here, is the efficiency of coupling between the optical waveguide modes and the fluorescent sources located on the surface of the waveguide. This is an important parameter in determining the sensor's sensitivity to the analyte. We calculate this efficiency for several different sensor configurations using the finite-difference time-domain numerical technique, and find that the efficiency of one-way coupling can vary widely depending upon the fluorescent source polarization, phase, and distance from the surface, as well as the waveguide mode number and thickness. In particular, we find that when the layer containing the fluorescent molecules is uniform in refractive index, the coupling efficiency is larger than when the local environment possesses an irregular index.

A novel approach to the production of biospecific, optical-fiber sensors is described which is based on a dynamic modification procedure. The fiber surface is first modified with a hydrophobic layer, and the specified agent is then associated with this surface through either its inherent or designed hydrophobicity. With this procedure, riboflavin, modified at the N-3 position with a C-8 moiety, is dynamically associated with a C-18 modified optical fiber surface. The dynamically modified sensor responds to riboflavin binding protein which quenches riboflavin fluorescence upon binding. By washing in an appropriate solvent, the modified riboflavin can be removed, and the sensing surface can be reproducibly renewed.

In this work a spectroscopic method to measure the degradation degree of paraffinic mineral oils is proposed. The behavior of some molecules (indoles) in solution versus their surrounding environment (solvent polarity and viscosity) is the basis of the method. Experimental results obtained for VESTAN oil, confirming this theoretical scheme, are presented.

In order to further understand signal variations observed on magnetic resonance imaging scans of interstitial laser heating, a commercial multichannel fluoroptic thermometer, equipped with fiber optic sensors, was employed in conjunction with the laser/MRI phototherapy system. Three calibrated fiber optic sensors of the thermometer were used to measure temperature changes in ex-vivo sheep's brain at various distances directly across from the beam of a Nd:YAG laser emitted from a bare fiber. Laser was operated at 5 W for 220 sec. Temperature was measured every 10 seconds and MR images were acquired during and after laser irradiation until temperature in all probes returned to the equilibrium level of prelaser irradiation. Image contrast analysis of the heated region showed that MRI signal variations, during heating and cooling periods, correlated well with the changes in temperature. It is concluded that direct thermometry of MRI-monitored laser application will aid in understanding the effects of high focal heating on the MRI signal.

New fluorescence-based disposable probes have been developed for monitoring of pH, carbon dioxide, and oxygen in extracorporeal loops and radial arteries. These low cost, small diameter sensors fit through most standard female Luer Lock fittings. The portable monitor and its interface to the probes are described. In-vitro performance of the sensor systems versus tonometered bovine blood and blood analyzer calibration ampules are shown, along with a brief discussion of clinical applications of these devices.

Laboratory bench testing of optical blood gas sensors is insufficient to completely predict capabilities. Sensor testing in animals offers advantages of known physiologic and regulatory mechanisms of hemodynamics to better predict sensor performance. The domestic rabbit, Oryctalogis Cuniculus, a lagomorph of the family Leporidae was used for sensor evaluation. The rabbits are ventilated and blood gases modulated by variations in FIO² and rate adjustments. Twenty gauge catheters are placed in the dorsal aorta, cartoid, and femoral arteries. Pressures are monitored via transducers on the arterial lines. The optical blood gas sensors are fitted within the catheters and blood samples are collected over them for bench analysis. Sensors are on 125 micrometers glass optic fibers. Proprietary prepolymers are applied on the fiber tips through in fiber photopolymerization. These sensors are then calibrated in tonometered water and blood. Sensor monitoring is accomplished through OSR microfluorimetry systems. We have used this model in 26 studies over the past six months evaluating over fifty blood gas sensors. These studies have lasted from six to twenty-four hours. Our correlation of sensor readings to assayed blood samples is r² equals .97 for pH values of 6.80 - 7.70, r² equals .94 for PCO₂ values of 10 - 175 mmHg and r² equals .94 for PO₂ values of 10 - 350 mmHg.