Is it time to scuttle the FEF25%-75%?

When we went through our hardware and software upgrade last August, one of the changes we made was to stop reporting the FEF25%-75% (AKA MMEF, MMFR, MMF). The pulmonary physicians had long since stopped using this value when assessing spirometry results and we had kept it on our reports as long as we did only for inter-laboratory compatibility. Along with other changes we made at that time we decided it was time to drop the FEF25%-75% off our reports.

FEF25%-75% has been used to assess “small airways disease” but more than one of our pulmonary physicians has said that they don’t believe there is such a thing. I’m not a clinician but I’ve always felt that tests and results need to be clinically useful in order to be performed or reported and more than one study has shown little correlation between anatomical findings and FEF25%-75%.

Regardless of whether or not small airways disease is an actual entity my first objection to the FEF25%-75% has to do with the concept that it measures flow in small airways when for most patients it lies within their FEV1. For this reason it has never been clear to me what the FEF25%-75% is measuring that the FEV1 isn’t. More importantly, I have significant concerns about the limitations involved in measuring the FEF25%-75% in the first place.

FEF25%-75% is measured by identifying the points at which 25% and 75% of the Forced Vital Capacity has been exhaled and then calculating the change in volume divided by the change in time:

FEF25-75_graph1 Using the FVC as the primary reference means that the measured FEF25%-75% is highly dependent on the FVC volume. Getting a truly maximal FVC from patients with lung disease requires a lot of effort and cooperation from the patient. An effort where the FVC is underestimated will cause the FEF25%-75% to be disproportionally overestimated. Small changes in FVC can have large changes in FEF25%-75%.

FEF25-75_graph2

This was also noted when the FEF25%-75% from pre- and post-bronchodilator spirometry efforts were compared. Numerous investigators saw that FVC and FEV1 could increase significantly post -bronchodilator but the FEF25%-75% often did not. The fact is that the FEF25%-75% from the pre- and post-bronchodilator efforts was being measured across a different set of lung volumes whenever the FVC increased post-bronchodilator. The solution has been to measure the post-bronchodilator FEF25%-75% at exactly the same volume points as the pre-bronchodilator effort. This is called volume adjustment and to some extent it makes sense but at the same time it calls into question exactly what the FEF25%-75% is measuring.

FEF25-75_graph3 To be honest, I think that adjusting the FEF25%-75% volume smacks of tweaking the results to meet the expectations. I will agree that there is a general correlation between flow rates and airway size during a forced exhalation but strictly speaking this is what the different flow-volume loop contours are all about. The primary problem with applying this concept to the FEF25%-75% is that the FEF25%-75% is an average flow rate that says nothing about the actual flow rates between the two values used to measure it.

FEF25-75_graph4

Since the FEF25%-75% lies within the FEV1 it is not surprising that it correlates well with airway obstruction. The correlation between FEV1/FVC ratio and FEF25%-75% is actually too good because at least one study showed that FEF25%-75% is always normal when the FEV1/FVC ratio is normal. This brings into question what additional information the FEF25%-75% adds, if any, towards assessing spirometry results.

As an alternate to FEF25%-75% some investigators have suggested that FEV3 and the FEV3/FVC ratio provides a much better window onto small airways. I did a pilot study on a few hundred patients using the FEV3/FVC ratio with ambiguous results. I found that like the FEF25%-75% the FEV3/FVC ratio was abnormal when the FEV1/FVC ratio was normal only rarely. It may be possible that with a larger sample of patients the FEV3/FVC can serve a purpose but so far none of our pulmonary physicians have shown an interest in it so its value remains speculative to me.

Our lab software is able to report over two dozen different values from a single forced vital capacity. Most of these values are not clinically useful. The FEF25%-75% has high inter-test and intra-test variability and is unduly affected by FVC. Given the limitations in how and what it measures it is not clear to me that the FEF25%-75% has much to do with identifying the site of airway obstruction. It is also unlikely that it provides any information not already provided by the FEV1 and FEV1/FVC ratio. I think it is time that everyone should think about dropping the FEF25%-75% from their reports too.

