Dry Rolling Seal Spirometers

Another type of volume-displacement spirometer is the dry rolling seal spirometer. A typical unit consists of a lightweight piston mounted horizontally in a cylinder. A rod that rests on frictionless bearings supports the piston (Figure 11-2, A and B). The piston is coupled to the cylinder wall by a flexible plastic seal. The seal rolls on itself rather than sliding as the piston moves. A similar type of rolling seal may also be used with a vertically mounted, lightweight piston that rises and falls with breathing (as in the dry-seal Stead-Wells described in the preceding section). The maximum volume of the cylinder with the piston fully displaced is usually 10–12 L. The piston has a large diameter so that excursions of just a few inches are all that is necessary to record large volume changes. The piston is normally constructed of lightweight aluminum to reduce inertia. Mechanical resistance is kept to a minimum by the bearings supporting the piston rod and by the rolling seal itself.

Although they can be used with a mechanical recorder, most dry rolling seal spirometers userolling seal spirometers use linear or rotary potentiometers. The potentiometer responds to piston movement to produce DC voltage outputs for volume and flow. For example, a 10 volt (V) potentiometer attached to a 10-L spirometer may produce an output of 1 V/L. On a separate channel, a flow of 1 L/sec may produce an output of 1 V. Flow in this case is proportional to the speed of the moving piston. These analog outputs for volume and flow are digitized so that a computer can store and manipulate the data.

The piston of the standard dry rolling seal spirometer (Figure 11-2, B) travels horizontally, eliminating the need for counterbalancing. The vertically mounted version (see Figure 11-1, B) depends on a lightweight piston and the rolling seal to reduce resistance to breathing. Temperature corrections (from ATPS to BTPS) are made by applying a correction factor to the digital value stored in the computer. A one-way breathing circuit and CO₂ scrubber may be added so that dry rolling seal spirometers can be used for rebreathing tests in much the same way as water-seal spirometers.

To perform studies such as the open-circuit nitrogen washout test, a “dumping” mechanism is attached to the spirometer. The dumping device empties the spirometer after each breath or after a predetermined volume has been reached. Addition of an automated valve and alveolar sampling device allows the dry rolling seal spirometer to be used for single-breath diffusion studies. Dry rolling seal spirometers are typically capable of meeting the minimum standards recommended by the ATS-ERS (see Chapter 12).

Common problems encountered with dry rolling seal spirometers are sticking of the rolling seal and increased mechanical resistance in the piston-cylinder assembly. These difficulties can usually be avoided by adequate maintenance of the spirometer. As for other types of volume-displacement spirometers, simple correction of volumes from ATPS to BTPS may not completely reflect physiologic flow or volume changes. Infection control of the dry rolling seal involves disassembling the piston-cylinder. The interior of the cylinder and the face of the piston are usually wiped with a mild antibacterial solution. The rolling seal itself is also wiped with disinfectant. Alcohol or similar drying agents may cause deterioration of the seal and should not be used. The seal should be routinely checked for leaks or tears. After reassembly, the piston should be positioned at the maximum volume position. When the rolling seal is extended completely, the material of the seal is less likely to develop creases that can result in uneven movement of the piston. With the previously described reservations, bacteria filters may be used to avoid contamination of the spirometer.

Bellows-Type Spirometers

A third type of volume-displacement spirometer is the bellows or wedge bellows spirometer. Both devices consist of a collapsible bellows that folds or unfolds in response to breathing excursions. The conventional bellows design is a flexible accordion-type container. One end is stationary and the other end is displaced in proportion to the volume inspired or expired. The wedge bellows operates similarly except that it expands and contracts like a fan (Figure 11-3, A and B) One side of the bellows remains stationary; the other side moves with a pivotal motion around an axis through the fixed side. Displacement of the bellows by a volume of gas is translated either to movement of a pen on chart paper or to a potentiometer. For mechanical recording, chart paper moves at a fixed speed under the pen while a spirogram is traced. For computerized testing, displacement of the bellows is transformed into a DC voltage by a linear or rotary potentiometer. The analog signal is routed to an A/D converter and then to a computer.

The conventional and wedge bellows may be mounted either horizontally or vertically. The horizontal bellows is mounted so that the primary direction of travel is on a horizontal plane. This design minimizes the effects of gravity on bellows movement. The horizontal bellows (either conventional or wedge) with a large surface area offers little mechanical resistance. This type is normally used in conjunction with a potentiometer to produce analog volume and flow signals. Small (approximately 7–8 L), vertically mounted bellows are available and may be used for portable spirometry and bedside testing. Most of these types offer simple mechanical recording, digital data reduction, or both by means of a small, dedicated microprocessor.

Both bellows-type spirometers (see Figure 11-3, A and B) can be used to measure vital capacity and its subdivisions, as well as FVC, FEV₁, expiratory flows, and MVV. Some bellows-type spirometers, especially those that are mounted vertically, are designed to measure expiratory flows only. These types expand upward when gas is injected, then empty spontaneously under their own weight. Horizontally mounted bellows can usually be set in a mid-range to record both inspiratory and expiratory maneuvers. This allows F-V loops to be recorded. With appropriate gas analyzers and breathing circuitry, bellows systems may be used for gas dilution FRC determinations and Dlco measurements. Most bellows-type spirometers meet ATS recommendations for flow and volume accuracy.

One problem that may occur with bellows-type spirometers is inaccuracy resulting from sticking of the bellows. The folds of the bellows may adhere because of dirt, moisture, or aging of the bellows material. Some bellows-type spirometers require the bellows to be partially distended when not in use. This technique allows moisture from exhaled gas to evaporate and prevents deterioration of the bellows. Leaks may also develop in the bellows material or at the point where the bellows is mounted. Leaks can usually be detected by filling the bellows with air, plugging the breathing port, and attaching a weight or spring to pressurize the contained gas.

Infection control of bellows-type spirometers depends on the method of construction. In some instruments, the bellows can be entirely removed; in others, the interior of the bellows must be wiped clean. Many bellows are made from rubberized or plastic-based material that can be cleaned with a mild detergent and dried thoroughly before reassembly. Bacteria filters may be used to avoid contamination of the bellows, with the reservations described previously.

Volume-displacement spirometers were once the main devices used for pulmonary function testing. Many such devices are still in use, and some manufacturers continue to produce sophisticated spirometers based on the volume-displacement principle. However, use of flow-sensing spirometers has become increasingly common, both in the pulmonary function laboratory and in small, portable devices for bedside or clinic use.

Turbines

The simplest type of flow-sensing device is the turbine, or respirometer. This instrument consists of a vane connected to a series of precision gears. Gas flowing through the body of the instrument causes the vane to rotate, registering a volume (Figure 11-5). The respirometer can be used to measure slow vital capacity (VC). It can also be used for ventilation tests such as V_T and . One such device is the Wright respirometer. This respirometer can measure volumes accurately at flows between 3 and 300 L/min. At flows greater than 300 L/min (5 L/sec), the vane is subject to distortion. Because of this limitation, it should not be used to measure FVC when the patient is capable of flows greater than 300 L/min. At low flows (less than 3 L/min), inertia of the vane-gear system may underestimate volume.

A special advantage of this type of respirometer is its compact size and usefulness at the bedside. Most respirometers can register a wide range of volumes using multiple scales. The standard Wright respirometer measures 0.1–1 L on one scale and up to 100 L on another scale. Turbine devices are also widely used for measurements of bulk flow in various dry gas meters.

An adaptation of the turbine flow device includes a photo cell and light source that is interrupted by the movement of the vane or impeller (Figure 11-6, D). Rotation of the vane interrupts a light beam between its source and the photo cell. This produces a pulse, with each pulse equivalent to a fixed gas volume. The pulse count is summed to obtain the volume of gas flowing through the device. The signal produced may not be linear across a wide range of flows because of inertia or distortion of the rotating vane.

The accuracy of turbine flow devices is usually limited by the factors described. For this reason, turbine devices often do not meet the ATS-ERS minimum recommendations for spirometers. These devices may be used for monitoring or screening. Because of their simplicity and small size, several such devices are marketed for home use. This type of spirometer allows FVC, FEV₁, and peak expiratory flow (PEF) to be monitored outside of the usual clinical setting.

Infection control of turbine-type respirometers depends on their construction and intended use. Devices such as the Wright respirometer usually must be gas sterilized. Water condensation from exhaled gas can damage the vane-gear mechanisms. Some turbine spirometers use disposable impellers. This avoids cross-contamination, but accuracy may be limited by the quality of the disposable sensor.

Pressure Differential Flow Sensors

Pressure differential flow sensors are among the most common implementations of flow sensing. They consist of a tube containing a resistive element. The resistive element allows gas to flow through it but causes a pressure drop (see Figure 11-6, A). The pressure difference across the resistive element is measured by means of a sensitive pressure transducer. The transducer usually has pressure taps on either side of the element (Figure 11-7, A). The pressure differential across the resistive element is proportional to gas flow as long as flow is laminar. This flow signal is integrated to measure volume (see Figure 11-4). Turbulent gas flow upstream or downstream of the resistive element may interfere with the development of true laminar flow. Most pneumotachometers attempt to reduce turbulent flow by tapering the tubes in which the resistive elements are mounted.

Although there are many designs for resistive elements, two types are commonly used. The Fleisch-type pneumotachometer uses a bundle of capillary tubes (or similar material) as the resistive element. Laminar flow is ensured by size and arrangement of the capillary tubes. The cross-sectional area and length of the capillary tubes determines the actual resistance to flow through the Fleisch pneumotachometer. The dynamic range of the Fleisch device must be matched to the range of flows to be measured. Different sizes (i.e., resistances) of pneumotachometers may be used to accurately measure high or low flows.

The other common type of pressure differential flow sensor is the Silverman (or Lilly) type. The Silverman pneumotachometer uses one or more screens to act as a resistive element. A typical arrangement has three screens mounted parallel to one another. The middle screen acts as the resistive element with the pressure taps on either side, whereas the outer screens protect the middle screen and help ensure laminar flow. The Silverman pneumotachometer usually has a wider dynamic flow range than the Fleisch type. As a result, it is better suited for measuring widely varying flows.

