Ifirst became acquainted with pulse oximeters the hard way — in a hospital Emergency Rooma few years ago. My wife and I were in the midst of a brief mini-vacation in theCalifornia High Sierras when I fell ill. After I spent a horrible sleepless night in thehotel, feeling nauseous and dizzy, my wife made a command decision and drove me to the ER,despite my protestations that I really didn’t need to see a doctor.
When I arrived at the hospital, the ER doc took one look at me,told me to lie down, and clipped some sort of probe onto my finger. The probe wasconnected by an electrical cable to a box with a digital readout on it, and the physiciannoted some readings from the instrument onto my chart. Then he stuck a cannula up my noseand started administering oxygen therapy.
The doctor told me that I was profoundly hypoxic — he classified my condition as cyanotic(meaning I was actually starting to turn blue) — and said that the instrument connectedto the finger clip probe was a pulse oximeter that was reading the percentage ofoxygen saturation in my arterial blood. Readings of 95% to 100% are normal at sea level,he explained, and even at the 8,000′ elevation of Mammouth Lakes, the oxygen saturationfor a flatlander like me should still be 90% or so. But when he first hooked me up, thereading was 75% — a severely hypoxic state that was literally life-threatening if it hadbeen left untreated for a few more hours.
After listening to my lungs with a stethescope, the ER doc diagnosed HAPE — HighAltitude Pulmonary Edema — which is a form of altitude sickness in which fluid startscollecting in the alveoli (tiny sacs) of the lung tissue, interfering with the lungs’ability to oxygenate the blood. The only remedy was for me to be transported immediatelyto lower altitude (while continuing to breathe supplemental oxygen) and stay there for 24to 48 hours until the fluid was reabsorbed and my lungs started to function normally onceagain.
My wife drove us down to Bishop, Calif., at 4,000-foot MSL and I checked into theBishop Hospital ER, where I was hooked to another pulse oximeter and pronounced out ofdanger. I was instructed to stay overnight in Bishop, remain on supplemental O2, andreturn the next day for another check. When I returned the next morning, my oxygensaturation readings were in the 90s and I was given a clean bill of health.
Oximetry for in-cockpit use?
Although my encounter with HAPE was scary and messed up my vacation, I was certainlyfascinated by the pulse oximeter technology that had been used to measure the oxygensaturation of my blood. As a pilot who does quite a lot of high-altitude flying inunpressurized airplanes, I was intrigued by the notion of carrying such an instrumentalong in the cockpit and using it to monitor my hypoxia level and that of my passengers.Such instrumentation would let me know precisely when I needed to start usingsupplemental oxygen, and precisely what O2 flow rates were necessary to preventimpairment or discomfort. I decided that I wanted one of these instruments for myairplane.
I did some checking with medical supply houses to see whether it might be possible topurchase such an instrument for in-flight use. The results were not encouraging. Theleast-expensive pulse oximeter I could find was rather bulky — about the size of abattery charger — and cost about $2,000. Furthermore, the distributors were notcomfortable selling such a device to anyone but a doctor or hospital, and told me that atthe very least I’d require a doctor’s prescription.
While discussing my HAPE scare with Dr. Brent Blue — a good friend and Senior AviationMedical Examiner based in Jackson Hole, Wyoming — I mentioned the idea of using aportable pulse oximeter in the cockpit. Brent thought it would be an excellent idea, andoffered to write me a presecription if I wanted to buy one. But $2,000 seemed awfullysteep, so I shelved the idea.
Enter the Nonin Onyx
Then, a few months ago, Dr. Blue sent me a “bingo card” from one of hismedical journals that advertised a new, micro-miniature pulse oximeter from Nonin Medical,Inc., in Plymouth, Minnesota. (In case you were wondering, the company’s name is short for”non-invasive.”) Through the magic of state-of-the-art custom integrated circuittechnology, Nonin Medical has managed to squeeze the pulse oximeter electronics into thefinger-clip probe itself, creating a self-contained finger pulse oximeter called the NoninOnyx . The Onyx is incredibly tiny (1.3″ x 1.3″ x 2.2″), weighs only twoounces, and is powered by two AAA alkaline batteries. Best of all, the Nonin Onyx sellsfor less than $400, clearly within the realm possibility for individual pilot use.
