`
`
`January 1989
`
`ANESAV
`
`ISSN 0003-3022
`
`
`
`The journal of
`
`The American Society of
`
`Inc.
`Anesthesiologists,
`
`
`Volume ’70
`
`0
`
`Number 1
`
`Complete Contents
`
`Pages 3, 5, 7, 9, and 11
`
`APPLE 1010
`
`1
`
`APPLE 1010
`
`
`
`Juanliesiohiy
`
`THE JOURNAL OF
`THE AMERICAN SOCIETY OF ANESTHESIOLOGISTS, INC.
`Editor-in-Chief
`LAWRENCE]. SAIDMAN, M.D., San Diego, California
`Editors
`
`David E. Longnecker, MD.
`julien F. Biebuyck, M.B., D.Phil.
`Philadelphia, Pennsylvania
`Hershey, Pennsylvania
`Dennis T. Mangano, Ph.D., MD.
`john]. Downes, M.D.
`San Francisco, California
`Philadelphia. Pennsylvania
`Kai Rehder, M.D.
`H. Barrie Fairley, M.B., B.S.
`Rochester, Minnesota
`Stanford, California
`Donald R. Stanski, MD.
`Simon Gelman, M.D., Ph.D.
`Stanford, California
`Birmingham. Alabama
`Michael M. Todd, MD.
`Carol A. Hirshman, MD.
`Iowa City, Iowa
`Baltimore, Maryland
`Warren M. Zapol, M.D., Boston, Massachusetts
`
`Associate Editors
`
`Carl Lynch III, M.D., Ph.D.
`Charlottesville, Virginia
`Mervyn Maze, M.B., Ch.B.
`Stanford, California
`
`Henry Rosenberg, M.D.
`Philadelphia, Pennsylvania
`Gary R. Strichartz, Ph.D.
`Boston, Massachusetts
`
`
`
`Charles W. Buffington, MD.
`Pittsburgh, Pennsylvania
`David H. Chestnut, MD.
`Iowa City, Iowa
`jeffrey E. Cooper, Ph.D.
`Boston, Massachusetts
`Dennis M. Fisher, M.D.
`San Francisco, California
`Tony L. Yaksh, Ph.D.
`Thomas F. Hornbein, M.D.
`San Diego, California
`Seattle, Washington
`
`editorial management, and Letters to the Editor should be mailed to Lawrence
` n, correspondence relating to
`a—-San Diego, San Diego, California 92093. Books for review should
`Manuscripts for publicatio
`J. Saidman, MD” Anesthesiology, T-015, University of Californi
`The Children's Hospital of Philadelphia, One Children's Plaza, 34th
`be mailed to John J. Downes. M.D., Department of Anesthesia,
`‘
`,
`SETECI and Civic Center Boulevard, Phlladelphia, PA 19104.
`.
`All articles are accepted for publication with the understanding that they are contributed exclusively to this journal and become the
`property of the American Society of Anesthesiologists, Inc. Statements or opinions expressed in the Journal reflect the views of the
`author(s) and do not represent official policy of the American Society of Anesthesiologists unless 50 stated,
`For details and recommendations concerning the preparation of manuscripts see GUIDE FOR AUTHORS, which appears in the january
`and u]
`’
`,
`.
`.
`Advgrliisssiii:and related correspondence should be addressed to Advertising Manager, ANBSTHESIOLOGV. J. B. Lippincott Company,
`East Washin ton Square Philadelphia, Pa. 19105. Publication ofan advertisement m ANESTHESIOLOGY does not constitute endorsement
`by ANESTHSSIOLOGY of the product described therein or of any representations made by the advertiser with respect to the product.
`_________________—
`0034022) is published 13 times a year {monthly except September. in which two issues are published)
`ANESTHESIOLOCY (ISSN 0
`incott Company, at Downsville Pike, Route 3, Box 20-3, Hagerstown,
`esthesiologists. lnc.. by J. E. Lipp
`.
`.
`.
`F01- the American Society of An
`East Washington Square. Philadelphia, PA 19105. Printed in the U. S. A. © Copyright 1939
`MD 21740. Business ofl‘ices are located at.
`5' 1'“- Smnd Class Pmuse Paid at Hagerstown. MD. and at additional mailing unite,
`by the American Society of Anesthesiologist
`ndence related to membership and change of address should be mailed to the Executive
`NDtice of Change of Address: All correspo
`of Anesthesiologists, 515 Busse Highway, Park Ridge, Illinois 50008. Correspondence regarding nonmember
`Sec
`-
`-
`subzsggéhfifligfig 3:231); to journal Fulfillment Department. Lippincott/Harper, Downsvlile Pike, Route 3. Box 20-3, Hagerstown.
