Errors and Limitations Associated With Use of Pulse Oximetry to Estimate Limb Occlusion Pressure (LOP)

An interesting article by Lima-Soares et al. [1] was recently published in the Journal of Strength and Conditioning Research.

The study compared a new method to estimate Arterial Occlusion Pressure (AOP) using visual inspection of a pulse oximeter, with an existing method using loss of sound with a handheld Doppler. The authors designed a study whereby AOP was evaluated in young male and female subjects in supine, seated and standing positions. In each individual, AOP was defined as loss of signal with the pulse oximeter and loss of sound with the Doppler equipment. Based on their results, the authors concluded that “the results presented herein strongly suggest the use of the portable pulse oximetry equipment as reliable when compared with handheld Doppler” [1] .

This conclusion is erroneous for several reasons relating to difficulties with the pulse oximeter estimation techniques and inherent limitations of pulse oximetry in relation to AOP. Furthermore, there are specific flaws in the study methodology that may have influenced the results. In response, we provide discussion on each of these aspects below. It is of note to the reader that since the term AOP is misleading, as indicated by McEwen and Hughes [2], the term Limb Occlusion Pressure (LOP) will be used throughout to denote the occlusion of all arterial vessels in a limb.

  1. Contradicting evidence cited in the study

In the discussion section, the authors quote a similar study in which pulse oximetry was compared with the handheld Doppler to estimate arterial circulation in patients with venous disease of the leg.  The authors noted that in this cited study, because of the presence of venous disease, only a “fair agreement between the handheld Doppler and the pulse oximeter was observed”. Hence, the cited study, and the restriction of the Lima-Soares et al. [1] study to LOP estimations in the upper limb, raises concerns with the reliability of pulse oximetry to estimate LOP in the lower limb, and to estimate LOP in patients with venous disease. These concerns are supported by another previous study showing that there was considerable disagreement in AOP estimation between doppler and pulse oximetry in the lower limbs of healthy men and women with no history of any vascular disorder [3].

  1. Venous congestion affecting accuracy of pulse oximeter estimation in the study

LOP was measured by inflating a cuff to 50m mmHg with a manual sphygmomanometer and gradually increasing the pressure in 10 mmHg steps with 5 seconds between each value to allow for improved oximeter visualization.  This slow increase in pressure could have led to venous congestion in subjects which may affect the accuracy of the pulse oximeter LOP estimation [4].

  1. The use of narrow cuffs, not representative of safe BFR practice

Narrow, 5 cm wide nylon cuffs were used in the study.  A narrow cuff requires and applies higher pressures and pressure gradients resulting in greater risk of patient injury [5]. It is unknown whether the conclusions of the study would have been different if cuffs having widths safe for BFR practice were used.

  1. Incorrect representation of LOP accuracy

In the results section, it was stated that the maximal difference found between the pulse oximeter and the Doppler was +/-10 mmHg.  This is misleading because the true LOP difference could be as much as 20 mmHg since each estimation had an accuracy of +0/ -10 mmHg.  For example, a case in which handheld Doppler LOP was determined to be 130 mmHg and pulse oximeter LOP of 120 mmHg, could have a had a true handheld Doppler LOP of 130 mmHg and a true pulse oximeter LOP of 110.1 mmHg, a spread of 20 mmHg.

Hence the conclusion that the maximal difference between the techniques is +/-10 is incorrect and a difference of up to 20 mmHg is possible.

  1. Lack of repeat measurements limiting strength of conclusions

Systolic blood pressure measurements in the study were “determined in duplicate” to confirm accuracy.  However only one pair of LOP estimates was taken at each body position.  The absence of repeated measurements limits the strength of the study conclusions.

  1. Narrow scope of study limiting strength of conclusions

BFR involves LOP determination and exercise on upper and lower limbs and in a range of patient age group and health conditions.  The present study limited LOP estimations to the right arm of subjects, and tested healthy, normotensive, and eutrophic subjects.  Hence, there is insufficient data in the study to make a general conclusion regarding the use of the portable pulse oximetry equipment as reliable when compared with handheld Doppler, particularly in the lower limb and in other patient populations. Furthermore, the study does not shed light on the reliability of pulse oximetry in BFR cuffs of different width or material.