My interest was definitely rekindled. I contacted Nonin and told them of my interest inusing their little pulse oximeter for in-flight monitoring of flightcrew hypoxia levels.Surprisingly, they seemed uninterested in talking to me. It seems that oximeters areFDA-approved medical devices, and Nonin’s policy is to offer them for sale only tohospitals and physicians. Interestingly enough, the company does sell their oximeters toveterinarians for use on animals. Pilots, however, are not on their radar screen.According to Nonin, the U.S. Food and Drug Administration (FDA) dictate these sillyrestrictions.
When I told Dr. Blue that I wasn’t getting anywhere with Nonin, he volunteered to callthem on my behalf. A few weeks later, a Nonin Onyx showed up on my doorstep. I was anxiousto put it through its paces and see how it worked in the cockpit.
Flight-testing the Nonin Onyx
I started out by familiarizing myself with the unit on theground. Operation of the Onyx the ultimate in simplicity. Unlike the big hospital unitsI’d seen before, the Onyx has no switches or controls. You simply clip the unit to thefingertip of your choice and the thing automagically turns itself on. After a few seconds,the “perfusion display” LED starts blinking in sync with your pulse. The colorof the blinking LED is green, yellow or red, indicating whether the unit is detectinggood, marginal or inadequate pulse amplitude. (If the indication is yellow or red, simplyreposition the clip or change to a different finger.)
After a few heartbeats, the two numeric LED displayslight up. The top number — labeled “%SpO2” — shows the percentage of oxygensaturation of your arterial blood, normally a figure between 95% and 100% at sea level,and progressively less at higher altitudes. The bottom number — labeled with a littleheart symbol — shows your pulse rate in beats per minute.
You can monitor yourselfcontinuously in flight, or if you prefer you can conserve batteries by doing periodicspot-checks throughout the flight. A pair of AAA alkaline batteries is good for 12 hoursof continuous operation, or about a thousand 45-second spot checks. The unit shuts itselfoff automatically ten seconds after you remove it from your finger, so there’s no chanceof accidentally running down the batteries because you forgot to turn the thing off. Ifthe batteries get low, the numeric displays start to flash once per second to warn you,but the unit continues to function normally for quite a while after that.
My initial flight tests with the unit produced some encouraging results, but also somepuzzling ones. At sea level, the readouts showed the oxygen saturation of my arterialblood to be normal (97% to 98%). And, just as I expected, I could see my O2 saturationgradually decline toward 90% (roughly the onset of measurable impairment) as the airplaneclimbed through 6,000 to 8,000 feet.
The instruction manual had cautioned that the oximetermeasurements might be affected by “excessive or rapid movement” and”fluctuating or flickering light.” So I tried intentionally subjecting the Onyxto in-flight vibration by putting it in contact with various portions of the aircraft (theglare shield, the windows, etc.), and to expose it to various light levels. Nothing I didseemed to have any effect on the readings, however, and I concluded that only slow,rhythmic movement or light pulses (that the unit could falsely interpret as a pulse) wouldbe a problem.
I found the LED digital displays to be somewhat difficult to read in bright, directsunlight. However, there was no problem reading them while my hand was in my lap, on thethrottle, or anywhere else that was shielded from direct sunlight by the instrument panelor glare shield. I also did some night flying with the Onyx, and found its displays to bevery pleasant to read (and not distractingly bright as I had feared).
Verifying the readings
The puzzling results came as I climbed above 10,000 feet into “hypoxiaterritory.” The oximeter’s saturation readings decreased into the 80s, as expected.But they also started to vary quite a bit, especially when I climbed to the 12,000 and13,000 foot levels, at which point the oscillations became quite pronounced. My initialassumption was that the instrument was not functioning correctly at these higher altitudesand lower O2 saturation levels.