`Maryland 21740 or call 1300-6538-3030: in Maryland. call collect 301-824-7300. Correspondence regarding nonmember subscriptions
`injapan includirig inquiries about subscription ratesand subscription orders, should be sent to Woodbell- Incorporated, 4-224 1. Kitakasai,
`Edogawaku, Tokyo 134.]8P3'1-
`Bangladesh. and Sn Lanka contact Universal Subscription Agency Pyt. Ltd., 101-102
`In India, Nepal.
`7
`d'
`C
`-
`,
`. S l: t,New Delhi—11001 , In ta.
`.
`’
`.
`‘
`.
`Agilflllgiilssgjigiiifrgdes:at: S. $40.00 individual. $50.00 institution; all other countries exceptJapan. India, Nepal, Bangladesh, and
`$8.00. Rates for air mail delivery available upon requesl'
`Sri Lanka, $7 5.00 individual, $85.00 institution. Single copies
`If the Society Office or the Publisher receives a request for repIaeetnent within (SO-days of the mailing in the U. S. or within 90 days in
`all other countries and the original issue is returned by the postal service as undeliverable, the issue will be replaced.
`POSTMASTER- Send address changes t
`Route 3, Box 20-3. Hagerstown, MD 21740_
`0 Anesthesiology, Downsville Pike,
`A [h
`g
`.
`'tems for in
`or the internal or personal use of specific clients, is granted by_The
`.
`ternal or personal use.
`Agteifigzdtéglcliéfy iffifigilieiiologists for libraries and other users registered with the Copyright Clearance Center (CCC) Transactional
`Reporting Service, provided that the base fee of $00.80 per copy. plus $00.05 per page is paid directly to CCC, 21 Congress St., Salem,
`t extend to other kinds of copying, such as copying for general distribution,
`MA 01970. 0003-3022/89 $00.80 + $00.05. This does no
`for advertising or promotional
`ollectlve works. or for resale.
`
`
`2
`
`
`
` Juestflesiolofly
`
`January 1989
`
`EDITORIAL VIEWS
`
`
`
`CONTENTS
`
`A Change in Format for ANESTHESIOLOGY.
`Lawrence ]. Saidman
`Changing Perspectives in Monitoring Oxygenation.
`H. Barrie Fairley
`Studies in Animals Should Precede Human Use of Spinally Administered Drugs.
`Tony L. Yaksh and]. G. Collins
`
`CLINICAL INVESTIGATIONS
`
`The Influence of Renal Failure on the Pharmacokinetics and Duration of Action of
`Pipecuronium Bromide in Patients Anesthetized with Halothane and Nitrous
`Oxide.
`james E. Caldwell, P. Ciaver Canfi’h’, Kay P. Castagnoii, Daniel P. Lynam, Mark R.
`Fahey, Dennis M Fisher, and Ronald D. Miller
`Effect of Intercostal Nerve Blockade on Respiratory Mechanics and CO2 Chemosen»
`sitivity at Rest and Exercise.
`Bernice R. Hecker, Robert Bjurstrom, and Robert B. Schogng
`ocardiographic, Mechanical, and Metabolic In-
`Comparison of Hemodynamic, Electr
`dicators of Intraoperative Myocardial Ischemia in Vascular Surgical Patients with
`Coronary Artery Disease.
`Sti‘ren Haggmark, Per Hohner, Margareta ostrnan, Arnold Friedman, George Diamond,
`Edward Lowenstein, and Sebastian Reiz
`mall Doses of Sufentanil or Fentanyl: Dose Versus EEG
`and Thiopental Requirement.
`
`Induction of Anesthesia with S
`Response, Speed of Onset,
`T. Andrew Bowdle and Richard j. Ward
`Relationship of Mivacurium Chloride in Humans during Nitrous
`OxideeFentanyl or
`Nitrous Oxide—Enflurane Anesthesia.
`Mark R. Fahey, Daniel P. Lynam, and
`james E. Caldwell, john B. Kim, Torn Heier,
`Ronald D. Miller
`
`The Dose-Response
`
`the Prostate, Serum Glycine Levels, and Ocular Evoked
`
`Transurethral Resection of
`Potentials.
`janice Mei-Li Wang, Donnell ]. Creei, and K. C. Wong
`
`2
`
`4
`
`7
`
`13
`
`19
`
`26
`
`3}
`
`36
`
`—/——-
`
`(Continued on page 5)
`
`3
`
`
`
`January 1989—ANESTHESIOLOGY
`
`CONTENTS
`
`(3‘
`
`(Continued from page 3)
`
`Lower Esophageal Contractility Predicts Movement during Skin Incision in Patients
`Anesthetized with Halothane, but Not with Nitrous Oxide and Alfentanil.
`Daniei I. Sessier, Randi Semen, Christine I. 0mm, and Franklin Chow
`
`Determination of Inna-abdominal Pressure Using a Transurethral Bladder Catheter:
`Clinical Validation of the Technique.