Most Fleisch and Silverman pneumotachometers use a heating mechanism to warm the resistive element to 37°C or higher. Heating the resistive element prevents condensation of water vapor from exhaled gas on the element. Condensation or other debris lodging in the resistive element changes the resistance across it, thus changing its calibration. A change in the resistive element such as condensation or a hole causes a change in the pressure-flow relationship. Pneumotachometers need to be recalibrated after cleaning or similar maintenance.

Some flow-sensing spirometers use resistive elements such as porous paper, rendering the flow sensor disposable (see Figure 11-7, B). These devices usually have a single pressure tap upstream of the resistive element. Pressure measured in front of the resistive element is referenced against ambient pressure. This design requires that the flow sensor be carefully “zeroed” before making any flow measurements. These types of flow sensors may be more susceptible to moisture or debris on the resistive element causing volumes and flows to increase with each successive blow. The flow sensor will need to be replaced if the technologist notices this occurring. The accuracy of spirometers using this type of flow sensor often depends on how carefully the disposable resistive elements are manufactured. If the resistance varies widely from sensor to sensor, each unit may need to be calibrated before use to ensure accuracy. Some manufacturers calibrate their disposable sensors and then provide a calibration code with each sensor. This code is used to identify a particular sensor by the software in the spirometer. One method of identifying the correct calibration factor for individual flow sensors is to imprint the sensor with a bar code (see Figure 11-7, B). The spirometer then includes a simple bar code reader to identify the appropriate calibration factors. Some portable spirometer systems that use precalibrated flow sensors do not provide for user calibration. However, verification of accuracy (using a 3-L syringe) is usually possible, even if the manufacturer has not provided for this in the software accompanying the spirometer.

Systems that use permanent pressure-differential flow sensors usually meet or exceed the ATS-ERS minimal recommendations for spirometers. Spirometers that use disposable sensors can meet or exceed the minimal requirements, depending on the quality of the sensor and the application software responsible for signal processing. Gas composition affects the accuracy of flow measurements in pressure-differential pneumotachometers. Correction factors for gases other than air can be applied by software so that these types of flow sensors can be used for most types of pulmonary function tests.

Infection control of pressure-differential flow sensors depends on their placement in the spirometer. In open-circuit systems in which only exhaled gas is measured, only the mouthpiece needs to be changed between patients. If inspiratory and expiratory flows are measured, the flow sensor itself may need to be disinfected between patients. Disassembly and cleaning of flow sensors usually require that the spirometer be recalibrated. Disposable or single-use sensors avoid this problem. In-line bacteria filters may be used to isolate the pneumotachometer from potential contamination. The spirometer should meet all ATS-ERS requirements for range, accuracy, and flow resistance with the filter in place (see Chapter 12). (See Figure 11-8.) If a filter is used, calibration with the filter in-line is usually required. The effect of bacteria filters on spirometric measurements has been reported to cause small errors in measured flows and volumes (30-50 mLs). Use of in-line filters can be an efficient and effective component of the PF laboratory’s infection control program.

Heated-Wire Flow Sensors

Heated-wire flow sensors are a third type of flow-sensing spirometer. They are based on the cooling effect of gas flow. A heated element, usually a thin platinum wire, is situated in a laminar flow tube (see Figure 11-6, B). Gas flow past the wire causes a temperature decrease so that more current must be supplied to maintain a preset temperature. The current needed to maintain the temperature is proportional to gas flow. The heated element usually has a small mass so that slight changes in gas flow can be detected. The flow signal is integrated electronically or by software to obtain volume measurements. The heated wire is usually protected behind a screen to prevent impaction of debris on the element. Debris or moisture droplets on the element can change its thermal characteristics. Some systems use two wires (Figure 11-9). One measures gas flow, and the second serves as a reference. Most heated-wire flow sensors maintain a temperature higher than 37°C. Heating prevents condensation from expired air that might interfere with sensitivity of the element.

Most heated-wire flow sensors meet or exceed ATS-ERS recommendations for accuracy and precision. Gas composition may affect the accuracy of flow measurements. Correction factors for gases other than room air can be applied via software. This allows heated-wire devices to accurately measure gases for pulmonary function tests using helium, oxygen (O₂), and other gases. Heated-wire sensors can be used for routine pulmonary function tests, exercise testing, and metabolic studies. Infection control for heated-wire sensors is similar to that for pressure- differential devices. Disposable or single-use devices avoid cross-contamination even when the sensor is located proximal to the patient’s airway.

Pitot Tube Flow Sensors

Pitot tube flow sensors are a fourth type of flow sensor. They use the Pitot tube principle. The pressure of gas flowing against a small tube is related to the gas’s density and velocity. Flow can be measured by placing a series of small tubes in a flow sensor and connecting them to a sensitive pressure transducer (see Figure 11-6, C). The pressure signal must be linearized and integrated as described for other flow-sensing devices. In practice, two sets of Pitot tubes are mounted in the same sensor so that bidirectional flow can be measured (Figure 11-10). A wide range of flows can be accommodated by using two or more pressure transducers with different sensitivities. Because this type of flow-sensing device is affected by gas density, software correction for different gas compositions is necessary. This is accomplished by sampling the gas, analyzing O₂ and CO₂, and applying the necessary correction factors. Software corrections for test gases used for various pulmonary function tests (e.g., Dlco) can be easily applied.

Pitot tube flow sensors meet or exceed ATS recommendations for accuracy and precision. Their practical applications include routine pulmonary function tests, metabolic measurements, and exercise testing. Infection control for this type of device includes single-use, or disposable, flowmeters. If Pitot tube flow sensors are cleaned, care must be taken that disinfectant solution or rinse water is completely removed from the small tubes used to sense flow.

Ultrasonic Flow Sensors

Gas flow can be detected and measured by passing high-frequency sound waves across the stream of gas. Ultrasonic transducers on either side of the flow tube transmit sound waves alternately across the tube. By passing the sound waves at an angle to the flow of gas in two different directions, bidirectional flow can be measured (see Figure 11-6, E). The sound waves are sped up by gas flowing in one direction and slowed down by gas flowing in the opposite direction. By measuring the “transit time” of the pulses with a very accurate digital clock, flow can be integrated to measure volume. Analyzing the change in frequency of the sound waves passing through the flowing gas has the advantage of not being affected by the gas composition, temperatures, or humidity. In addition, there are no moving parts or elements to become occluded when measuring exhaled gas.

A distinct advantage of measuring gas flow by means of ultrasonic pulses is that a disposable flow tube can be inserted between the transducers, thus eliminating problems with cross-contamination between subjects. A further advantage of this design is that the disposable flow tube does not require calibration because it simply acts as a transparent barrier separating exhaled gas from the sensing transducers (Figure 11-11).

Flow Sensor Summary

Flow-sensing spirometers have some advantages over volume-displacement systems. When combined with appropriate gas analyzers and breathing circuits, flow-sensing spirometers can be used to perform lung volume determinations by the open-circuit method. Diffusing capacity can be measured with flow-sensing spirometers as well. Pressure-differential, heated-wire, and Pitot tube pneumotachometers are used to measure flow and volume in body plethysmographs, exercise testing systems, and metabolic carts. Because flow sensors require electronic circuitry to integrate flow or sum volume pulses, flow-based spirometers are microprocessor controlled. Some flow-sensing spirometers provide their analog signal (flow, volume, or both) to a strip chart or X-Y recorder. However, most flow sensors are integrated into a spirometer system and use computer-generated graphics to produce volume-time or flow-volume tracings.

Most flow-based spirometers can be easily cleaned and disinfected. Some flow sensors can be immersed in a disinfectant without disassembly. As noted, many systems use inexpensive, disposable sensors that can be discarded after one use. The use of in-line bacteria filters to prevent contamination of flow-based spirometers may result in changes in the operating characteristics of the spirometer. Any resistance to airflow through the filter will be added to the resistance of the spirometer. For this reason, the spirometer may need to be calibrated with the filter in place. Although the resistance offered by most filters is low, it may change with use. This may occur if water vapor from expired gas condenses on the filter media. Use of barrier filters does not eliminate the need for routine decontamination of spirometers.

Most types of the flow sensors produce a signal that is not linear across a wide range of flows. Some systems use two separate flow sensors (or variable orifices) to accommodate both low and high flows. Better accuracy can be obtained for flow and volume by matching the flow range of the sensor to the physiologic signal. Most flow-based spirometers “linearize” the flow signal electronically or by means of software corrections. In many systems, a simple “look-up table” is stored in the computer. The flow signal is continuously checked against the table and corrected. By combining a calibration factor (see Chapter 12) with the look-up table corrections, very accurate flow and integrated volume measurements are possible. Flows and volumes are corrected before variables such as FEV₁ are measured.

Turbine, pressure-differential, heated-wire, and Pitot tube flow-sensing spirometers are affected by the composition of the gas being measured. Changes in gas density or viscosity require correction of the transducer signal to obtain accurate flows and volumes. In most systems, these corrections are performed by computer software with a stored table. A flow-sensing spirometer may be calibrated with air but then used to measure mixtures containing helium, neon, oxygen, or other test gases. Some gases cause a linear shift in flow proportional to their concentrations. Corrections are usually made by applying a simple multiplier to the signal.

The accuracy of flow-based spirometers depends on the electronics, software, or both that process the flow signal. Pulmonary function variables measured on a time base (e.g., FEV₁ or FEV₆) require precise timing and accurate flow measurement. Detection of the start or end of the test is critical in flow-based spirometers. Timing is usually triggered by a minimum flow or pressure change. Signal integration begins when flow reaches a threshold limit, usually 0.1–0.2 L/sec. Spirometers that initiate timing in response to volume pulses usually have a similar threshold that must be achieved to begin recording. Contamination of resistive elements, thermistors, or Pitot tubes by moisture or other debris can alter the flow-sensing characteristics of the transducer and interfere with the spirometer’s ability to detect the start or end of test. Similar problems can occur when flow drops to very low levels near the end of a forced expiration, causing measurement of flow (and volume) to be terminated prematurely. As a result, the volume (e.g., FVC or VC) may be underestimated.