To find out for sure, I madearrangements with Dr. Blue to borrow a couple of clinical pulse oximeters from a localhospital, one a suitcase-sized unit that cost about $5,000, and the other a smaller modelthat cost around $2,000. Brent and I set up these units in my Cessna T310 — the big onein the back seat and the smaller one between the seats — and the two of us took off on atest flight to compare results from the two hospital oximeters with those from my littleNonin Onyx. Shortly after takeoff, I put the airplane on autopilot and hooked my righthand to all three pulse oximeters, the three finger-clip probes clipped to three differentfingers. (Boy, did that feel weird!)
We step-climbed all the way up to 17,000 feet, leveling off at each 1,000-foot altitudeand noting the readings of the three oximeters. All three O2 saturation readingsconsistently agreed within one percentage point, indicating that the inexpensive littleOnyx was every bit as accurate as the big, expensive hospital units. Interestingly enough,as we climbed above 10,000 feet, all three started to oscillate — the readings from thethree pulse oximeters remained in very close agreement, and the oscillations on all threeunits were perfectly synchronized.
NOTE: We later discovered that the oscillations were not an instrumentationartifact, but a genuine oscillation of my blood oxygen saturation level at altitude. Thisfascinating and potentially important finding is the subject of a companion article:
As we climbed above 14,000 feet, Brent and Idonned conserving cannulas and turned on the flow of supplemental oxygen. After levelingat 17,000 feet, we started experimenting with various oxygen flow rates (using acalibrated vernier flowmeter) to determine the impact on oxygen saturation. We found thatwithout supplemental oxygen, O2 sat readings decreased into the mid-70s (extremeimpairment), but that saturation could be brought up to a very acceptable level (low 90s,equivalent to a physiological altitude of 6,000 or 8,000 feet) by using an extremely lowO2 flow rate (0.5 liter/minute or less). Turning up the flow rate had no significantbeneficial effect whatever.
Based on our oximeter results, we concluded that it would be prudent to start usingsupplemental oxygen at altitudes well below what the FARs require. O2 sat levelsdefinitely drop to those associated with measurable impairment at cabin altitudes as lowas 10,000 feet, and supplemental O2 is probably warranted at altitudes as low as 6,000 or8,000 feet if you’re a smoker or if you have respiratory congestion (e.g., a cold) oremphysema or any number of other conditions that can reduce the efficiency of yourrespiratory system. Using a conserving cannula and a vernier flowmeter, together with apulse oximeter, it’s often possible to get by with drastically lower flow rates ofsupplemental oxygen than those that are normally used. On the other hand, there are timeswhen the O2 flow will need to be turned up. The oximeter makes it possible to administerprecisely the required amount of O2 needed for your altitude and physical condition.
All in all, the Onyx performed astonishingly well alongside thelarge and expensive hospital oximeters. This test flight gave us great confidence in theaccuracy of the Nonin instrument, despite its tiny size and relatively low cost.
About my only gripe with the Nonin Onyx is the fact that its digital readout isdesigned for attended-care monitoring of a patient, and is oriented “upsidedown” for self-monitoring. In order to monitor your own readings, you must either (1)turn your palm up and bend your instrumented finger toward your palm, or (2) learn to readthe display upside-down. In my early trials with the Onyx, I used the finger-bend method,but after using the instrument for awhile, I found that the upside-down-reading method wasjust as easy. If the use of pulse oximeters in the cockpit becomes popular, perhaps we canpersuade Nonin to produce a model with an inverted display. In the meantime, it’s not aserious problem, just a mild annoyance.
Theory of operation
The technology that makes possible the non-invasive measurement of the oxygensaturation of arterial blood is quite fascinating — simple in concept, but complex inexecution. Pulse oximetry takes advantage of the fact that blood changes color dependingon whether the hemoglobin in the red blood cells are oxygenated or deoxygenated.Oxygenated blood is bright red, while deoxygenated blood is dark red in color, borderingon purple. It is therefore possible to deduce the degree of oxygen saturation of bloodfrom its color. The basic idea behind a pulse oximeter is to shine a red light through avascular bed (such as a fingertip or earlobe) and measure how much of the red light isabsorbed. Dark-red or purple deoxygenated blood will absorb most of the red light, while bright-red, oxygenated blood will allow the red light to pass right through.
