I did a bunch of searches, trying to see what level of oxygen saturation is okay when doing intense workouts. I like for some of my workout to be at my personal HR max 165+ (not "fat burning" nor "aerobic" but anaerobic). I have no idea what my sats are at these points (during certain songs of zumba, or when I do a hard bouldering route), but based on my very heavy panting and slight quease, I know they're not 99%.
This says that anaerobic sprints can get a person under 95%, women in particular.
http://www.ncbi.nlm.nih.gov/pubmed/20204812
That is considered "significant." So 73% sounds bad, but I would suggest you do a test on a bike, where the measurements are more accurate. See how high you can get your HR, and see what you O2 is at that HR.
This talks about O2 sat measurement issues during exercise.
http://www.lakesidepress.com/pulmonary/books/physiology/chap12_1.htm
Here is the section:
PULSE OXIMETRY
Apart from measurement of heart rate, probably the most commonly measured exercise variable is oxygen saturation by pulse oximetry (SpO2). This is because exercise oximetry is so widely used to assess need for supplemental oxygen in patients with dyspnea, whether or not they are being considered for exercise training. Such a test may consist of nothing more than walking the patient in the hospital corridor while measuring finger pulse oximetry.
Whatever the purpose for exercising a patient, it is essential to screen for exercise-induced hypoxemia, because it may be the cause of symptoms and is treatable. Above 85% true oxygen saturation, SpO2 is accurate to within about +/- 3% of the blood oxygen saturation as measured with a co-oximeter (SaO2) (Escourrou 1990). However, there are several potential pitfalls to using pulse oximetry for exercise testing, including improper capture of pulse, inaccuracy at very low levels of SaO2, intravenous dyes, skin pigment, and poor signal response. In a review of 10 studies utilizing both ear and finger pulse oximetry during exercise, Mengelkoch, et. al. found that only 67% of the pulse oximeters studied were considered accurate when SaO2 was > 85% in non-smokers (Mengelkoch 1995). However, the current generation of finger pulse oximeters appear to be more accurate than the older ear-probe equipped models (Mengelkoch 1995). It is interesting that most of the studies reviewed used cycle ergometry, since it produces less oximetry artifact than a treadmill.
An avoidable pitfall of pulse oximetry can occur when there is excess carboxyhemoglobin. In contrast to blood co-oximeters, which utilize four wavelengths of light to separate out oxyhemoglobin from reduced hemoglobin, methemoglobin (MetHb) and carboxyhemoglobin (COHb), pulse oximeters utilize only two wavelengths of light (Powers 1989; Principles of Pulse Oximetry 1991). As a result, pulse oximeters measure COHb and part of any MetHb along with oxyhemoglobin, and combine the three into a single reading, the SpO2. (MetHb absorbs both wavelengths of light emitted by pulse oximeters, so that SpO2 is not affected as much by MetHb as for a comparable level of COHb). Powers, et. al. showed that in subjects who smoked and had COHb levels of >4%, pulse oximeters significantly overestimated SaO2 (Powers 1989).
Whereas excess methemoglobin is an uncommon finding clinically, excess carboxy-hemoglobin is present in all cigarette and cigar smokers. A resting SpO2 should be correlated with a measured SaO2 and (if a blood co-oximeter is available), COHb and methemoglobin levels. If the measured SaO2 does not agree with SpO2, the fact should be noted, reason(s) sought, and then accounted for during the exercise test. If a measured SaO2 cannot be correlated with SpO2, exercise testing should not be done in current smokers, so as to avoid falsely high SaO2 readings. (The half-life of CO breathing ambient air is about 6 hours, so 24 hours after smoking cessation the CO level should be normal, i.e., less than 2.5%.)
Carbon monoxide can also be measured in exhaled air as ppm (parts per million) and correlated with a blood carboxyhemoglobin level (e.g., 10 ppm roughly equals 2% COHb). Also, if a co-oximeter is available, carboxyhemoglobin can be reliably measured on a venous blood sample; the value is the same as arterial. If venous COHb is elevated, its value can be subtracted from the SpO2 to get a truer reading of the patient's SaO2. Attention to CO is important if one is to obtain accurate estimation of the patient's blood oxygen status.
Finally, here is a complicated study, but I think it is interesting. It talks about how doing anaerobic (short intense bursts) exercise should be emphasized as well as aerobic exercise. Lung function was not correlated with ability to do anaerobic exercise, nutritional status/muscle mass was, and higher anaerobic tolerance is associated with easier daily life tasks (stairs, carrying things).
http://ajrccm.atsjournals.org/content/157/4/1145.full
So, you can get your HR up, build muscle, and takes breaks btw to keep your o2 sats up. I know a couple people who post here and lift weights have talked about this.