Problems related to electronic “drift” require flow sensors to be “zeroed” frequently. Zeroing is simply a one-point calibration in which the output of the transducer is set to zero under a condition of no flow. Many systems “zero” the flow signal immediately before a measurement. Zeroing corrects for much of the electronic drift that occurs. A true zero requires no flow through the flow sensor. Thus, the flow sensor must be held still or occluded during the zero maneuvers. Most flow-based systems use a 3-L syringe for calibration. By calibrating with a known volume signal, the accuracy of the flow sensor and the integrator can be checked with one input. Calibration and quality-control techniques for volume-displacement and flow-sensing spirometers are included in Chapter 12.

Portable (Office) Spirometers

The widespread availability of microprocessors has resulted in a large number of small, portable spirometers based on various flow-sensing principles. These spirometers can be separated into two general categories: (1) those that interface with a laptop or desktop computer and (2) stand-alone devices that incorporate a dedicated microprocessor. In both cases, these devices may be referred to as screening or office spirometers.

Many flow-sensing spirometers interface directly with personal computers (Figure 11-12, A-C), typically a laptop computer. Other spirometers use an interface card that plugs directly into a personal computer (PC). With the appropriate software installed on the PC, spirometry can be performed. Other spirometers place the necessary electronic components in the flow sensor head or in an external adapter. This implementation allows the flow sensor to be connected to a serial port or USB (universal serial bus) port, both of which are standard on most computers. The pressure transducer and electronics for flow-based spirometry can also be mounted on a removable card. These cards allow spirometry to be performed with handheld computers, laptop computers, or personal digital assistants (PDAs). Many flow-based spirometers use dedicated microprocessors (Figure 11-13). Some of these are very compact so that the entire device is not much larger than a calculator. These designs allow the units to be handheld and portable.

Many portable spirometers use disposable flow sensors. Disposable sensors can provide accurate measurements if they are manufactured according to rigid specifications. Disposable flow sensors are usually precalibrated at the time of manufacture. Some manufacturers include calibration codes (or bar codes) on the sensor to be used in conjunction with the spirometer software. Ideally, each spirometer should provide a means for calibration using a 3-L syringe. At a minimum, the software in portable spirometers should allow verification of volumes, even if precalibrated flow sensors are used.

Interfacing a flow sensor to a PC or laptop computer makes spirometry available in a variety of clinical settings. Handheld or PC-based systems provide a relatively inexpensive way to perform spirometry, before and after bronchodilator studies, and even bronchial challenges. Increasingly sophisticated software allows spirometric data to be stored, manipulated, and displayed graphically. In spite of the availability of small, accurate spirometers, spirometry is still not widely used in primary care. Because spirometry is effort dependent, poorly performed maneuvers can result in misclassification (i.e., obstruction versus restriction versus normal). The choice of inappropriate reference equations and incorrect interpretation of results further limits the usefulness of spirometry.

The National Lung Health Education Program (NLHEP) provides recommendations for office spirometers to be used in primary care settings (Box 11-1). The goal of these recommendations is to standardize spirometry to promote early detection of chronic obstructive pulmonary disease (see Figure 11-14). Office spirometers should be simple and designed to measure three important parameters: FEV₁, FEV₆, and the FEV₁/FEV₆ ratio. To provide accurate measurements, office spirometers should display automated messages describing the acceptability and repeatability of efforts. Automated interpretation of simple spirometry can be performed if test quality is acceptable and appropriate reference values are used. Display or printouts of spirograms are optional. Office spirometers should meet the accuracy recommendations of the ATS-ERS (see Chapter 12).

Pressure Plethysmographs

The pressure plethysmograph is based on an adaptation of Boyle’s law (see Evolve site http://evolve .elsevier.com/Mottram/Ruppel/). Volume changes in a sealed box are inversely related to pressure changes if temperature is constant. A sensitive pressure transducer monitors box pressure changes. Pressure change is related to volume change by calibration (see Chapter 12 for calibration techniques). Pressure changes result from compression and decompression of gas within both the patient’s chest and the box. If box temperature remains constant, each unit of pressure change equals a specific volume change. For example, a volume change of 15 mL may result in a pressure change of 1 cm H₂O. After the box has been calibrated empty, the calibration factor changes slightly when a patient enters the plethysmograph. This change is easily corrected using an estimate of the volume displaced by the patient (based on the patient’s weight).

The pressure plethysmograph must be essentially leak free. Most pressure boxes use a solenoid to vent the box and maintain thermal equilibrium. In some implementations, the vent remains open until the pressure measurement begins, so that the box is continually being vented. Making V_TG and Raw measurements with the patient panting reduces unwanted pressure changes caused by thermal drift, leaks, or background noise. Some pressure plethysmograph systems use a controlled leak to facilitate thermal equilibrium. The leak allows gas to escape as the interior of the box warms but does not interfere with high-frequency changes such as those that occur with panting. Similarly, connecting the atmospheric side of the box pressure transducer to a container within the box dampens the effects of thermal drift. Both methods reduce the effect of temperature changes within the box and maintain good frequency response. Pressure plethysmographs are best suited to maneuvers that measure small volume changes (i.e., 100 mL or less). Measurements of VC or FVC can usually be made only with the door open or the box adequately vented to the atmosphere.

Flow Plethysmographs

The flow plethysmograph uses a flow transducer in the box wall to measure volume changes in the box. Gas in the box is compressed or decompressed, causing flow through the opening in the box wall. Flow through the wall is integrated, corrections are applied, and volume change is recorded as the sum of the volume passing through the wall and the volume compressed. In one implementation, the patient breathes through a pneumotachometer connected to the room (transmural breathing). The transmural pneumotachometer allows larger gas volumes (i.e., the VC or flow-volume curves) to be measured while the patient is enclosed in the plethysmograph. The transmural flow is redirected to the plethysmograph for Raw measurements so that the ratio of flow to box volume can be plotted. For V_TG measurements, the flow transducer in the plethysmograph wall is occluded so that the device works like a pressure box. The flow-type plethysmograph requires that the pressure, volume, and flow signals be measured in phase. Although thermal changes must be accounted for, the flow plethysmograph does not need to be rigorously airtight. The flow box’s primary advantage is the ability to measure flows at absolute lung volumes (i.e., corrected for gas compression).

In both types of plethysmograph, a flow sensor (i.e., pneumotachometer) is needed to measure airflow at the mouth (Figure 11-17). Flow measurement is required to compute Raw. The flow signal is also used to determine end-expiration for shutter closure in V_TG measurements. The pneumotachometer must be linear across the range of flows encountered in spontaneous breathing and panting (±2 L/sec) and should meet all ATS-ERS requirements for spirometers (e.g., range, accuracy, frequency response) for measurement of VC and/or FVC. Heated Fleisch or Silverman types of pressure-differential pneumotachometers are often used in the plethysmograph. Pitot tube or heated-wire flow transducers can also be used.

A mouth pressure transducer is normally coupled to a shutter mechanism. The shutter can be an electric solenoid, a scissors-type valve, or a balloon valve. The mouth pressure transducer records pressures in the range of −20 to +20 cm H₂O when the airway is occluded but should be able to measure pressures of more than ±50 cm H₂O, with a flat frequency response greater than 8 Hz. Some systems require the technologist to close the shutter by remote control at end-expiration. This may be accomplished by observing the tidal breathing maneuver on a display and actuating the shutter at end-expiration. However, most systems automatically close the shutter at a preselected point in the breathing cycle. The technologist initiates a sequence in which the computer monitors flow and closes the shutter when expiratory flow becomes zero. Automated shutters allow the airway to be occluded for a fixed length of time or for a specified number of panting breaths.

Plethysmograph or box pressure is recorded by a sensitive pressure transducer connected to the box chamber. This transducer needs to be able to accurately measure pressure changes as small as ±0.2 cm H₂O with a flat frequency response that accommodates the panting rates that are commonly encountered (e.g., >8 Hz). The box pressure transducer should have a range that can accommodate the pressures typically encountered because of thermal drift, tidal breathing, and so forth. If a slow leak is incorporated into the box to promote thermal stability, it should have a time constant of approximately 10 seconds.

Recording of plethysmographic maneuvers is usually performed by a computer. Breathing efforts are displayed in real time, allowing the technologist to elicit proper maneuvers from the patient. The real-time display assists in ensuring that panting maneuvers are performed correctly; some systems display prompts or flags so that the patient can be coached to pant at the correct frequency. The computer then stores the data and performs the necessary calculations to compute thoracic gas volume and airway resistance. Because the computer can track volume changes in the body box, V_TG and airway resistance are sometimes measured from the same maneuver. The patient breathes normally to establish the end-expiratory level, the shutter is closed and then the patient pants. When the pre-set number of seconds has elapsed, the shutter opens (see Chapter 4). The volume change between the established end-expiratory level and the point at which the shutter was closed is then used to “correct” the measured V_TG so that it equals the patient’s FRC. If V_TG is measured at lung volumes other than FRC, the alveolar pressure should also be corrected for any difference from ambient pressure (i.e., P_B).

Computerized plethysmographs compute a “best-fit” line to determine the slope (i.e., tangent) for both the open-shutter (flow versus box pressure) and closed-shutter (mouth pressure versus box pressure) panting maneuvers. The technologist can also manipulate the tangent by means of the computer keyboard or mouse. This allows some degree of correction for efforts in which the patient panted incorrectly or noise was introduced into the recorded signal. Plethysmography software provides lung volume and airway resistance data immediately after completion of the maneuver. This aids in selecting acceptable maneuvers to report. The test can also be repeated as required when questionable values are obtained. Using computer-displayed panting frequency, the technologist can coach the patient to maintain a desired rate.