`Thomas j Iberii, Cherries E. Lieber, and Ernest Benjamin
`
`LABORATORY INVESTIGATIONS
`
`Epidural Clonicline Analgesia in Obstetrics: Sheep Studies.
`janies C. Eisenaeh, Maria I. Castro, David M. Dawn, andjames C. Rose
`
`The Enhancement of Proton/Hydroxyl Flow across Lipid Vesicles by Inhalation An-
`esthetics.
`
`Dongios E. Rainer and David S. Cafiso
`The Influence of Dextrose Administration on Neurologic Outcome after Temporary
`Spinal Cord lschernia in the Rabbit.
`john C. Druminond and Suzanne S. Moore
`Tachyphylaxis to Local Anesthetics Does Not Result from Reduced Drug Effectiveness
`at the Nerve Itself.
`Peter Lipferi, Hoiger Holihnsen, andjoachim O. Amati
`Comparison of the Effects of Halothane on Skinned Myocardial Fibers from Newborn
`and Adult Rabbit. I. Effects on Contractile Proteins.
`Eiiiot]. Krane ondjndy Y. Su
`Regional Differences in Left Ventricular Wall Motion in the Anesthetized Dog.
`Johan Diederieks, Bruce f. Leone, and Pierre Foe'x
`Effects of “Nitrendipine” on Nitrous Oxide Anesthesia, Tolerance, and Physical De-
`pendence.
`S. ] Dolin and H. ] Liiiie
`
`42
`
`47
`
`51
`
`57
`
`64
`
`71
`
`76
`
`32
`
`91
`
`MEDICAL INTELLIGENCE ARTICLE
`
`Pulse Oximetry.
`Kevin K. Tremper and Steven _]. Barker
`
`98 /
`
`(Continued on page 7}
`
`f
`
`4
`
`
`
`January Igag—ANES'rl-n-LSIOLOGY
`
`CONTENTS
`
`(Continued from page 5)
`
`LABORATORY REPORTS
`
`Laudanosine Does Not Displace Receptor-specific Ligands from the Benzodiazepinergic
`or Muscarinic Receptors.
`Yeshayahu Kai: and Moshe Gaoish
`
`Effects of Mathemoglobinemia on Pulse Oximetry and Mixed Venous Oximetry,
`Steven j. Barker, Kevin K. Tremper, and john Hyatt
`Hyperbilirubinemia Does Not Interfere with Hemoglobin Saturation Measured by Pulse
`Oximetry.
`Francis Veyckemans, Philippe Boole, J" E- Guillaume, Eric Williams, Annie Robert, and
`Thierry Cierbaux
`
`Evaluation of a Blood Gas and Chemistry Monitor for Use during Surgery.
`0- Bashein, W955?) K- Grgydanus, and Margaret A. Kenny
`
`A Model for Determining the Influence of Hepatic Uptake of Nondepolarizing Muscle
`Relaxants in the Pig.
`johann Motsch, Pim ]. Hanna's, Franz Alto Zimmermann, and Sandor Agoston
`Transtracheal Doppler: A New Procedure for Continuous Cardiac Output Measure-
`ment.
`jeroine H. Abrams, Roland E. Weber, and Kenneth D. Hoimgn
`
`CASE REPORTS
`
`Management of Acute Elevation of Intracranial Pressure during Hepatic Transplan—
`tation.
`D. Brajzbord, R. 1'. Parks, M. A. Ramsay, A. W. Paulsen, T. R. Valek, T, H, Stoygerz,
`and G. B. Klintmalin
`Treatment of Isorhythmic A-V Dissociation during General Anesthesia with Propran-
`olol.
`Russell F. Hit!
`
`Fiberoptic Endobronchial Intubation for Resection ofan Anterior Mediastinal Mass.
`Dirk Younker, Randal! Clark, and Lewis Cooeler
`Postpartum Seizure after Epidural Blood Patch and Intravenous Caffeine Sodium Ben—
`
`zoate.
`
`109
`
`1 12
`
`118
`
`123
`
`I28
`
`134
`
`1 39
`
`141
`
`144
`
`146
`
`ft
`
`(Continued on page 9)
`
`Vincent E. Boi
`
`on, Craig H. Leickt, and Thomas S. Stanton
`
`5
`
`
`
`January l989—-ANESTHESI(')[.OGY
`
`9
`
`CONTENTS
`
`.
`-
`I
`Caudal Epidural AnestheSIa in an Infant With Epidermolysis Bullosa.
`Lawrence L. Yee, jaet B. Gunter, and Charter B. Manley
`
`((Jnntinued from page 7)
`149
`
`Recurrent Respiratory Depression after Alfentanil Administration.
`Rory S. jafi and Dennis Coateon
`
`Pain of Delayed Traumatic Splenic Rupture Masked by Intrapleural Lidocaine.