  1. Similar/greater limitations in pulse oximetry compared to handheld Doppler

The pulse oximetry technique does not overcome many limitations existing in the handheld Doppler method.  Both methods can result in error from improper sensor placement, both take a significant amount of time to measure LOP, both require user skill in correct incrementation of cuff pressure and detecting the point at which the pulse is no longer heard/seen. Both techniques can result in erroneously low LOP readings due to a failure to observe delayed pulses near the end of the measurement, and both can overshoot the LOP if cuff pressure is not carefully incremented

Using pulse oximetry to estimate LOP may provide additional limitations compared to the Doppler method. A previous study examined the use of perfusion index to monitor arterial occlusion compared to doppler [6]. The perfusion index is the ratio of pulsatile blood flow to static blood in peripheral tissue. It can be obtained non-invasively from a pulse oximeter and provides a measure of peripheral perfusion. The study by Wall and colleagues [6] found the PI is less accurate than the doppler method, and is an even less accurate indicator of LOP than the standard plethysmograph waveform. This provides further evidence of the inappropriateness of pulse oximeters for determining LOP.

  1. Additional limitations in pulse oximetry related to peripheral circulation

There are a number of limitations with estimating LOP using pulse oximetry that are not present in handheld Doppler LOP estimation.

A pulse oximeter will only function properly if it can detect a modulation in transmitted light.  In cases where perfusion is decreased and amplitude is small the device may provide an erroneous estimation.  A low signal-to-noise ratio in low perfusion states create a signal that can be changed by artifact, thus making it difficult to distinguish between pulsation and background noise [4].

Many factors can cause poor peripheral perfusion, particularly cold or hypotension.  Other factors affecting peripheral perfusion include edema or venous congestion of the limb, peripheral artery disease, low cardiac output, vasoconstriction, and vasoactive drugs (dobutamine or dopamine) [4].

  1. Additional limitations in pulse oximetry related to motion artifact

Motion artifact can also interfere with signal detection and interpretation of the signal by a user because of an unstable waveform. Improperly seated sensors, shivering, or pressure changes in the cuff, can cause movement, creating an inaccurate LOP estimate.  Adjustment of the device to a longer signal averaging time may reduce the effects of motion artifact [4]. However, this may increase estimation time and also reduce accuracy of LOP estimation.

  1. Additional limitations in pulse oximetry related to patient anatomy and physiology

Standard pulse oximeter probes are designed for fingers, not toes, hence it can be difficult to apply pulse oximeter sensors on toes in a way that will provide an accurate LOP estimation.  Furthermore, the pulse amplitudes in lower limb extremities are typically smaller than those seen on the upper limb extremities, making it more difficult to visually detect the point of occlusion on the lower limb compared to the upper limb.  Finally, misshaped and missing digits, dystrophic nails, and nail polish can all prevent pulses from accurately being determined using a pulse oximeter.

References:

[1] Lima-Soares F, Pessoa KA, Torres Cabido CE, Lauver JD, Cholewa J, Rossi F, et al. Determining the Arterial Occlusion Pressure for Blood Flow Restriction: Pulse Oximeter as a New Method Compared With a Handheld Doppler. J Strength Cond Res 2020. doi:10.1519/JSC.0000000000003628.

[2] McEwen J, Hughes L. Pressure prescription for blood flow restriction exercise. Medicine & Science in Sports & Exercise. 2020 Jun 1;52(6):1436.

[3] Zeng Z, Centner C, Gollhofer A, König D. Blood-flow-restriction training: Validity of pulse oximetry to assess arterial occlusion pressure. Int J Sports Physiol Perform 2019;14:1408–14. doi:10.1123/ijspp.2019-0043.

[4] DeMeulenaere S. Pulse Oximetry: Uses and Limitations. J Nurse Pract 2007;3:312–7. doi:10.1016/j.nurpra.2007.02.021.

[5] Noordin S, McEwen JA, Kragh JF, Eisen A, Masri BA. Surgical tourniquets in orthopaedics. J Bone Jt Surg – Ser A 2009;91:2958–67. doi:10.2106/JBJS.I.00634.

[6] Wall PL, Buising CM, Nelms D, Grulke L, Renner CH. Masimo Perfusion Index Versus Doppler for Tourniquet Effectiveness Monitoring. J Spec Oper Med 2019;19:44–6.

2020-05-25T10:23:12-08:00