Most plethysmographs include the necessary signal-generating devices to perform physical calibration (see Chapter 12). A pressure manometer (or fluid-filled U-tube) may be mounted on the box for calibration of the mouth pressure transducer. Some systems provide a fixed pressure signal for mouth pressure calibration. A small syringe (30–50 mL) driven by an electric motor allows calibration of the box pressure transducer. The motorized syringe usually produces a sine-wave flow with frequency that can be varied. This allows checking of box pressure calibration at various frequencies. Most computerized plethysmographs use a standard 3-L syringe to calibrate the pneumotachometer or flow sensor. However, a flow generator and rotameter (i.e., a flowmeter) may be included for flow calibration. Computerized plethysmograph systems provide automated calibration of transducers. The output of the transducer (i.e., its amplified signal) is measured, and the computer generates a software correction factor. This correction is then applied to every measurement made with the transducer. Some manufacturers also supply quality control (QC) devices such as an isothermal lung analog (see Chapter 12). These devices provide QC to verify calibration of transducers and appropriateness of software correction factors.

The ease with which a patient can enter the plethysmograph and perform the required maneuvers is an important feature. Some patients may experience claustrophobia when inside the plethysmograph. Older boxes used a plywood cabinet to provide the necessary rigidity so that pressure changes were not attenuated. Boxes made of durable plastics are largely transparent and less confining for the patient (see Figure 11-16) while maintaining the necessary rigidity. Most plethysmographs contain 500–700 L of volume and can accommodate even large patients. Careful design allows the patient to easily enter the box. Some plethysmographs are large enough to accommodate patients in wheelchairs. Others use a clamshell design so that the patient may be seated and the box closed around him or her. Most plethysmographs provide an internal switch or mechanism that allows patients to open the box from inside if they become uncomfortable.

Equally important is a communication system that allows both voice and visual contact with the patient. Panting against a closed shutter may be difficult for some individuals, and continuous coaching is often necessary to elicit valid maneuvers. An intercom system that provides continuous two-way communication is essential.

Free Breathing and Demand Valves

The simplest type of valve allows the patient to be switched from breathing room air to breathing gas contained in a spirometer or special breathing circuit. Free breathing valves are often used in both open-circuit and closed-circuit FRC determinations. The free breathing valve is designed so that the patient can be “switched in” to the system either manually or by computer controls at a specific point in the breathing cycle.

The typical free breathing valve consists of a body with two or more ports. A drum in the valve body rotates to connect different combinations of ports. Because these valves are used mainly for tidal breathing or slow vital capacity (SVC) maneuvers, resistance to flow is usually not critical. Most have ports with diameters of 1.5–3 cm. For studies involving gas analysis, such as FRC determination, the valve must be free of leaks. The dead space of valves used in measurements of FRC should be less than 100 mL.

Some systems use a “breathing manifold” that consists of multiple ports and valves all connected to the patient mouthpiece. The different ports allow inspired or expired gas to be directed to the spirometer or gas-sampling devices. The valves may be electric or gas-powered solenoids, balloon valves that inflate with compressed air, or scissors-type valves that pinch flexible tubing to control flow. Under computer control, different combinations of valves and ports are opened and closed. This type of manifold permits spirometry, dilutional lung volumes, and Dlco tests to be performed with the same breathing circuit. A similar manifold is used in many body plethysmograph systems to permit measurement of V_TG and Raw along with Dlco.

Infection control of free breathing valves and multiple-port manifolds involves disassembly, cleaning, and disinfection or sterilization. Because cleaning between patients may not be practical, in-line filters may be used to prevent contamination of these devices.

Demand valves (also sometimes referred to as demand-flow regulators) are used in many circuits in which the patient inspires a test gas (e.g., Dlco, FRCn₂). The primary considerations for demand valves are the pressure required to trigger gas flow and the adequacy of flow once the valve opens. Most demand valves consist of a valve body that contains a sensitive diaphragm. The diaphragm moves in response to the patient’s inspiratory effort, opening the valve and allowing gas to flow from a high-pressure source. Maximal flow is controlled by the valve and usually depends on the driving pressure (typically 20–50 psi). Demand valves are often adjustable, allowing the sensitivity to be set so that minimal pressure is needed to trigger gas flow. Problems typically encountered with demand valves include inadequate source pressure (turned off or not connected) and sticking or incorrectly adjusted diaphragms. Demand valves used in Dlco circuits should be able to deliver 6 L/sec flow with less than 10 cm H₂O pressure.

Directional Valves

Directional (one-way and two-way) valves are used in many types of breathing circuits. The simplest type consists of a flap or diaphragm that opens in only one direction. The valve is then mounted in a rigid tube that can be inserted in a breathing circuit. Because gas is permitted to flow in one direction only, these valves are called one-way valves.

Another common valve design is that used to separate inspired from expired gas, often called a two-way nonrebreathing valve. This type of directional valve consists of a T-shaped body with three ports and two separate diaphragms (Figure 11-18). The diaphragms allow gas to flow in one direction only. The patient connection is between the diaphragms, effectively separating inspired from expired gas. Two-way nonrebreathing valves are used in exercise testing, metabolic studies, or any procedure requiring collection, measurement, or analysis of exhaled gas. The valve body may contain a tap for the connection of gas sample tubing. This tap is typically placed between the diaphragms so that both inspired and expired gas can be sampled.

Two factors must be considered in the selection of appropriate directional valves: dead space volume and flow resistance. For one-way valves, only flow resistance is a concern. In two-way nonrebreathing valves, dead space is the volume contained between the two diaphragms along with the volume of any connectors (e.g., a mouthpiece). Most manufacturers supply information about the dead space of individual valves. Sometimes the dead space value is printed on the valve body. Unknown dead space can be determined by blocking two of the three ports and measuring the water volume required to fill the dead space portion of the valve.

Low dead space valves (<50 mL) may be required for children or if the patient already has increased dead space, particularly if only tidal breathing is being assessed. Valves with large-bore ports and low-resistance diaphragms usually have larger dead space volumes. For exercise testing, large-bore nonrebreathing valves are used to minimize resistance at high flows. These valves may have a dead space volume of 100–200 mL. Selection of an appropriate-sized nonrebreathing valve should be based on the maximal flow anticipated during the test. For example, a maximal exercise test for a healthy adult patient may include flows greater than 100 L/min. A large-bore valve would be selected to accommodate the high flow. Valve dead space would be less of a concern because large tidal volumes are necessary to generate the increased flow. Mechanical (i.e., valve) dead space must be accurately determined for use in calculations involving gas analysis, such as physiologic dead space measurements.

Low resistance to flow is also a critical characteristic of one-way and two-way valves. Resistance to flow through most valves is nonlinear and depends on the cross-sectional area of valve leaflets or diaphragms (Figure 11-19). Resistance is usually not critical for tests in which flows of less than 1 L/sec occur. Small-bore (15–22 mm) directional valves can be selected based on an appropriate dead space volume. Most small-bore nonrebreathing valves have a resistance in the range of 1–2 cm H₂O/L/sec at flows of up to 1 L/sec (60 L/min). However, if the patient breathes through the valve for long intervals, even a small resistance may result in respiratory muscle fatigue and changes in the ventilatory pattern. Applications such as exercise testing typically involve increased flows. Large-bore, two-way valves are indicated when flows greater than 1 L/sec (60 L/min) can be expected to develop. Pressures less than 3 cm H₂O can be maintained even at flows of 5 L/sec (300 L/min) with large-bore valves. Saliva and moisture may collect in valves during prolonged tests (e.g., exercise or eucapnic hyperventilation). Valves with “saliva traps” may be needed for these procedures. If a valve is used in a spirometry circuit, it must have very low resistance to meet the ATS-ERS recommendation of less than 1.5 cm H₂O/L/sec at flows of 12 L/sec. Valves used in a Dlco circuit should produce a total resistance of less than 1.5 cm H₂O/L/sec at flows of 6 L/sec.

Any valve can cause increased resistance if not properly maintained. Rubber, plastic, or silicon leaflets and diaphragms can stick or become rigid with age. Valves should be disassembled and cleaned after each use, according to the manufacturer’s directions, and allowed to dry thoroughly before reassembly. Care must be taken when reassembling valves to ensure that all diaphragms are oriented properly. Valves should be visually inspected to make sure that diaphragms are mounted correctly and that they open and close properly before being used.

Gas-Sampling Valves

Specialized valves (valve manifolds) may be used to sample gas during tests such as the single-breath Dlco. These valves open and close during the breathing maneuver to allow the patient to inspire test gas and expire to a gas collection device. The same mechanisms are often used to occlude the airway near the patient’s mouth for breath-holding maneuvers or to measure mouth pressure. Many of these valves use electrically or pneumatically powered solenoids to direct flow to a spirometer or sample collection device. Other systems use scissors-type valves to pinch compressible tubing. The primary concern with gas-sampling valves is smooth operation with appropriate direction of the gas to be sampled. Electrically activated solenoids may deteriorate with age, particularly if exposed to high humidity conditions (e.g., expired air). Replacement of O-rings or similar types of seals may be necessary to ensure uncontaminated gas samples. Some sampling valves use balloons that inflate to block or direct the flow of gas. These balloons require periodic replacement because small leaks can prevent the balloon from occluding flow. When this occurs, gas may not be directed to the appropriate device, or the gas itself may be contaminated. Scissors-type valves offer an advantage in that the compressible tubing can be changed easily between patients.

Infection control for sampling valves usually requires disassembly and cleaning. Some complex valve manifolds may be difficult or impossible to disassemble. In such devices, an in-line filter may be needed to avoid cross-contamination.

Oxygen Analyzers

Oxygen analysis can be performed by several different methods. Table 11-2 lists some types of O₂ analyzers available. Two types are used for rapid analysis of O₂ such as breath-by-breath exercise tests: the polarographic electrode and zirconium fuel cell. The other O₂ analyzers listed are used for specialized applications, including patient monitoring.

Infrared Absorption (CO₂, CO) Analyzers

Several types of respiratory gas analyzers are based on absorption of infrared radiation to measure gas concentrations. Infrared absorption is used in CO analyzers for Dlco tests. Infrared CO₂ analyzers are used for exercise testing, metabolic studies, and bedside monitoring (capnography) in critical care (Figure 11-20, A-C).