`Wiitiarn W' Pond, Gregory M' Somerviile, Siong H- Thong, James A. Ranochak, and
`Gregory A. Weaver
`
`Dose-response Relationship for Succinylcholine in a Patient with Genetically Determined
`Low Plasma Cholinesterase Activity.
`Charles E. Smith, Geraént Lewis, Francois Donati, and David R. Bevan
`
`CORRESPONDENCE
`
`Datermination of Decay Constants from Time-varying Pressure Data.
`Charles Beattie, Linda S. Humphrey, and Gary Mamscha}:
`Reply. Ctgfiom’ R. Swanson and Witit'arn W. Muir III
`
`Use Caution when Extrapolating from a Small Sample Size to the General Population.
`David ]. Benefiei, Edward A. Easter, and Rodger Shepherd
`Reply. Imad H. AbduE-Rasoot, Daniel H. Sears, and Renata L. Km;
`Succinylcholine and Trismus.
`Fredertc A. Berry and Carl Lynch HI
`Reply. Henry Rosenberg
`An Infant Model to Facilitate Endotracheal Tube Fixation in the Pediatric ICU Patient.
`Patrick K. Birmingham and Babette Horn
`
`An Alternative Method for Management of Accidental Dural Puncture for Labor and
`Delivery.
`Shout Cohen, jonathan S. Daitch, and Paul L. Gotdiner
`
`High-pressure Uterine Displacement.
`Michael ]. Dorsey and Waiter L. Mittar
`
`Calculating the Potency
`
`of Mivacurium.
`
`Aaron F. Kopman
`Reply. john]. Savarese
`
`151
`
`154
`
`156
`
`159
`
`160
`
`160
`
`16]
`161
`
`162
`163
`
`164
`
`165
`
`166
`
`166
`
`f
`
`(Continued on page ] l)
`
`6
`
`
`
`january 1989——A N I‘ISTI'l ESIO LOGY
`
`CONTENTS
`
`.
`.
`‘
`‘
`Midazolam 1n a Malignant Hyperthermia-susceptible Patient.
`‘
`jultana FL]. Brooks
`
`I 1
`
`(Continued from page 9)
`167
`
`Exchange Autotransfusion Using the Cell Saver during Liver Transplantation.
`Mare R. Brown, Michael A. E. Ramsay, and Thomas H. Sanger:
`
`Air Entrainment Through a Multiport Injection System.
`Dean Gilbert, Theodore]. Sanford, Jr, and Brian L. Partridge
`Reply. Thelma Maoedo
`
`A Tracheal Tube Extension for Emergency Tracheal Reanastomosis.
`Robert S. Holzman
`
`The Relationship Between Malignant Hyperthemia and Neuroleptic Malignant Syn-
`drome.
`Haggai Homesh, Dov Aizenberg, Margo Lapidot, and Hanan Mung;
`Reply. Stanley N, Carof, Stephan C. Mann, Henry Rosenberg, jqfrey E, match”, and Terry
`D. Herman-Patterson
`
`Appmpriate Facilitation of Intravenous Regional Techniques in RSD.
`Kevin Foley, Linda Schatz, and Randall L. Martin
`
`REPORT OF SCIENTIFIC MEETING
`
`ANNOUNCEMENT
`
`168
`
`159
`
`170
`
`170
`
`171
`
`172
`
`17 3
`
`174
`
`173
`
`issue.
`
`The Guide for Authors is published in the january and july issues. It may be found on page 33A of this
`
`GUIDE '10 AUTHORS
`
`A NESAV is re rode word (“rorh'n”) mm" by the Chemiml' xi‘hflmrf Sena-re m idemflj the journal
`
`——#—
`
`7
`
`
`
`This material may be protected by Copyright law (Title 17 U.S. Code)
`
`8
`
`
`
`Amdtesiology
`V70, N01,]an1989
`
`PULSE oxmaTRv
`
`99
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`%0‘SATURATION
`
`became a standard clinical and laboratory tool in pul-
`monary medicine. Although it was demonstrated to be
`accurate for intraoperative monitoring.m its size and ex-
`pense. and the cumbersome nature of the ear probe pre-
`vented its acceptance as a routine monitor. At this time.
`all oximeters produced various light source wavelengths
`by filtering white light. The filtered light was then trans
`mitted to and from the tissue through fiberoptic cables.
