I agreed to conduct the testing, but expressed my skepticism about this test (more on that later).
Francesco Conconi suggested that in endurance sports, heart rate increases fairly linearly with work or velocity (J Appl Physiol. 1982 Apr;52(4):869-73.). If you graph work or velocity on one axis and heart rate on another, the result should be a line--more or less--up to a deflection point at which point the line flattens out. That point (heart rate deflection point), according to Conconi, very closely approximates the anaerobic threshold and the work or speed associated with anaerobic threshold. Below is what that might look like for heart rate plotted against watts on a Concept II rowing ergometer:
The plotted red dots represent the coordinates of watts and heart rate. You can see that the line connecting the dots very closely matches the thicker trend line (in Excel) until heart rate reaches 180. Then, while watts are increasing linearly from 220 to 240, the heart rate flattens (remains essentially the same for three measurements). Thus, by Conconi's theory, 180 beats per minute represents anaerobic threshold (AT) intensity for this person.
The person tested above happens to be me. Interestingly, my Conconi-derived AT closely resembled my heart rate at threshold intensity (178) as ascertained in lab tests by Dr. Julie Downing. This apparent corroboration piqued my interest. My initial skepticism remained, however, as we performed the test on others.
We had people warm up on the ergometer fairly easily for 5 minutes or so (and practice holding a steady 90 watts or less just to have the experience of maintaining a certain power). Each rower on the ergometer had a spotter. Each rower wore a heart rate monitor, and each spotter had a corresponding heart rate watch, notepad and pencil.
We separated rowers by at least 5 feet away so the monitors wouldn't pick up others' heart rate signals. The spotters asked the rowers to row for a minute at a very easy power level (110 or so watts) to start and told the rowers to keep the pace consistent for a minute.
Every minute, the spotters asked the rowers to increase the power by 10 watts. The spotter noted the heart rate at the end of each minute and the difference (delta) between the current heart rate and the previously noted heart rate (see example below). For most, the test lasted less than 20 minutes.
This is what a test result might look like:
Mins Watts HR Delta
1 100 90
2 110 95 5
3 120 107 12
4 130 113 6
5 140 118 5
6 150 125 7
7 160 132 7
8 170 138 6
9 180 143 6
10 190 146 3
11 200 149 3
12 210 151 2
test over because the delta is flattening out.
test over because the delta is flattening out.
Interpreting the Results:
Using the above data, we can guess without plotting the data, that the Conconi point is around 190 watts at a heart rate of 146. We can guess this because the delta between heart rates is starting to conspicuously diminish. Plotting the data may make the results more tangible.
Within the group of eight men we tested, we found that any two of us might interpret the data differently. We found that some of us also projected onto the data what we thought was our anaerobic threshold, i.e., we were introducing experimenter/subject bias. The graphs varied between relatively easy-to-read and not so.
Some graphs seemed somewhat clear (e.g., this person with a deflection point and Conconi AT at heart rate of 158):
Some seemed to have multiple potential deflection points (HR of 110, 130, 145 or 160?):
Several of the graphs looked something like this with very little deflection (Conconi deflection point of HR 140 or 150?):
In the end, we did agree on some heart rate and power (watts) and Conconi-AT for each of the eight people who participated. In hindsight, I am not sure how we arrived at these values because they are not all obvious. It is hard to say if this was due to the phenomenon of "group think" or something else. At any rate, we then tried to corroborate our findings. One of us had read an Ed McNeely artcile somewhere (I think in this Rowing News article) that the average pace in an 20-minute all out effort closely resembled anaerobic threshold pace. So, we took our most recent 20-minute pace and compared it to the pace corresponding to the watts at the Conconi-derived anaerobic threshold. The result was a rather remarkable .92 correlation between the two.
Source of Skepticism
I'd like to be able to say that we proved the Conconi test to be a reliable indicator of anaerobic threshold, but we don't really have the data to support that. While we achieved a fairly remarkable correlation between 20-minute test pace and Conconi-calculated anaerobic threshold, it's not clear what that actually means, if anything. Below are some of the sources of skepticism about the Conconi test and our execution of it:
- Looking for deflection points reminds me of stock market technical analysis or tea-leaf prognostication, where you may start to see things that aren't there.
- In Boll Soc Ital Biol Sper. 1980 Dec 15;56(23):2504-10., Conconi et al studied 320 runners and found that the "deflection velocity and anaerobic threshold (established through blood lactate determination) were coincident in 10 runners". One is tempted to add "just" as in "coincident in just 10 runners." That 3% rate is a rather flimsy basis for any physiological phenomenon. In fact, it sounds like the exception rather than the rule. And yet, this early research seems to be the springboard for the Conconi AT test.
- Various studies of runners have found the Conconi test unreliable. For instance, in Int J Sports Med. 1995 Nov;16(8):541-4., authors Jones AM, Doust JH concluded that "the Conconi test [is] unsuitable for reliable evaluation of AT." In Int J Sports Med. 1998 Nov;19(8):553-9., Bourgois J, Vrijens J. found that Conconi's heart rate threshold (ATHR) "does not reflect the anaerobic threshold and is therefore not relevant for monitoring continuous endurance training in rowing." This was a study of younger rowers.
- Using heart rate as a source of information seems potentially
flawed. If some of us had our morning coffee, would our heart rates be more
elevated? If we were nervous about how we performed relative to our peers, would our heart rates more elevated? Were we dressed
too warmly? It seems that many factors influence heart rate and could
have altered our results.
- The granularity of our data may have caused us to miss the physiological changes. We measured heart rate every minute, but might we have gotten better data if we had more continuous heart rate data (e.g. have the heart rate monitor record every 5 seconds) and, if possible, have more continuous watt data as well.
- The whole notion of anaerobic threshold is open to definitional discussion and taking Conconi's protocol definition (more or less) and then corroborating with possibly a different definition from McNeely is not rigorous science. If we had agreed on, say, 4mmol of lactate as a definition of anaerobic threshold and had been blood-lactate tested as well, we might have had a more firm basis for some conclusions.
In conclusion, the non-invasive indices were comparable with the invasive index and could, therefore, be used in the assessment of AT during rowing ergometer use. In this population of elite rowers, Conconi threshold (Con-AT), based on the measurement of HRDP tended to be the most adequate way of estimating AT for training regulation purposes.I'm not sure what "most adequate" means here--maybe cheapest and easiest(?)--but I think you get the drift. The appeal of a non-invasive test means you can still get a Conconi test at places like the UC Davis Sports Medicine facility. In any event, this test is very easy to perform. Why a lab would charge $100 for this test seems curious, but then they may have a better protocol or value-added services.