Certain gases (e.g., CO₂ and CO) absorb infrared radiation. A common type of infrared analyzer uses two beams of infrared radiation directed through parallel cells. One cell contains sample gas, while the other contains a reference gas. The two beams converge on a single infrared detector (Figure 11-21). A small motor rotates an interrupter or “chopper” between the infrared source and the cells. The chopper blades alternately interrupt the infrared radiation passing through the sample and reference cells. If the sample and reference gases have the same concentration, the radiation reaching the detector is constant. However, when a sample with a different gas concentration is introduced, the radiation reaching the detector varies in a rhythmic fashion. This causes a vibration in the detector that is translated into a pulsatile signal proportional to the difference between the two beams.

Infrared analyzers can measure small changes in gas concentrations such as the difference between inspired and expired CO in Dlco tests. Infrared analyzers respond rapidly once the gas has been transported to the measuring chamber. Gas can be sampled either continuously or discretely with infrared analyzers. For continuous sampling, the gas flow must be constant. The analyzer must be calibrated using the same flow at which measurements are made. Pump settings and the sample line should not be altered after calibration. Water condensation or other debris in the sample line can significantly alter the flow and affect the accuracy of the measurement. Water vapor in the sample will dilute the gas being analyzed. Water vapor can be removed if the response time is not critical. For rapid response times, as required for breath-by-breath analysis, the effects of water vapor can be corrected mathematically by assuming that expired gas is fully saturated, or by using semipermeable tubing that equilibrates the sample with ambient saturation (see the section on gas-conditioning devices).

Infrared analyzers used to measure CO (and, in some cases, tracer gas) during the Dlco test need to have a linear output because the calculation of Dlco commonly uses the ratio of CO to tracer gas (see Chapter 3). The analyzer should display nonlinearity of 0.5% or less of full scale across the range of gas concentrations typically used (approximately 0.3% for CO). Although the output of the infrared detector may be nonlinear, it can be easily corrected electronically or by means of a look-up table in software. For measuring CO and tracer gases during Dlco tests, the analyzer needs to be stable (i.e., no drift) for at least as long as the test may last (typically 30 seconds). Because water vapor and CO₂ affect the measurement of CO, these gases need to be removed (“scrubbed”) before reaching the infrared detector.

Common problems occurring with infrared analyzers involve the chopper motor, sample cell, and infrared detector. Motors turning the chopper blades may wear out or work intermittently. Some analyzers use a nonmechanical means of alternating the infrared beams, thus eliminating the problem. The sample cell can easily become contaminated. Water or other debris can be aspirated into the analyzer and contaminate the cell “window,” interfering with transmission of the infrared beam. Infrared detector cells degrade over time and become less sensitive. Both contamination of the sample cell and detector aging can alter response time or make the analyzer impossible to calibrate.

Emission Spectroscopy Analyzers

The single-breath and multiple-breath N₂-washout tests (open-circuit FRC determination) use N₂ analysis. The Geissler tube ionizer is an N₂ analyzer based on the principle of emission spectroscopy (Figure 11-22). This instrument consists of an enclosed ionization chamber that contains two electrodes and a photo cell. A vacuum pump creates a constant low pressure in the ionization chamber by bleeding gas through a needle valve. The needle valve draws gas to be sampled from the breathing circuit. When current is supplied to the electrodes, the N₂ between them is ionized and emits light. After being filtered, this light is monitored by a photo detector. The intensity of the light is directly proportional to the concentration of N₂ in the sample. The current, distance between electrodes and gas pressure must remain constant. The photo detector converts the light signal into a DC voltage. This analog signal is then amplified, linearized, and directed to an appropriate meter or analog-to-digital converter. The Geissler tube ionizer allows continuous and rapid analysis of N₂ with response times typically less than 100 msec.

Emission spectroscopy analyzers used for measuring FRC by N₂ washout should have a range of 0%–80% and should be linear (≤0.2% error) across this range. N₂ analyzers need to have a resolution of 0.01% or less with a 95% response time of 60 msec or less. Because these analyzers measure rapidly changing concentrations of N₂, correction for phase delay in the N₂ signal may be required.

Analyzers using emission spectroscopy usually require a vacuum pump. Vacuum pressure must be maintained at a stable level to ensure accuracy and linearity. Leaks in the seals around the needle valve or in the pump itself may occur. Inability to zero or span the analyzer (i.e., adjust the gain) often indicates a leak or faulty vacuum source. The photo detector, ionizing electrodes, and light filter may all degrade over time. Periodic linearity checks allow adjustment for small changes in these components.

Thermal Conductivity Analyzers

Measurement of FRC by the closed-circuit method requires He analysis. Some Dlco systems also analyze helium as the tracer gas. Thermal conductivity analyzers measure gas concentrations in a sample by detecting the rate at which different gases conduct heat. Heated wires or beads (thermistors) are exposed to the gas sample. The concentration of a specific gas can be detected by measuring the change in electrical resistance of the thermistors. Two glass-coated thermistors serve as sensing elements connected by a Wheatstone bridge circuit (Figure 11-23). Thermistors change temperature and electrical resistance as a function of the molecular weight of the gases surrounding them. One thermistor serves as a reference. A difference in the concentration of gases between two thermistors can be detected because differences in heat conducted away alter the electrical resistance in the circuit. He analyzers use a reference cell containing no helium (He). Other gases can be analyzed by means of thermal conductivity if no interfering gases are present. Thermal conductivity analyzers are used in conjunction with gas chromatography (see the section on gas chromatography). Water vapor and CO₂ must be removed before He analysis. Thermal conductivity analyzers can be used for continuous or discrete measurements but have response times in the range of 10–20 seconds. Thermal conductivity analyzers cannot be used to detect rapid changes in gas concentration.

When a thermal conductivity analyzer is used for FRC_He measurements, a range of 0%–10% full scale is required, with a resolution of 0.01% or less over this range. The 95% response time should be 15 seconds or less for step changes in He concentration of 2%. The analyzer should show minimal drift (≤0.02%) for as long as the test may last (up to 10 minutes).

Thermal conductivity analyzers are very stable. Other than the pump used to draw gas into the analyzer, there are no moving parts. Unless the thermistor in the sampling chamber is contaminated or physically damaged, the analyzer remains accurate for an extended period. Gas pressure and temperature should be maintained at levels similar to those used during calibration in order for the analyzer to produce accurate results. Water vapor or CO₂ in the sample circuit (caused by malfunctioning absorbers) is a common cause of errors with this type of analyzer. Some He analyzers also use a water absorber in line with the reference thermistor. This allows dry room air to be used to zero the analyzer. Exhaustion of this absorber can result in calibration errors.

Gas Chromatography

Gas chromatography combines a means of separating a sample into component gases and a detector for measuring concentrations of the components. The detector is usually a thermal conductivity analyzer as previously described. Most chromatographs use the principle of column separation to segregate the component gases of the sample (Figure 11-24). The chromatograph column contains a packing material that impedes movement of gas molecules, depending on their size. The material is usually a high-surface-area inorganic or polymer packing. Some columns also use materials that combine chemically with specific gases. A combination of columns allows a wide range of gases to be analyzed with a single detector. He is used as a carrier gas because of its high thermal conductivity. The sample gas, along with the He carrier gas, is injected into the column. Component gases exit the column at varying rates and are detected by a thermal conductivity analyzer. The concentrations of each gas can be determined by comparing the output of the thermal conductivity analyzer with a known calibration gas. When He is used as the carrier gas, it cannot be used as an inert indicator for lung volume determinations or diffusing capacity measurements. Neon, which is relatively insoluble, may be substituted for He in these tests. Water vapor and CO₂ are usually removed from the sample to prevent contamination of the separator column.

Gas chromatographs are well suited to applications requiring analysis of multiple gases, such as Dlco determinations. If gas chromatography is used for Dlco measurements, the linearity of the device should be 0.5% of full scale for both CO and the tracer gas. As for infrared analyzers, gas chromatographs in Dlco systems need to be stable (i.e., no drift) for the interval between calibration and test gas analysis. Gas chromatograph response times are typically from 15–90 seconds, depending on the gas to be detected and the flow of carrier gas.

Chromatography is very accurate and is widely used for analysis of certified reference gases. Column material must be replaced when exhausted to maintain accuracy. Some chromatographs heat the column to enhance separation. Failure of the heating mechanism can lead to inaccurate analyses. Exhaustion of water or CO₂ absorbers can also cause the column to become contaminated.

Chemoluminescence Analyzers

Chemoluminescence analyzers are based on the principle that when two reactants are mixed to form an excited intermediate state, the intermediate may emit light as it decays back to a lower energy level. Chemoluminescence is routinely used to measure nitric oxide (see Fe_NO in Chapter 10) to assess airway inflammation. Ozone may be combined with NO to form NO₂ (nitrogen dioxide) in an excited state:

< ?xml:namespace prefix = "mml" />NO+O3→NO2[⋄]+O2

where:

NO = nitric oxide

O₃ = ozone

NO₂ = nitrogen dioxide

[◊] = represents decay to a lower energy state

O₂ = oxygen

The activated NO₂ luminesces (i.e., emits light) in visible and infrared wavelengths as it decays to a lower energy state. A photomultiplier or charge-coupled device (CCD) counts photons emitted at a specific wavelength, which are proportional to the amount of NO in the sample. Figure 11-25 shows an example of one commercially available chemoluminescence analyzer for measurement of NO.

Specific recommendations for chemoluminescence NO analyzers have been published by the ATS-ERS (Table 11-3). Calibration of NO analyzers is critical because of the small gas concentrations typically involved (parts per billion [ppb]). Chemoluminescence analyzers used to measure FeNO should be calibrated with gases that span the range of values encountered in clinical practice. These ranges differ depending on whether airway or nasal NO is being measured. In addition to appropriate calibration gases, a zero gas that is free of NO is also required. Ambient air can be drawn through an NO scrubber to provide the zero gas. Chemoluminescence analyzers are sensitive to changes in ambient conditions, especially temperature, and may require recalibration if conditions change. Sample inlet flow also affects the temperature of the reaction chamber and must be carefully controlled.