`In the mid 19705, Takuo Aoyagi. an engineer working
`for Nihon Kohden Corporation. made an ingenious dis
`covery regarding oximetry. He was developing a method
`to estimate cardiac output semi—noninvasively by detecting
`the washout curve of dye injected into a peripheral vein
`as it perfused the car. This washout curve was measured
`in the ear with a red and infrared light densitometer sim-
`ilar to the Millikan ear oximeter. He noticed that his
`
`washout curves contained pulsations due to the arterial
`pulse in the ear. To more easily analyze the dye washout
`curve, he subtracted these pulsations from the curve, and
`in doing so he discovered that the absorbance ratio of the
`pulsations at the two wavelengths changed with arterial
`hemoglobin saturation. He soon realized that he could
`build an ear oxirneter that measured arterial hemoglobin
`saturation without heating the ear by analyzing pulsatile
`light absorbances? This first pulse oximeter, developed
`by Nihon Kohden, used filtered light sources similar to
`Millikan’s ear oximeter. The device was evaluated clini-
`cally in the mid 19705 and marketed with little success.2
`In the late 19705. Scott Wilber in Boulder. Colorado,
`developed the first clinically accepted pulse oximeter by
`making two modifications of
`the Nihon Kohden
`method“: First, he produced a lightweight sensor by
`using light emitting diodes (LEDs) as light sources and
`photodiodes as detectors. Consequently, the instrument
`was connected to its earciip sensor only by a small electrical
`cable. Wilber also improved the saturation estimates by
`using a digital microprocessor to store a compiex calibra-
`tion algorithm based on human volunteer data”: This
`method will be discussed in more detail below. This device
`
`was developed by Biox Corporation of Boulder. Colorado,
`and was successfully marketed to pulmonary function lab-
`oratories in the early 19805.
`The clinical utility of the noninvasive oximeter in the
`operating room was rediscovered in the 19805 by William
`New, an anesthesiologist at Stanford University. Realizing
`that a continuous, noninvasive monitor of oxygenation
`would be useful to anesthesiologists, New developed and
`marketed a pulse oximeter to this group.155 The Nellcor
`model N100 had by 1985 become almost synonymous
`with the term "pulse oximeter."
`
`i Wilber 5: Blood constituent measuring device and method. US
`Patent #4. 407. 290 April 1. 1981.
`
` mo ‘ PI' u 4: H's- " a: la 0° 12'“ :2”
`
`
`
`
`
`FIG. 1. Thisfigure is from a 1951 article in ANES’I‘HESIOI.OGY. It
`reveals dramatic desaturation in a 4-yr-old patient during a tonsillec-
`tomy. Reproduced from Stephen RC. Slater HM. johnson AL, Sekelj
`P: The oximeter—A technical aid for the anesthesiologist. ANESTHE—
`SIOLOGY 12:548. 1951. with permission.
`
`The Physics and Physiology of Pulse Oximetry
`
`BEER’S Law
`
`In the 19305, Matthes used spectrophotometry to de-
`termine hemoglobin oxygen saturation.2 This method is
`based on the Beer-Lambert law, which relates the con-
`centration ofa solute to the intensity oflight transmitted
`through a solution.
`
`1trans = Iine‘n
`
`A = DC:
`
`(l)
`
`(1a)
`
`where 1......“ = intensity oftransmitted light; I... = intensity
`of incident light; A = absorption; D = distance light is
`transmitted through the liquid (path length); C : con-
`centration of solute (hemoglobin); E- = extinction coeffi~
`cient of the solute (a constant for a given solute at a Spec-
`ifted wavelength). Thus, if a known solute is in a Clear
`solution in a cuvette of known dimensions, the solute con—
`centration can be calculated from measurements of the
`incident and transmitted light intensity at a known wave-
`length. The extinction coefficient e is a property of light
`absorption for a specific substance at a specified wave-
`length. In a one-component system. the absorption A is
`the product of the path length, the concentration, and
`the extinction coefficient. equation la. If multiple solutes
`
`9
`
`
`
`K. K. TREMPER AND S. _|. BARKER
`
`Anesthesiole
`V 70. No ].Jan 1989
`
`By the above definition of oxygen saturation, the two
`forms of hemoglobin that do not bind oxygen (COHb
`and MetHb) are not included. This is the origin of what
`is now referred to as “functional hemoglobin saturation,”
`defined as (Severinghaus jW, personal communication):
`
`OgHb
`Functional SaOz = 6m x 100%.
`
`(2)
`
`With the advent of multiwavelength oximeters that can
`measure all four species of hemoglobin, “fractional sat-
`uration” has been defined as the ratio of oxyhemoglobin
`to total hemoglobin:
`
`reduced
`hemoglobin
`
`
`
`960
`
`Il||l
`
`i
`/ ’ .— l- "'
`
`I oxyhemoglobin
`/
`nun-"uuu'.
`
`
`
`HEMOGLOBIN EXTINCTION cu nves
`
`
`Fleet
`l
`
`/"'\ i
`
`
`'
`
`Infrgred
`i
`I
`
`methemoglobln
`
`100
`
`10
`
`1
`
`
`
`ExtinctionCoefiicient
`
`
`
`
`
`carboxyhemoglobin~._.