Unfortunately, there lots of other things that affect how much red light is absorbed:finger thickness, skin thickness, skin pigmentation, bone thickness, fingernail thickness,the presence of nail polish, etc. To account for these things, the pulse oximeter shines asecond light beam of different color through the finger (actually, an infrared beam) andmeasures the differential attenuation of the two wavelengths. Extraneous sources of lightattenuation are thus cancelled out, allowing accurate determination of red lightabsorption by the blood relatively unaffected by finger-to-finger variations.
This still leaves the problem of how the pulse oximeter can distinguish betweenarterial blood (which is what we want to measure) and venous blood (which is alwaysdeoxygenated). Here’s where the “pulse” part comes into play. The oximeter takesadvantage of the fact that arterial blood flows in pulses, while venous blood flow issteady. By locking onto the pulse and measuring only the differences in red lightabsorption between the high and low points of pulse fluctuations (systolic vs. diastolic),the oximeter is able to cancel out the effects of steady venous flow and measure only thecolor of the pulsating arterial blood.
All this magic is done through complex digital signal processingby a tiny microprocessor and sophisticated software, using elaborate exponential smoothingand empirical approximation algorithms to come up with a reading that correlates veryclosely with the results of laboratory blood gas analysis over a wide range of conditions.
The technology involved is really cool. The fact that this technology can now bereduced to a two-ounce gizmo that slips into your shirt pocket is nothing short ofremarkable.
Limitations of pulse oximetry
From my perspective as a pilot, perhaps the most important limitation of pulseoximeters is that they will not detect carbon monoxide (CO) poisoning. When CObinds with the hemoglobin in your blood, the cells turns bright red…just as if they hadbeen oxygenated. The resulting molecule (carboxyhememglobin) is incapable of carrying O2to your cells, but is indistinguishable in color from oxygenated blood so far as the pulseoximeter is concerned. Consequently, it’s important for pilots to carry a
NOTE: AVweb has completed an extensive investigation into CO monitors for aircraft use. We do not recommend the inexpensive chemical-spot detectors available from most pilot supply stores.
By the same token, a pilot or passenger who smoked a cigarette shortly before takeoffwill tend to be more hypoxic than the pulse oximeter readings indicate, because smokingcauses CO poisoning which the oximeter cannot detect.
A second limitation is that pulse oximeters tend to become inaccurate at extremely lowlevels of oxygen saturation (below about 75%). This is not a serious problem for in-flightuse, because accuracy at those levels is not important. Who cares whether your O2 sat is76% or 73%? If it’s below 80%, you’re in big trouble, and you better crank up the O2, getdown to a lower altitude, or both…fast!
A third limitation is that the pulse oximeter depends on the presence of a good pulse.People with unusually low blood pressure or impaired blood flow to the fingers may havedifficulty getting valid oximeter readings. Conditions that cause constriction of theblood vessels in the extremities (e.g., cold temperatures, profound hypoxia) can alsointerfere with oximeter readings, as can drugs that are vasoconstrictors or vasodilators(e.g., nitroglycerine), or drugs that affect blood color (e.g., sulfonamides). In mostsuch cases, the oximeter will warn of inadequate perfusion (a yellow or red LED in thecase of the Nonin Onyx). Once again, these problems are not often seen in the cockpit.
Hypoxia and oximetry
It’s one thing to comply with the FARs, but it’s quite another thing to fly safely andcomfortably. Hypoxia is insidious and highly physiologically variable from pilot topilot…and in the same pilot from day to day. The FAA-mandated requirements forsupplemental oxygen use may be too liberal or too conservative, and there is no objectiveway for a pilot to know without using a pulse oximeter to measure the actual level ofoxygen saturation. An overweight, out-of-shape, middle-aged smoker will become hypoxic ata far lower altitude than a young, athletic non-smoker.