An electrochemical sensor has been developed, based on the amperometric technique that can measure exhaled nitric oxide. The technique uses the production of a current when a voltage potential is applied between two electrodes (basic principle used in the PO₂ electrode). The utilization of this analyzer allowed for the development of a hand-held (portable) exhaled nitric oxide device (Figure 11-26).

Gas-Conditioning Devices

Interference from water vapor or CO₂ in expired gas is common to many types of gas analyzers. These two gases are usually removed by chemical “scrubbers.”

CO₂ may be absorbed by passing the sample through granules containing either barium hydroxide (Ba[OH]₂) or sodium hydroxide (NaOH). Granules containing NaOH usually have a light brown appearance that changes to white when saturated with CO₂. Ba(OH)₂ (Baralyme) is usually supplied with an indicator (ethyl violet), which changes from white to purple when saturated with CO₂. Both NaOH and Ba(OH)₂ are mildly corrosive and may generate heat if exposed to high concentrations of CO₂. Both generate water as a product of combination with CO₂. Therefore, they should be placed upstream of any water vapor absorber used in the same circuit.

Water vapor is absorbed by passing the humidified gas over granules of anhydrous calcium sulfate (CaSO₄ [Drierite]) or silica gel. These substances are termed desiccants. CaSO₄ usually contains an indicator that changes from blue to pink when saturated with water vapor. Some analyzers use silica gel to remove water vapor.

Conditioning of gas that contains water vapor may also be accomplished with special sample tubing (Nafion). This tubing is permeable to water vapor. Sample gas passing through the tubing equilibrates its water vapor pressure with that of the surrounding atmosphere. Water vapor is not removed; it remains constant at a known level. If wet gas (e.g., expired air) is passed through the tubing, the sample falls to ambient humidity. If dry gas (e.g., calibration gas) is passed through the tubing, the sample rises to ambient humidity. This allows corrections for water vapor pressure to be accurately applied, as long as the ambient humidity level is known.

Failure to adequately remove water vapor or CO₂ from a gas sample usually results in dilution of the remaining gases. Dilution lowers the fractional concentration of the gas being analyzed. In some types of analyzers (e.g., thermal conductivity), CO₂ or water vapor may also directly alter the output of the analyzer. Chemical scrubbers or permeable tubing should always be replaced according to the manufacturer’s recommendations.

pH Electrodes

The traditional glass pH electrode contains a solution of constant pH on one side of a glass membrane. The sample to be analyzed is brought into contact with the other side of the pH-sensitive glass (Figure 11-27). The difference in pH on either side of the glass causes a potential difference, or voltage. To measure this potential, two half-cells are used: one for the constant solution and one for the sample. The constant solution half-cell (i.e., the measuring electrode) is usually a silver–silver chloride wire. The external half-cell is usually a saturated calomel (i.e., approximately 20% KCl) electrode called the reference electrode. The reference electrode makes contact with the unknown solution by means of a permeable membrane or a liquid junction. These half-cells are connected to a voltmeter calibrated in pH units. The voltage difference between the two electrodes is proportional to the pH difference of the solutions. Because the pH of one solution is constant, the developed potential is a measure of the pH of the sample. This type of analysis is thus referred to as a potentiometric because a potential is measured. Potentiometric electrodes are widely used in instruments with permanent electrodes as well as in some disposable cartridge–based systems.

pH can also be measured by an optical pH indicator. The indicator is an azo-dye substance in a cellulose membrane; the indicator exists in an acidic and basic form. Each of the two forms absorbs light in different regions of the spectrum (i.e., blue and red). The acidic form absorbs blue light while the basic form absorbs red light. Light is passed through the blood sample and analyzed at different wavelengths. Increased absorbance in the blue wavelength indicates a lower pH, while increased absorbance in the red portion of the spectrum indicates a higher pH. By analyzing the absorptions at multiple wavelengths, the actual pH can be determined. This method is used in some cartridge-based blood gas analyzers.

A third method used for measuring pH in blood gas analysis uses a fluorescent chemosensor or optode. The optode has a pH-sensitive fluorescent indicator dye that exists in two forms (protonated and deprotonated). The deprotonated form fluoresces (emits light), but the protonated form (higher H⁺ concentration) does not. By measuring the fluorescence, the pH can be measured.

Protein contamination of the pH-sensitive glass is a common problem that increases with the number of specimens analyzed. Routine cleaning with a proteolytic agent (e.g., bleach) reduces buildup of protein on the electrode tip. KCl depletion or blockage of the reference junction can also cause pH electrode malfunction. Contamination of reagents used for pH electrode calibration may also result in measurement errors. Daily (or more frequent) use of suitable quality control (QC) materials can detect these and other problems (see Chapter 12). pH measurements made with optical methods are dependent on how well the sample is mixed. Poorly mixed specimens may result in absorbances that do not accurately reflect the acid-base status of the patient. Fluorescent optodes are designed to separate the blood specimen from the sensor itself using an isolation layer to prevent contaminants from affecting the sensor.

Pco₂ Electrodes

Traditional Pco₂ electrodes (Severinghaus electrodes) measure Pco₂ potentiometrically using an adaptation of the pH electrode (Figure 11-28). A combined pH-reference electrode is placed inside of a membrane-tipped plastic jacket. The jacket is filled with a bicarbonate electrolyte. The membrane is usually Teflon or a similar material permeable to CO₂ molecules. A spacer or wick made of nylon is sometimes placed between the pH-sensitive glass and the membrane. The spacer ensures that a thin layer of bicarbonate electrolyte is in contact with the electrode. When the blood sample is introduced at the tip of the electrode, CO₂ diffuses across the membrane. CO₂ is hydrated in the electrolyte according to the following equation:

CO2+H2O↔H2CO3↔H++HCO3–

The higher the Pco₂, the more the equation is driven to the right. The change in H+ concentration is proportional to the change in Pco₂. The electrode detects the change in Pco₂ as a change in pH of the electrolyte (i.e., a potentiometric measurement). The voltage developed is exponentially related to Pco₂. A tenfold increase in Pco₂ is approximately equal to a decrease of 1 pH unit. Partial pressure of CO₂ can be determined by calibrating the pH change when the electrode is exposed to gases with known Pco₂ values.

Another method of measuring Pco₂ uses infrared absorption of dissolved CO₂ at three different wavelengths. By modulating the length of the light path through the specimen, absorption caused by factors other than the blood itself can be factored out. Once the concentration of CO₂ has been determined (solving for three different wavelengths) the Pco₂ can be calculated by dividing the concentration by the solubility factor for CO₂.

Pco₂ can also be measured with optical fluorescence. An optode similar to that used for pH measurements (see the preceding section) is covered by a membrane that is permeable to CO₂. CO₂ diffuses into the optode where it forms carbonic acid and lowers the pH. As the pH decreases (more H+), there is less fluorescence, and Pco₂ can be measured indirectly in much the same way as in the traditional electrode.

The most common problem with traditional Pco₂ electrodes is degradation or contamination of the permeable membrane. Protein or debris deposited on the membrane slows diffusion of CO₂. Equilibrium between the sample and electrode may not be achieved. Electrolyte depletion or exhaustion in the jacket around the electrode may also occur with extended use. Careful attention to shifts in electrode performance, during either calibration or control runs, can detect these common problems. Routine maintenance includes replacing the membrane and refilling the electrode with fresh electrolyte. Most manufacturers provide a kit that contains a disposable jacket with a membrane and fresh electrolyte. Analyzers that use single-use electrodes eliminate most of the listed problems related to the membrane and electrolyte in the electrode.

Analyzers that use the infrared photometric methodology typically use single-use measurement cuvettes, eliminating problems of contamination of the measuring chamber. However, photometric absorption measurements require that the specimen be well mixed. Fluorescent optode analyzers use an optical isolation layer to prevent contaminants and stray light from entering the optode. Guidelines for QC of blood gas electrodes are included in Chapter 12.

PO₂ Electrodes

The standard PO₂ electrode (Clark electrode) consists of a platinum cathode that is usually a thin wire encased in plastic or glass, together with a silver–silver chloride (Ag-AgCl) anode (see Figure 11-28.) Both anode and cathode are placed inside a plastic jacket that is tipped with a polypropylene or polyethylene membrane. This membrane is semipermeable and allows diffusion of oxygen molecules. The jacket is filled with phosphate–potassium chloride buffer. A polarizing voltage of approximately 630 mV is applied to the electrode. The cathode is slightly negative with respect to the anode. Because the electrode is polarized, it is referred to as a polarographic electrode. Oxygen is reduced (i.e., takes up electrons) at the cathode according to the following equation:

O2+2H2O+4e–→4OH–

Electrons (e in the preceding equation) are supplied by the Ag-AgCl anode. Electrons flow from the anode to the cathode with a current proportional to the number of molecules of O₂ reduced. Each O₂ molecule can take up four electrons, and the greater the number of O₂ molecules present, the greater the current. The membrane causes a diffusion limitation to the number of molecules reaching the electrode. The greater the partial pressure on the sample side of the membrane, the higher the rate of diffusion. The measurement of the current (amperometric) developed within the electrode is therefore proportional to PO₂.

Optical methods of measuring PO₂ include phosphorescence and fluorescence quenching. In each of these methods, a dye that emits light and is sensitive to the presence of O₂ is used (Figure 11-29). The higher the PO₂ is in the sensor or optode, the lower the phosphorescence or fluorescence (i.e., increased quenching of emitted light).

As with the Pco₂ electrode, contamination or degradation of the membrane alters diffusion of O₂ and can result in erratic measurements. Depletion of the phosphate buffer can also cause a change in electrode sensitivity. Most polarographic electrodes use a platinum wire of small diameter to reduce the actual consumption of O₂ at the tip of the electrode. The exposed surface of the platinum cathode gradually becomes plated with metal ions and must be periodically polished to maintain its sensitivity. Alternatively, the entire electrode can be replaced.