`"" '
`-
`sec
`880
`
`son
`
`920
`
`.01
`sun saw 630
`
`720
`
`760
`
`Wavelength )t {rim}
`
`Fractional Sa02
`
`FIG. 2. Transmitted light absorbance spectra of four hemoglobin
`species: oxyhemoglobin. reduced hemoglobin, wrboxyhemoglobin, and
`methemoglobin. Adapted from Barker 5] and Tremper KK: Pulse
`Oximetry: Applications and limitations, Advances in Oxygen Moni-
`toring, International Anesthesiology Clinics. Boston. Little, Brown and
`Company. 1987. pp. 155—175.
`
`are present, A is the sum of similar expressions for each
`solute. The extinction coefficient can vary dramatically
`with the wavelength of the light. The extinction coeffi-
`cients for various hemoglobin species in the red and in-
`frared wavelength range are shown in figure 2.
`Laboratory oximeters use this principle to determine
`hemoglobin concentration by measuring the intensity of
`light transmitted through a cuvette filled with a hemo-
`globin solution produced from lysed red blood cells.“1
`For Beer's law to be valid, both the solvent and the cuvette
`
`must be transparent at the wavelength used, the light path
`length must be known exactly, and no absorbing species
`can be present in the solution other than the known solute.
`It is difficult to fulfill these requirements in clinical devices;
`therefore, each instrument theoretically based on Beer’s
`law also requires empirical corrections to improve accu-
`racy.
`
`HEMOGLOBIN SATURATION DEFINITIONS
`
`Adult blood usually contains four species of hemoglo-
`bin: oxyhemoglobin (OgHb), reduced hemoglobin (Hb),
`methemoglobin (MetHb), and carboxyhemoglobin (CO-
`Hb) (fig. 2). The last two species are in small concentra-
`tions, except in pathologic conditions. There are several
`definitions of hemoglobin saturation. Historically, “oxy-
`gen saturation” was first defined as the oxygen content
`expressed as a percentage of the oxygen capacity. The
`oxygen content (cc of oxygen per 100 cc of blood) was
`measured volumetrically by the method of Van Slyke and
`Neill (1924).15 The oxygen capacity was defined as the
`oxygen content after the blood sample had been equili-
`brated with room air (158 mmHg oxygen at sea level).
`
`02H!)
`= ——--———-—-—-— X 100 .
`021'“) + Hb + COHb + MetHb
`%
`
`(3)
`
`The fractional hemoglobin saturation is also called the
`“oxyhemoglobin fraction,” or “oxyhemoglobin %.“”
`When oximetry is used to measure hemoglobin satu-
`ration, Beer‘s law must be applied to a solution containing
`four unknown species: OgHb, Hb, COHb, and MetHb.
`Expanding equation 1a to a four-component system results
`in an absorption given by:
`
`A = chlel ‘l' D2026? + D3C3€3 'l" 1340464.
`
`(1b)
`
`The subscripts 1 through 4 correspond to the four he-
`moglobin species. If the path lengths are the same, then
`D can be factored out:
`
`A. = D(C1E1 ‘i‘ 0263 "i’ 0363 'l' Catt).
`
`(1C)
`
`The extinction coefficients 6; through a, are constants at
`a given wavelength A (fig. 2). The absorption defined in
`equation 1c is determined from equation 1 by measuring
`the incident and transmitted light intensities. From equa-
`tion 1c, we see that four wavelengths of light are needed
`to produce four equations to solve for the unknown con-
`centrations Cl through Cs. If COHb and MetHb were
`not present, their contributions to the absorption would
`be zero and functional hemoglobin saturation could be
`determined by a two-wavelength oximeter (two equations
`and two unknowns). If two wavelengths existed for which
`the extinction coefficients for COHb and MetHb were
`
`zero, then these absorption terms would again be zero
`and a two-wavelength oximeter could measure functional
`saturation. Unfortunately, as illustrated in figure 2, the
`extinction coefficients for COHb and MetHb are not zero
`
`in the red and infrared range, and their presence will,
`therefore, contribute to the absorption. Even though the
`definition of functional hemoglobin saturation involves
`only two hemoglobin species (OgHb and Hb), when
`_ MetHb and COHb are present, four wavelengths are re-
`quired to determine either functional or fractional he-
`moglobin saturation.”
`
`10
`
`10
`
`
`
`PULSI'Z OXIMI‘TI'RY
`
`1 0 1
`
`Anesthesiology
`V T0. N0 Ljan 1909
`
`PULSE OxIM 1?:er
`
`Noninvasive oximeters measure red and infrared light
`transmitted through a tissue bed, effeCtively using the fin—
`ger or ear as a cuvette containing hemoglobin. There are
`several technical problems in accurately estimating SaOa
`by this method. FirstI there are many absorbers in the
`light path other than arterial hemoglobin, including skin,
`soft tissue, and venous and capillary blood. The early ox-
`imeters subtracted the tissue absorbance by compressing
`the tiSsue during calibration to eliminate all the blood,
`and using the absorbance of bloodless tissue as the base-
`line. These oximeters also heated the tissue to obtain a
`
`signal related to arterial blood with minimum influence
`of venous and capillary blood.
