Pulse oximeters estimate blood oxygen saturation by shining two wavelengths of light through tissue and measuring how much each wavelength is absorbed, but dozens of factors can skew that measurement. Skin pigmentation, nail polish, cold fingers, movement, carbon monoxide exposure, anemia, and even the type of device you use all influence the number on the screen. Some of these errors are small enough to ignore; others can be dangerously misleading in exactly the situations where accurate readings matter most.
Skin Pigmentation
Pulse oximeters work by comparing how red and infrared light pass through living tissue. Melanin, the pigment that determines skin tone, absorbs light too, and it does not absorb both wavelengths equally. In people with darker skin, the overall light absorption in the outer layer of skin is higher, and the way light scatters through tissue also changes. Photons tend to scatter more in a forward direction in darker skin, which shifts the ratio the device uses to calculate oxygen saturation. Because most pulse oximeters are calibrated on study populations that skew lighter-skinned, the built-in algorithms may not account for these optical differences well enough, leading to readings that overestimate true oxygen levels in people with darker pigmentation.1Physiological Measurement. A review of the effect of skin pigmentation on pulse oximeter accuracy
This is not a theoretical concern. During the COVID-19 pandemic, several studies highlighted that Black and Hispanic patients were more likely to have falsely reassuring pulse oximeter readings, sometimes masking dangerously low oxygen levels that would have prompted treatment. The root issue is not that darker skin makes pulse oximetry impossible, but that the calibration curves baked into most commercial devices were developed with insufficient diversity. Regulatory bodies have since pushed for more inclusive testing standards, but many devices currently in use predate those guidelines.
Nail Polish, Acrylic Nails, and Other Coatings
Because fingertip probes shine light through the nail bed, anything coating the nail can absorb or scatter that light before it reaches the detector. Classic research found that black, blue, and green nail polish lowered saturation readings noticeably, with blue and green causing larger drops than purple or red. The degree of error correlated strongly with how differently the polish absorbed the two wavelengths the oximeter uses.2PubMed. The effect of nail polish on pulse oximetry A more recent systematic review across 20 studies confirmed that dark-colored polishes, particularly black, blue, brown, and purple, produced small but statistically real drops in displayed saturation. One study in that review reported that the oximeter failed to get a reading at all from fingers with black polish nearly nine times out of ten.3PubMed Central. Impact of Fingernail Polish on Pulse Oximetry Measurements: A Systematic Review
In practice, light-colored and sheer polishes are less likely to cause problems. If you need an accurate reading quickly and are wearing dark polish, removing it from one finger or placing the probe sideways across the finger (so light passes through the nail bed from side to side rather than top to bottom) can reduce the interference. Thick acrylic or gel nails create a similar issue by adding an extra layer of material between the light source and the blood vessels.
Cold Hands and Poor Circulation
Pulse oximeters do not just measure light absorption; they detect the tiny pulse of blood expanding your arteries with each heartbeat and use that pulsation to separate arterial blood from everything else. When blood flow to your fingertips drops, the pulsatile signal gets weaker, and the device has less to work with. Cold fingers are the most common culprit. Research has shown that cold skin temperatures significantly reduce the quality of the light signal the sensor picks up, because vasoconstriction shrinks blood volume in the fingertip and weakens the pulse wave the device relies on.4Biomedical Signal Processing and Control. Analysing the effects of cold, normal, and warm digits on transmittance pulse oximetry The same issue crops up at high altitude, where cold ambient temperatures combine with lower oxygen levels to compound the problem.5PubMed Central. The Use of Pulse Oximetry in the Assessment of Acclimatization to High Altitude
It is not just cold weather. Low blood pressure, shock, peripheral artery disease, and medications that constrict blood vessels all reduce perfusion to the extremities. In these situations the oximeter may display erratic numbers, give no reading at all, or quietly become less accurate without any warning on the screen. Some newer devices display a “perfusion index” that tells you how strong the pulse signal is, which gives you a sense of whether the reading should be trusted. Warming the hand or switching to a better-perfused site like the earlobe can help when fingertip readings look unreliable.
Movement and Motion Artifacts
Shivering, fidgeting, or even tapping your finger while wearing a probe can fool the sensor into misidentifying noise as a real pulse signal. The device expects a clean, rhythmic waveform driven by the heartbeat, and any additional movement creates competing signals. A study characterizing motion artifacts in 350 patients found that twitching and shaking (including shivering and tremors) were common, though they tended to affect readings less severely than other types of motion because the disturbances were either low in amplitude or short-lived.6Anesthesia & Analgesia. A characterization of motion affecting pulse oximetry in 350 patients Larger, sustained movements, like a restless patient shifting their hand, tend to cause bigger and longer-lasting errors.
Modern oximeters have signal-processing algorithms designed to filter out motion noise, and they are considerably better at it than devices from a decade or two ago. Still, no algorithm is perfect. If you are checking your own saturation at home and the number seems off, holding your hand still on a flat surface for 15 to 30 seconds and waiting for the reading to stabilize is the simplest fix.