Because the membrane causes a diffusion limit to O₂ molecules reaching the cathode, the electrode performs differently when exposed to liquid versus gas samples. Many blood gas systems use gas to calibrate the PO₂ electrode. Noticeable differences may result when the electrode is then used to analyze the tension of O₂ dissolved in a liquid (e.g., blood). These differences are usually compensated for by correcting the PO₂ with an empirically determined gas-to-liquid factor.

Laboratory Analyzers

Although the gas-measuring (PO₂ and Pco₂) electrodes and the pH electrode system can each be used separately, all three are usually implemented together in a blood gas analyzer (Figure 11-30). The three electrodes are mounted in a single measuring chamber. This allows a small blood sample (≤200 μl) to be analyzed. Most blood gas analyzers are microprocessor controlled. Sample aspiration, rinsing, and calibration can all be done automatically with program control. Standardization of these functions, especially calibration, reduces measurement error and improves precision. The microprocessor can calculate other blood gas values derived from pH, Pco₂, and PO₂, including HCO3–, total CO₂, and base excess (BE). In addition to pH, Pco₂, and PO₂, most modern laboratory analyzers incorporate a hemoximeter to provide total Hb and its derivatives (e.g., O₂Hb, COHb). Computerized analyzers can monitor automated calibrations and electrode performance to alert the technologist of existing or impending problems. The computer can monitor the level of reagents and calibrating solutions/gases as well. Some analyzers automatically perform quality control runs (i.e., auto QC) and monitor the results.

Point-of-Care Analyzers

To provide rapid results of critical analytes (e.g., blood gases and electrolytes), portable or bedside analyzers (i.e., point-of-care [POC]) have become widely available (Figure 11-31). These devices are typically designed for use in the emergency department, critical care unit, or outpatient clinic. Most can be battery-operated, but some POC instruments require standard power. Blood gas measurement techniques differ slightly among models. Some POC blood gas analyzers use microelectrodes similar to those described previously, whereas others use spectrophotometry, infrared spectroscopy, or fluorescence quenching methods. Reagents, calibration materials, and waste containers are usually contained in disposable cartridges. Some POC systems use cartridges that allow a fixed number of analyses, whereas others use a single-patient sample chamber. Calibrations for POC systems that use multiple-specimen cartridges are usually performed in the traditional manner (see Chapter 12). These POC instruments use aqueous buffers in the single-patient chamber to perform calibration immediately before sample analysis. Single-use devices often have the sample chamber precalibrated by the manufacturer. Many POC instruments include ion-specific electrodes for analysis of electrolytes (e.g., K⁺, Na⁺, and Ca⁺) along with other metabolites and hematocrit. A few POC analyzers also incorporate hemoximetry in addition to pH, Pco₂, and PO₂.

The accuracy and precision of most POC blood gas analyzers appear comparable to that obtained with standard laboratory instruments. Routine analysis of multiple levels of quality control (QC) material is required to assess precision. However, because of the design of the sensors and the fact that reagents and reference electrodes may not be required, some cartridge-based systems require QC only when the cartridge is changed. Analysis of unknown specimens and comparison with other instruments or laboratories (proficiency testing) is required to determine accuracy.

Transcutaneous PO₂ and Pco₂ Electrodes

The transcutaneous O₂ electrode (tcPO₂) operates on a principle similar to that of the polarographic electrode. The tcPO₂ electrode typically consists of a ring-shaped silver anode heated by a coil to increase blood flow at the skin placement site. Inside the circular anode is a platinum cathode (Figure 11-32). All elements are enclosed in a plastic case. The face of the sensor is covered by a Teflon membrane. Electrolyte (KCl) is placed between the membrane and the sensor. A second layer of electrolyte and a cellophane membrane are added to form a double membrane. The current between the silver anode and platinum cathode is proportional to the PO₂ diffusing through the skin and membrane. A feedback controller keeps the temperature constant at the skin site. This also compensates for changes in capillary blood flow and stabilizes the measurement.

The gradient between tcPO₂ and Pao₂ is relatively constant in patients with normal cardiac output. In neonates there is a close correlation between transcutaneous and Pao₂. In hemodynamically stable adults, tcPO₂ is approximately 80% of Pao₂. Measurement of tcPO₂ can trend oxygenation when this gradient has been established. In patients with reduced cardiac output, the gradient between tcPO₂ and Pao₂ widens. Conditions that affect perfusion to the skin may also alter the gradient between arterial and tcPO₂.

Transcutaneous measurements of Pco₂ are possible with a sensor that uses a CO₂-permeable membrane like that in the traditional blood gas electrode. CO₂ diffuses through the skin and into the sensor. tcPCO₂ and tcPO₂ sensors typically require calibration. This is usually accomplished by attaching the sensor to a port that exposes it to a calibration gas. Several types of transcutaneous monitors combine tcPCO₂ with tcPO₂ or pulse oximetry (SpO₂).

Most transcutaneous monitors heat the skin site from 40°C–45°C. The increased temperature “arterializes” capillary blood flow. The fastest response times are usually attained at the highest temperature setting. However, this necessitates moving the electrode every 3–4 hours to prevent burns. Changing sensor sites is particularly important in neonates because of the reduced thickness of their epidermis. Periodic recalibration of the electrode is necessary even if the sensor site has not been changed. After placement of the sensor, an interval of 5–30 minutes may be required for equilibration to be reached.

Spectrophotometric Oximeters

The spectrophotometric oximeter uses light absorption to analyze saturation of hemoglobin (Hb) with O₂. The concentration of carboxyhemoglobin (COHb) or other forms of Hb (e.g., methemoglobin, sulfhemoglobin) can also be determined. This type of spectrophotometer is sometimes called a co-oximeter or hemoximeter. In addition to various forms of Hb, spectrophotometric measurement principles (absorption of light) can be used to measure pH and Pco₂, as described previously.

The hemoximeter analyzes the absorption of light in a blood sample at multiple wavelengths. At certain wavelengths, two or more forms of Hb have similar absorbances (Figure 11-33). These common wavelengths are termed isobestic points. An isobestic point for oxyhemoglobin (O₂Hb), reduced Hb (RHb), and COHb is 548 nm. At this wavelength, the absorbance of a mixture of the three pigments is directly proportional to the total concentration of Hb. An isobestic point for O₂Hb and RHb is 568 nm. The absorbance of COHb at this point is considerably higher. A change in absorbance at 568 nm compared with 548 nm indicates a change in the concentration of COHb relative to the sum of the concentrations of the other two species. The isobestic point for RHb and COHb is 578 nm, with O₂Hb absorbance being considerably greater. The difference in absorbance at 578 nm indicates the concentration of O₂Hb relative to the other two pigments. The total Hb concentration, O₂Hb, COHb, and methemoglobin (MetHb) saturation can be determined by analyzing absorbances and solving simultaneous equations.

The hemoximeter provides the true O₂Hb saturation (see the section on oxygen saturation in Chapter 6). This is particularly important if increased concentrations of COHb, MetHb, or other abnormal hemoglobins are present. Some automated blood gas analyzers calculate O₂Hb saturation. Calculated saturation is based on the measured PO₂ and pH at 37°C. This calculation assumes that the Hb has a normal P₅₀ (see Chapter 6). Calculated O₂Hb may significantly overestimate true saturation in the presence of increased levels of COHb or methemoglobin. The hemoximeter provides the most accurate estimate of the actual O₂ saturation. Most laboratory blood gas analyzers and some POC analyzers combine blood gases and hemoximetry. These instruments provide pH, PCO₂, PO₂ and spectrophotometric measurements of Hb saturation, all performed with the same blood sample.

Hemoximeters may give erroneous Hb, O₂Hb, or COHb readings if forms of hemoglobin are present that the instrument does not recognize. For example, blood from a newborn (e.g., containing fetal Hb) will give erroneous values if analyzed by an oximeter set up for adult blood. Substances that cause light scattering in the specimen (e.g., lipids resulting from lipid therapy) may also cause false readings. To function properly, the hemoximeter must hemolyze the sample so that Hb molecules are suspended in solution rather than contained within the red cells. Hemolysis is accomplished by chemical or mechanical disruption of red cell membranes. Incomplete hemolysis results in light scattering within the sample rather than simple absorption. Sickle cells are not easily disrupted, particularly by chemical lysis, and may result in false readings for O₂Hb and COHb. Incomplete hemolysis may be difficult to detect unless whole-blood QC or proficiency testing is performed (see Chapter 12). Blood specimens used for hemoximetry must be well mixed or the concentration of Hb (i.e., total Hb) may be incorrect. Most hemoximeters report errors such as incomplete hemolysis or light scattering.

In addition to measurements of O₂Hb, COHb, and MetHb saturations, the hemoximeter can calculate oxygen content and P₅₀. P₅₀ can be estimated by measuring the actual saturation of a specimen (usually a venous sample with a saturation of less than 90%) and comparing this value with the calculated saturation based on PO₂ and pH of the same blood. This simplified method compares favorably with tonometering of the blood sample with various low-oxygen concentrations and constructing a dissociation curve.

Pulse Oximeter

Pulse oximeters (Figure 11-34) are commonly used to assess oxygenation noninvasively. Pulse oximeters treat Hb as a filter that allows only red and near-infrared light to pass. The Lambert-Beer law relates total absorption in a system of absorbers to the sum of their individual absorptions.

In principle, the standard pulse oximeter measures absorption of a mixture of just two substances, O₂Hb and RHb. The concentration of either one can be determined if their extinction is measured while the path length stays constant. The wavelengths of light used in standard pulse oximetry are near 660 nm in the red region of the spectrum and near 940 nm in the near-infrared region. Extinction curves for O₂Hb and RHb show that reduced Hb has absorption 10 times higher than oxyhemoglobin at 660 nm, whereas O₂Hb has a higher absorbance (two to three times) at 940 nm. Calculating all possible combinations of the two forms of Hb (i.e., varying the saturation from 0%–100%) allows the ratio of absorbances at the two wavelengths to be determined. As a result, a calibration curve can be constructed. The capillary bed does not follow the optical principles exactly as described by the Lambert-Beer law, so the calibration curve is derived empirically. The ratio of absorbances at the two distinct wavelengths is expressed as follows:

R=A660nm/A940nm

A series of R values (e.g., the calibration curve) is determined by relating this ratio to actual saturation measurements. Unlike the hemoximeter, which measures absorption in a hemolyzed blood sample, the pulse oximeter measures light passing through living tissue. The transmitted light is not only absorbed but also refracted and scattered. This causes the absolute accuracy of the pulse oximeter to be less than the accuracy of a hemoximeter.