At sea level in a healthy person, the oxygen saturation istypically 95% to 100%. At 6,000 feet, the normal oxygen saturation drops to the 90% to 95%range, and continues to decrease as one goes to higher altitudes. As O2 sat decreases intothe 80s, measurable impairment of cognitive and physical performance begins. Those changesdon’t occur immediately, but vary with the speed of ascent and the duration of exposure.
When oxygen saturation levels drop, bad things happen that are rarely perceived by thevictim (at least in the early stages). Visual changes occur, including “tunnelvision” and a marked decrease in night vision. Other common symptoms of hypoxiainclude headaches, anxiety, panic sensation, inability to perform mathematical problemsaccurately, inability to program equipment such as a GPS, dizziness, nausea, headache, andconfusion. Symptoms are different for each person, and can occur at altitudes far lowerthan most people would predict. (An excellent way to get in touch with your own hypoxicsymptoms is to schedule a altitude chamber “ride” at the FAA Civil AeromedicalInstitute in Oklahoma City, or at an Air Force base near you.) Symptoms of hypoxiagenerally remain consistent for a particular person, but the altitude at which the onsetof impairment occurs is highly variable from day to day.
One of the earliest physiological effects of hypoxia is a change in respiration fromsteady to cyclical. This change in involuntary breathing patterns interfere withrespiratory efficiency and exacerbate the hypoxic effect of high-altitude flight. For moreinformation about this fascinating subject, see the companion article
The availability of small, low-cost pulse oximeters suitable foruse in the cockpit provides an enormous leap forward in detecting and dealing within-flight hypoxia. Although not perfect, the pulse oximeter which can be worn on afingertip by both pilot and passengers gives an almost instantaneous oxygen saturationreading. Oximetry has now become so important in the hospital setting that doctors oftenrefer to it as “the fifth vital sign.” (The first four are pulse rate, bloodpressure, respiratory rate, and temperature.)
Dr. Blue offers the following guidelines are offered with respect to pulse oximeter usein the cockpit:
- Always abide by the oxygen-use requirements of the FARs (O2 required above 12,500 feet for more than 30 minutes, and at all times above 14,000 feet), but treat them as minimum requirements that may often be inadequate to prevent hypoxic impairment.
- Always use supplemental oxygen at the first sign of hypoxic symptoms (visual impairment, headache, dizziness, nausea, anxiety, panic, confusion, etc.), and adjust the oxygen flow to alleviate those symptoms.
- If no symptoms of hypoxia are present and a pulse oximeter is available, use supplemental oxygen and adjust the oxygen flow rate to maintain an oxygen saturation level no less than 10 percentage points lower than your stable saturation level at home. (“Home” is defined as the elevation at which you have lived for the majority of time during the past 180 days.) For example, if your home O2 sat reading is typically 97%, use supplemental O2 to maintain an in-flight reading of no less than 87%.
How to get one
Basedon our recent in-flight testing and research, Dr. Brent Blue and I have come to theconclusion that the use of a pulse oximeter in the cockpit is an essential safety measurefor any pilot who flies at cabin altitudes of 10,000 feet and above. We also believeoximeter use is prudent — regardless of cruising altitude — for any pilot who smokes,flies with a cold or cough, or suffers from any other respiratory condition. In short, wethink that most pilots would be wise to carry a pulse oximeter in their flight bag.
Because of the difficulties Iencountered in obtaining my oximeter (medical companies don’t want to talk to you ifyou’re not a doctor or hospital administrator), I asked Dr. Blue to set up adistributorship called Aeromedix.com for the purpose ofmaking the Nonin Onyx pulse oximeter (and other relevant medical products) easilyavailable to pilots at discount prices and with the FDA-required prescription thrown in atno charge. Aeromedix.com has now set up a web page where youcan order the Onyx oximeter online at a discounted price, significantly less than the $425or so that you’d pay at a medical supply house (assuming you had a doctor’s prescriptionso they’d sell one to you). You can also order by telephone (888-362-7123).
In the interests of full disclosure, I should point out that I serve as a paid consultantto Brent’s company, and have been involved in the selection and evaluation of most of theproducts he offers for sale through Aeromedix.com.