Ambient Light
Pulse oximeters work by detecting tiny amounts of light that pass through your tissue, which means bright external light reaching the sensor’s detector can contaminate the signal. Surgical lights, heat lamps, direct sunlight, and even some fluorescent fixtures have all been documented as sources of interference.7PubMed. Potential errors in pulse oximetry. III: Effects of interferences, dyes, dyshaemoglobins and other pigments Heat lamps, in particular, produce intense broadband light that can overwhelm the sensor entirely. The fix is straightforward: shielding the probe from external light with an opaque cover or simply repositioning the sensor away from the light source eliminates the problem in most cases.8PubMed Central. Optical crosstalk and other forms of light interference in pulse oximeter comparison studies
Electrosurgery equipment in operating rooms can also introduce electromagnetic interference that shows up as erratic readings. This is less relevant outside a hospital setting but worth knowing if you are a clinician puzzling over impossible-looking numbers during a procedure.
Carbon Monoxide Poisoning
This is arguably the most dangerous limitation of standard pulse oximetry. Carbon monoxide binds to hemoglobin about 200 times more readily than oxygen does, forming carboxyhemoglobin. The problem is that carboxyhemoglobin absorbs red light almost identically to oxyhemoglobin, so a two-wavelength pulse oximeter cannot tell them apart. In one landmark study, as carboxyhemoglobin levels rose and actual oxygen-carrying hemoglobin fell below 30%, both pulse oximeters tested continued to display saturations above 90%.9PubMed. The effect of carbon monoxide inhalation on pulse oximetry and transcutaneous PO2
Subsequent work confirmed the pattern: pulse oximeters failed to drop below 96% even when carboxyhemoglobin levels reached 44%, and the size of the overestimate roughly equaled the amount of carboxyhemoglobin present.10PubMed. The pulse oximetry gap in carbon monoxide intoxication This “pulse oximetry gap,” the difference between the displayed number and the true oxygen saturation, means that a patient who looks fine by pulse oximeter could be profoundly oxygen-deprived. The reading may even appear normal in heavy smokers, who carry chronically elevated carboxyhemoglobin.11PubMed Central. Accuracy of Carboxyhemoglobin Detection by Pulse CO-Oximetry During Hypoxemia Specialized CO-oximeters that use additional wavelengths of light can distinguish carboxyhemoglobin from oxyhemoglobin, but the standard finger-clip devices used in most homes and many clinical settings cannot.
Methemoglobin, another abnormal form of hemoglobin caused by certain medications and chemical exposures, also interferes with readings, though it has a different spectral signature and tends to push the displayed value toward the mid-80s regardless of the patient’s actual saturation.
Anemia
Pulse oximeters measure the percentage of hemoglobin that is carrying oxygen, not the total amount of oxygen in the blood. In theory, an anemic patient could show a perfectly normal saturation of 98% while still delivering far less oxygen to tissues than someone with a full complement of red blood cells. That is already a conceptual limitation, but the accuracy of the reading itself also degrades when hemoglobin drops low enough.
Research found that at oxygen saturation levels around 54%, the error in pulse oximeter readings increased roughly linearly as hemoglobin fell, going from essentially zero error above 14 g/dL to about a 14-percentage-point underestimation at hemoglobin levels of 8 to 9 g/dL.12PubMed. Effect of anemia on pulse oximeter accuracy at low saturation At near-normal saturations (around 97%), anemia did not produce measurable error. In other words, severe anemia and low oxygen levels compound each other to make the oximeter less reliable exactly when accuracy matters most. Laboratory testing using artificially diluted blood confirmed that cheaper fingertip devices were the least accurate under severe anemia, while higher-end benchtop instruments held up better.13PubMed Central. Quantifying pulse oximeter accuracy during hypoxemia and severe anemia using an in vitro circulation system
Where You Place the Sensor
Not all body sites give equally reliable readings. Finger probes are the most common and, under most conditions, the most accurate. A comparison of finger, ear, nose, and forehead probes found that under poor-perfusion conditions, finger probes outperformed the alternatives. Only two out of eight ear probes, and none of the nose or forehead probes tested, were expected to stay within 4% of the reference value 95% of the time.14PubMed. Pulse oximeter probes. A comparison between finger, nose, ear and forehead probes under conditions of poor perfusion
There are situations where the finger is not the best option. During surgery on the hand, when fingertip access is blocked, or when peripheral perfusion is so poor that the finger gives no signal, ear probes may respond faster to changes in oxygen levels because the earlobe has a relatively rich blood supply and is closer to the central circulation. Forehead sensors, despite performing poorly in the study above, have been refined in more recent designs and are sometimes used in operating rooms. The key point is that switching probe sites is not a neutral move: each location has its own accuracy profile, and the device’s calibration may or may not be validated for that site.