`Pulse oximeters deal with the effects of tissue and ve-
`
`nous blood absorbances in a completely different way.
`Figure 3 schematically illustrates the series of absorbers
`in a living tissue sample. At the top of the figure is the
`pulsatile or AC. component. which is attributed to the
`pulsating arterial blood. The baseline or DC component
`represents the absorbances of the tissue bed, including
`venous blood, capillary blood, and nonpulsatile arterial
`blood. The pulsatile expansion of the arteriolar bed pro-
`duces an increase in path length (see equation lb), thereby
`increasing the absorbance. All pulse oximeters assume that
`the only pulsatile absorbance between the light source
`and the photodetector is that of arterial blood. They use
`two wavelengths of light: 650 nanometers (red) and 940
`nanometers (near infrared). The pulse oximeter first de—
`termines the AC component of absorbance at each wave-
`length and divides this by the corresponding DC com-
`ponent to obtain a “pulse-added" absorbance that is in-
`dependent ofthe incident light intensity. it then calculates
`the ratio (R) of these pulse-added absorbances, which is
`empirically related to Sa02:
`
`R : ACsso/Dcsso
`Acgqo/ D6940
`
`(4)
`
`Figure 4 is an example of a pulse oximeter calibration
`curve.”5 The actual curves used in commercial devices
`
`are based on experimental studies in human volunteers.
`Note that when the ratio of red to infrared absorbance
`
`is one, the saturation is approximately 85%. This fact has
`clinical implications to be discussed later.
`It is a fortuitous coincidence of technology and phys-
`iology that allowed the development of solid-state pulse
`oximeter sensors.” Light emitting diodes (LEDs) are
`available over a relatively narrow range of the electro-
`magnetic spectrum. Among the available wavelengths are
`some that not only pass through skin but also are absorbed
`by both oxyhemoglobin and reduced hemoglobin. For
`best sensitivity, the difference between the ratios of the
`absorbances of OgHb and Hb at the two wavelengths
`
`»W i
`
`“c
`
`Absorption clue to pulsatiie anerial blood
`g
`Absorption due to non-pulsalile arterial blood
`2
`' 0 Absorption clue to venous and capillary blood
`..
`E E.
`r w "v r
`2* "
`”3.0.03.9...0'9
`A i
`‘- D“ m «’0‘
`«
`i
`”4
`ow
`iittoiotoioioioi
`Time
`
`Absorption clue to tissue
`
`FIG. 3. This figure schematically illustrates the light absorption
`through living tissue. Note that the AC signal is due to the pulsatile
`component of the arterial blood while the DC signal is comprised of
`all the nonpulsatile absorbers in the tissue; nortpulsatile arterial blood,
`venous and capillary blood, and all other tissues. Adapted from Ohnteda
`Pulse Oximeter Model 3700 Service Manual. 1986, p. 22.
`
`should be maximized. As we see in figure 2, at 660 nano-
`meters. reduced hemoglobin absorbs about ten times as
`much light as oxyhemoglobin. (Note that the extinction
`coefficients are plotted on a logarithmic axis.) At the in-
`frared wavelength of 940 nanometers, the absorption
`coefficient of OgHb is greater than that of Hb.
`
`Engineering Design and Physiologic Limitations
`
`. Although the theory on which pulse oximetry is based
`is relatively straightforward. the application of this theory
`to the production ofa clinically useful device involves a
`
`0.4 0.6 0.8 1.0 1.2 1.4 1.6 1.3 2.0 2.2 2.4 2.6 2.8 3.0 3.2 3.4
`
`H = —
`A Caring D C 560
`Acetoch one
`
`FIG. 4. This is a typical pulse oxinteter calibration curve. Note that
`the $30! estimate is determined from the ratio (R) of the pulse-added
`red absorbance at 660 nanometers to pulse-added infrared absorbance
`at 940 nanometers. The ratios of red to infrared absorbance-5 vary
`from approximately .4 at 100% saturation to 3.4 at 0% saturation.
`Note that the ratio of red to infrared absorbance is one at a saturation
`
`of approximately 85%. This curve can be approximately determined
`on a theoretical basis but, for accurate predictions of SpOg. experi-
`mental data are required. Adapted frontjA Pologe: Pulse oximetry:
`Technical aspects of machine design, International Anesthesiology
`Clinics. Advances in Oxygen Monitoring. Edited by Tremper KK,
`Barker S]. Boston, Little. Brown and Company, 1987. p 142.
`
`11
`
`11
`
`
`
`102
`
`K. K. TREMPER AND S. j. BARKER
`
`Anesthesiolo
`V w. No 1.1m 1939
`
`significant engineering effort. This section will present in
`general terms the clinical and physiologic problems of
`oximeter design and their engineering solutions. The dis-
`cussion is divided into four areas: dyshemoglobins and
`dyes, LED center wavelength variability, signal artifact
`management, and accuracy and response. The reader
`should be aware that these problems can interact with
`one another.