The transmitted light at each wavelength consists of two components, the AC and DC components (Figure 11-35). The AC component varies with the pulsation of blood. The DC component represents light absorbed by tissue and venous blood. The DC component is larger than the AC and is relatively constant. The amplitude of both AC and DC levels depends on the intensity of the incident light. The AC component represents the arterial blood because the arterioles pulsate in the light path. By dividing the AC level by the DC level at each of the two wavelengths, the AC component is effectively corrected. The AC component then becomes a function of the extinction of O₂Hb and RHb. The ratio just described then becomes:

R=AC1/DC1AC2/DC2

where:

1 = red wavelength (660 nm)

2 = near-infrared wavelength (940 nm)

Correcting the pulsatile component (AC) in this manner allows the pulse oximeter to “ignore” absorbances caused by venous blood, tissue, and skin pigmentation.

The light source used in pulse oximetry is the Keytermlight-emitting diode (LED). LEDs are capable of emitting a very bright light near the 660-nm and 940-nm wavelengths required for analysis of Hb saturation. Light intensity is controlled by a feedback circuit that regulates the driving current to the LED. The greater the DC component resulting from pigmentation or venous blood, the greater the current supplied to the LED. One problem with LEDs is that the exact wavelength of light emitted varies with individual diodes. Each LED has its own center wavelength that may differ from 660 nm or 940 nm by as much as 15 nm. To overcome this variation, each oximeter must have a series of calibration curves programmed into it so that it can accommodate a range of LEDs. The extinction curves for RHb and O₂Hb are steep and different at 660 nm, so 10 or more calibration curves are typically required for the red-light range. Slight variations in center wavelength are less critical in the 940-nm region because the extinction characteristics of O₂Hb and RHb are the same from 800–1000 nm.

A photo diode detects transmitted light in the pulse oximeter. A single photo diode senses both red and near-infrared light. The microprocessor that controls the oximeter cycles the LEDs on and off separately 400–500 times per second. The oximeter also turns both LEDs off during each cycle. This allows the photo diode to detect ambient light caused by scattering and to offset the LED signals.

Pulse oximeter accuracy tends to decrease at low saturations. Low saturations occur as the concentration of RHb increases. RHb has a much higher absorbance at 660 nm than does O₂Hb. Therefore, slight variations in the center wavelength of the red LED (as described) exaggerate the error in measured saturation. This is one reason pulse oximeters exhibit decreasing accuracy at lower saturations.

Because the AC, or pulsatile component, is usually much smaller than the DC component, detecting it can sometimes cause problems. Low perfusion or poor vascularity can cause the oximeter to be unable to measure the pulsatile component. Most oximeters display a warning message if the photo detector senses inadequate light levels. Motion artifact can also cause inaccuracy with pulse oximeters. Movement, especially shivering, often occurs in the same frequency range as the signal to be detected (i.e., arterial pulsations). If the motion is consistent and lasts long enough, it introduces a signal of approximately the same amplitude into both red and infrared channels. The pulse oximeter senses motion artifact as part of the DC component. This adds a large value to both the numerator and denominator of the ratio (R). The motion signal forces R toward a value of 1, which is equal to a saturation of 85% on the typical oximeter calibration curve. Some oximeters use multiple digital filters to discriminate motion artifact and produce a more reliable signal.

Most pulse oximeters use the AC signal from one channel (660 nm or 940 nm) to calculate pulse rate. An algorithm implemented by the microprocessor locates peaks in the waveform of the AC signal and counts them (see Figure 11-35). Some oximeters use this signal to display graphic representations of pulse waveforms. Pulse oximeters that measure absorption at multiple wavelengths (see Figure 11-34) can detect Hgb (SpHb), COHb (SpCO), and MetHb (SpMet), in addition to the standard SpO₂.

Reflective Spectrophotometers

KeytermReflective spectrophotometry is based on the variable reflection of light by O₂Hb and RHb at different wavelengths. Just as the light absorbed by O₂Hb and RHb is a function of wavelength, so is the intensity of reflected or back-scattered light. Carefully spaced optical fibers can be used as transmitting and receiving paths for light. Reflective spectrophotometry can be used to monitor arterial saturation in a manner similar to that of pulse oximetry. Reflective spectrophotometry can also be incorporated into a pulmonary artery catheter to measure mixed venous oxygen saturation (Sv¯o₂).

A reflective sensor can be placed at a site where a thin layer of tissue covers bone, such as the forehead. The oximeter then measures O₂Hb in a manner similar to that used for pulse oximetry but uses reflected rather than absorbed light. Pulse oximeters are available that can use either the regular sensor (i.e., absorption) or a reflective sensor.

A specially designed pulmonary artery (Swan-Ganz) catheter contains fiberoptic bundles (Figure 11-36). This catheter has regular pressure-sensing ports, a balloon tip for flotation through the right side of the heart, and capability for cardiac output determinations. Three LEDs similar to those used in pulse oximeters illuminate blood flowing past the tip of the catheter via one of the optical fibers. A photodetector senses the reflected light and converts its intensity into a signal. A microprocessor calculates two independent ratios of reflected light intensities from the three wavelengths. Combining two reflected light intensity ratios reduces the instrument’s sensitivity to pulsatile blood flow or changing hematocrit. This design also minimizes changes caused by light scattering from red cell surfaces and the walls of the blood vessel. Sv¯o₂ is calculated from the light ratios using programmed calibration curves, similar to those used for a pulse oximeter. As in a pulse oximeter, saturation measured is the saturation of functional Hb (see Chapter 6). Sv¯o₂ determination by this method tends to be higher than that measured by a hemoximeter, especially if large amounts of COHb or MetHb are present. Sv¯o₂ is then displayed and may be printed using a trend recorder (Figure 11-37).

The reflective spectrophotometer must be routinely calibrated to ensure that observed changes in Sv¯o₂ are the result of physiologic phenomena rather than instrument drift. The catheter is usually standardized by calibrating it against an absolute color reference before insertion. After the catheter is in place, calibration is accomplished by adjusting the output to match saturation measured by a co-oximeter. This type of calibration is accurate at the time it is performed but may change if there are shifts in pH or hematocrit. Because reflective spectrophotometers measure reflected light in whole blood that is flowing rather than transmitted light in a hemolyzed blood sample, their absolute accuracy is less than that of a hemoximeter.

Data Acquisition and Instrument Control

Computerized pulmonary function systems process analog signals from spirometers, plethysmographs, and gas analyzers. Equally important is the computer’s capacity to control instrument functions, such as switching valves or recording signals. Computer control allows the technologist to manage complex test maneuvers. Data acquisition and instrument control are implemented with an interface (Figure 11-39) between the computer and pulmonary function equipment. Similar principles are applied in large laboratory systems and in small handheld spirometers.

An important component of the pulmonary equipment interface is the analog-to-digital (A/D) converter. The A/D converter senses an analog signal (such as pressure or flow) and transforms it into a digital value. The analog signal is usually a DC voltage in the range of either 0–10 V or ±5 V. A/D converters are classified by the number of bits (binary digits) into which they convert the signal. The higher the number of bits, the greater the resolution of the input signals. A 12-bit converter can transform a voltage into a number represented by 000000000000 to 111111111111 as a binary number. In the decimal notation, this corresponds to a range of 0–4096, or 2¹². For example, a 10-L spirometer may produce an analog signal ranging from 0–10 V (i.e., 1 V = 1 L). If the spirometer is connected to a 12-bit converter, the signal can be divided into 4096 parts. This provides a resolution of approximately 0.0024 V or 2.4 mL. The smallest volume change that can be detected would be 2.4 mL for this spirometer system. For most volume and flow sampling applications, 12-bit converters are adequate. For greater resolution, a 16-bit A/D converter may be used. Some systems use transducers (i.e., pressure or flow) that directly produce a digital output. These transducers do not require an A/D converter, simplifying the measurement.

The rate at which data are sampled also affects accuracy. High-speed A/D converters can perform more than 20,000 conversions per second on a single channel. Most computerized pulmonary function systems sample data 100 times/sec (i.e., 100 Hz) or greater. These high rates exceed the frequency bandwidth of breathing maneuvers by a factor of more than two. Accuracy is attained by matching analog output of a device (e.g., a spirometer) to an A/D converter with appropriate sampling rate and voltage resolution.

Another means of sampling volume or flow signals is measuring the time required for a known change in volume or flow. For example, the number of clock ticks that occur for a volume change of 100 mL can be counted and flow calculated. This technique requires a spirometer that uses a position encoder or generates pulses for each volume increment. The accuracy of encoder-pulse systems depends on clock resolution and volume increment, especially when measuring high flows.

Most devices (e.g., pulse oximeters or capnographs) use dedicated microprocessors and A/D boards to process data. These instruments often include a communication port so that data can be sent to a PC or printer. Most computers have serial ports (USB or conventional) that can be used to interface various instruments. Some instruments (e.g., monitors) include Ethernet ports so that the device can communicate as a network device.

Pulmonary Function Data Storage and Programs

Computerized pulmonary function systems, even small, portable spirometers, generate large volumes of patient data. Managing these large amounts of data is relatively easy via computer networks. Almost all PC-based systems support networking to some extent. Networked computers allow multiple users to share data, as well as peripheral devices such as printers. The network typically consists of a primary computer or server. Other computers are then linked by one of several types of networks. Data can be transferred between any two computers linked by the system. The server usually maintains programs and data that all of the networked users need.

The format in which test data are stored is determined by (1) complexity of the data, (2) volume of data (i.e., number of tests done per month or year), and (3) how data will be accessed. Three methods for data management are used with computerized pulmonary function systems.