The Ceiling Effect Above 97%
Pulse oximeters are designed to detect dangerously low oxygen, not to distinguish between adequate and excessive oxygenation. Once the displayed saturation reaches about 97 to 100%, the device effectively hits a ceiling. A patient breathing supplemental oxygen could have an arterial oxygen tension of 100 mmHg or 500 mmHg, and the pulse oximeter would show the same number in both cases.15PubMed Central. The oxygen reserve index (ORI): a new tool to monitor oxygen therapy This matters because excessive oxygen (hyperoxia) carries its own risks, including lung injury and, in premature infants, retinal damage. Clinicians managing supplemental oxygen need arterial blood gas measurements or newer tools like the oxygen reserve index to titrate oxygen safely once saturation is in the high-normal range.
Venous Pulsation and Heart Valve Problems
Pulse oximeters assume that the only thing pulsating in the tissue is arterial blood. In some heart conditions, that assumption breaks down. Severe tricuspid regurgitation, where blood flows backward through the valve between the right side of the heart, creates pulsation in the venous system. The oximeter may then pick up venous blood, which is lower in oxygen, as part of its signal. A study of patients with severe tricuspid regurgitation found that pulse oximeter readings were consistently lower than laboratory-measured values, with discrepancies as large as 11 percentage points.16PubMed. Inaccuracy of pulse oximetry in patients with severe tricuspid regurgitation Severe right-sided heart failure, tight-fitting blood pressure cuffs cycling on the same arm, and even some mechanical ventilator settings can create similar venous pulsation effects.
Intravenous Dyes and Medications
Certain dyes used in medical procedures absorb light at wavelengths close to those used by pulse oximeters. Methylene blue is the classic example: injecting it intravenously causes an immediate and often dramatic drop in the displayed saturation, sometimes plunging the reading into the 80s or lower. In most cases, this is a spurious artifact that resolves within a few minutes as the dye is cleared from the bloodstream. However, methylene blue can also cause real physiological desaturation in some patients, meaning clinicians cannot simply dismiss the drop as artifactual every time.17PubMed Central. Oxygen desaturation following methylene blue injection: Not always spurious Indocyanine green and patent blue dye, used in cardiac testing and sentinel lymph node biopsies, can cause similar transient interference.
Consumer Wearables Versus Medical Devices
Smartwatches and fitness trackers increasingly offer blood oxygen readings, but the technology differs in important ways from a medical fingertip probe. Most wearables use reflectance oximetry, bouncing light off the top of the wrist rather than shining it through tissue. The wrist is not an ideal measurement site: the arteries are deeper, there is more tissue and bone between the sensor and the blood, and the device sits on a part of the body that moves constantly. Research investigating common consumer smartwatches found that their accuracy varied and that factors like fit, skin tone, and motion during measurement introduced additional uncertainty.18PLOS Digital Health. Investigating the accuracy of blood oxygen saturation measurements in common consumer smartwatches
These devices are useful for spotting trends over time, like noticing that your nighttime oxygen levels dip consistently, which might prompt a conversation with your doctor about sleep apnea. They are not reliable enough to make clinical decisions. If a smartwatch reading worries you, confirm it with a medical-grade fingertip pulse oximeter or, better yet, a healthcare provider. The FDA clears medical pulse oximeters to a defined accuracy standard (typically within 2 to 3 percentage points of the true value across a saturation range of about 70 to 100%), while most consumer wearables are marketed as wellness devices and do not undergo the same validation.
Altitude and the Environment
At high altitude, arterial oxygen saturation genuinely drops because there is less oxygen in the air. That is not an error; the pulse oximeter is doing its job. The complication is that the environmental conditions accompanying altitude, especially cold temperatures, can simultaneously degrade the device’s accuracy by reducing finger perfusion. A hiker at 4,000 meters with cold, blue fingers and a reading of 82% might actually be at 85% or 78%; the uncertainty grows when the signal quality is poor. Warming the measurement site before taking a reading, using a device that reports a perfusion index, and taking multiple readings over a few minutes are all recommended strategies for getting a trustworthy number in the field.5PubMed Central. The Use of Pulse Oximetry in the Assessment of Acclimatization to High Altitude
When Anatomy Itself Gets in the Way
The pulse oximeter assumes that the arterial blood supply to the finger is intact and flowing normally. In most people, the hand has two main arteries feeding overlapping networks, so blocking one does not cut off supply. But anatomical variants exist. A study using pulse oximetry alongside ultrasound to assess the palmar arches found that about 4% of apparently healthy participants had an abnormal oximetry response when one artery was compressed, suggesting an incomplete arterial arch.19European Journal of Vascular and Endovascular Surgery (EJVES Vascular Forum). Duplex Ultrasound and Pulse Oximetry Increase a Diagnostic Probability of Incomplete Palmar Arch Peripheral vascular disease, Raynaud’s phenomenon, and arterial grafts can all alter blood flow patterns in the hand enough to affect readings, either by weakening the pulsatile signal or by creating conditions where the oximeter misidentifies the source of pulsation.
For patients with known vascular disease in the extremities, clinicians often try multiple fingers or switch to an earlobe probe. The goal is the same: find a site with a strong, clean pulse wave so the device has reliable data to work with.