`
`DYSHEMOGLOBINS AND DYES
`
`Being two-wavelength devices, pulse oximeters can deal
`with only two hemoglobin species. As noted above, this
`would be adequate to measure functional 5302 if MetHb
`and COHb did not absorb red or infrared light at the
`wavelengths used. Unfortunately, this is not the case, and
`therefore both MetHb and COHb will cause errors in the
`
`pulse oximeter reading. It is not intuitively obvious how
`a pulse oximeter will behave in the presence of dyshein-
`oglobins. With respect to carboxyhemoglobin, we can gain
`some insight from the extinction curves of figure 2. In
`the infrared range (940 nm), COHb absorbs very little
`light: whereas, in the red range (660 nm), it absorbs as
`much light as does OgHb. This is clinically illustrated by
`the fact that patients with carboxyhemoglobinemia appear
`red. Therefore, to the pulse oximeter, COHb looks like
`021'”) at 660 nm; while, at 940 nm COHb is relatively
`transparent. The effect of COHb on pulse oximeter values
`has been evaluated experimentally in dogs.” In this study,
`the pulse oximeter saturation (SpOg) was found to be given '
`approximately by:
`
`OgHb
`
`+ . X
`09 C0Hb><lOO%.
`total Hb
`
`Spof :
`
`(5)
`
`The effects of methemoglobinernia on pulse oximetry
`are also partially predictable from the extinction curves
`(fig. 2). MetHb has nearly the same absorbance as reduced
`hemoglobin at 660 nm, while it has a greater absorbance
`than the other hemoglobins at 940 nm. This is consistent
`with the clinical observation that methemoglobinemia
`produces very dark, brownish blood. Thus, it would be
`expected to produce a large pulsatile absorbance signal
`at both wavelengths. The effect of Metl-Ib on pulse ox-
`imeter readings has also been measured in dogs.13 As
`methemoglobin levels increased, the pulse oximeter sat-
`uration (SpOg) tended toward 85% and eventually became
`almost independent of the actual 53.09 .1“ In other words,
`in the presence of high levels of MetHb, SpOg is erro-
`neously low when SaOE is above 85%, and erroneously
`high when 83.02 is below 85%. This may be explained by
`the fact that MetHb causes a large pulsatile absorbance
`at both wavelengths, thereby adding to both the numer-
`ator and denominator of the absorbance ratio R (equation
`4) and forcing this ratio toward unity. As shown in figure
`
`12
`
`4, an absorbance ratio of one corresponds to a saturation
`of 85% on the calibration curve. Pulse oximeter error
`
`during methemoglobinemia has also been confirmed in a
`clinical report.19
`In neonatal blood, a fifth type of hemoglobin is present,
`fetal hemoglobin (HbF). HbF differs from adult Hb in
`the amino acid sequences of two of the four globin sub.
`units. Adult Hb has two a- and two Jfl-globin chains, while
`HbF has two a and two f chains. This difference in globin
`chains has little effect on the extinction curves and there-
`
`fore should not affect pulse oximeter readings.§'1l This is
`indeed fortunate because the fraction of HbF present in
`neonatal blood is a function of gestational age and cannm
`be accurately predicted. HbF does produce a small error
`in in vitro laboratory oximeters; OgHbF may be inter-
`preted as consisting partially of COHb.“
`The absorbance ratio R (equation 4) may be affected
`by any substance present in the pulsatile blood that absorbs
`light at 660 or 940 nm and was not present in the same
`concentration in the volunteers used to generate the cal-
`ibration curve (fig. 4). Intravenous dyes provide a good
`example of this principle.21 '22 Scheller at all. evaluated the
`effects of bolus doses of methylene blue, indigo carmine,
`and indocyanine green on pulse oximeters in human vol-
`unteers?‘ They found that methylene blue caused a fall
`in SpOg to approximately 65% for 1—2 min. Indigo car~
`mine produced a very small drop in saturation, while in-
`clocyanine green had an intermediate effect. The detec-
`tion of intravenous dyes by pulse oximeters should not
`be surprising. because it was this effect that led Aoyagi
`to the invention of pulse oximetry.2
`
`LED CENTER WAVELENGTH VARIABILITY
`
`The LEDs used in pulse oximeter sensors are not ideal
`monochromatic light sources: there is a narrow spectral
`range over which they emit light. The center wavelength
`of the emission spectrum varies even among diodes of the
`same type from the same manufacturer. This variation
`can be 3:15 nanometers.” As seen in figure 2, a shift in
`LED center wavelength will change the measured ex-
`tinction coefficient and thus produce an error in the sat-
`uration estimate. This source wavelength effect will be
`greatest for the red (660 nm) wavelength, because the
`extinction curves have a steeper slope at this wavelength.
`Manufacturers have found