References:

Berend N, Wright JL, Thurlbeck WM, Marlin GE, Woolcock AJ. Small airways disease: Reproducibility of measurements and correlation with lung function. Chest 1981; 79: 263-268

Cockcroft DW, Berscheid BA. Volume adjustment of maximal midexpiratory flow. Importance of changes in Total Lung Capacity. Chest 1980; 78: 595-600

Gelb AF, Williams AJ, Zamel N. Spirometry. FEV1 vs FEF25-75 percent. Chest 1983; 84: 473-474

Hansen JE, Sun XG, Wasserman K. Discriminating values and normal values for expiratory obstruction. Chest 2009; 136: 369-377

Sherter CB, Connolly JJ, Schilder DP. The significance of volume-adjusting the maximal midexpiratory flow in assessing the response to a bronchodilator drug. Chest 1978; 73: 568-571

Creative Commons License
PFT Blog by Richard Johnston is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.

28 thoughts on “Is it time to scuttle the FEF25%-75%?

  1. Hi. Interesting analysis.

    My Spirmetry test had values of: FVC normal @ 124%, 5.77(L),
    FEV1 mildly reduced at 77%, 2.48 (L), and FEV1/FVC ratio severely reduced @ 43%.

    However, post bronchdilator there was a significant improvement of 28% @ the FEF25/75 level. NO data was given on the report of the PD for the FEV1 and FVC, and I’m assuming that there was no change n these values PD.

    This seems to suggest that there may be some use of FEF 25/75 values if the FVC is quite above normal.

    Would my FEV1 be higher Post Bronchdilator if my FEF 25/75 was up by that much (28%). I’m thinking that there must be some significance in this “reversibility” aspect of the PD.

    Just like to add that, I suspect that the FVC is so high due to my own great effort in inhaling. I really put the rib cage/shoulders into it on the “intake” of air. Maybe my FEV1/FVC ratio was affected-downward because of the “artificially” high FVC.

    Thanks. Very interesting blog.

    • Up to about 20 years ago some of the physicians I worked with used an increase in FEF25-75 of 40% as an indication of bronchodilation. I can’t point to a specific research study that changed their mind but they stopped using FEF25-75 and stopped teaching its use to fellows and residents at least that long ago. The current recommendations from the ATS-ERS is to assess bronchodilator response only by changes in FVC and FEV1 (with a threshold of 12% and 200 ml for both). That FEF25-75 can improve following bronchodilator was acknowledged in the ATS-ERS recommendations but they also noted that due to the way it is measured FEF25-75 can increase for a number of reasons not associated with bronchodilation and is not a reliable indicator.

      I don’t know why your report did not include your post-bronchodilator FVC and FEV1. My lab always reports the same set of results for baseline and post-bronchodilator tests. The only reason we wouldn’t report something is because the test quality was inadequate not because the results didn’t show a significant change.

      I am glad you put good effort into your test. To some extent your FVC of 122% of predicted shows the limitations of reference equations based solely on height. For most people there is a reasonably good relationship between height and lung volume but humanity follows a bell-shaped curve and the lung volume for a small number of people is going to be either above or below that “normal” range and still be normal for them. To some extent this is what the issue of ethnicity plays in reference equations. However when FVC is elevated, FEV1 is usually elevated by the same amount. In a way you’re right that your elevated FVC is reducing your FEV1/FVC ratio more than it “should”, but this is in part because the current recommendation from the ATS-ERS is to use the FEV1 to assign severity, not the FEV1/FVC ratio. Playing devil’s advocate I could argue that because your normal FVC is elevated what would be a mild decrease in FEV1 for somebody else should be a moderate or severe decrease for you. FEV1 is a better indicator of clinical severity however, and the best use of the FEV1/FVC ratio is simply as an indication for the presence of airway obstruction. The amount the FEV1/FVC ratio is decreased doesn’t matter as much as the fact that it is decreased.

  2. Thanks for the detailed reply. Very much appreciated.

    With regard to only the post-bronchodilator FEF 25/75 values stated, and not the PB FEV1 and FVC also given on the report, I can say that the report has a statement on it that reads:
    …………………………………………………….
    “This document is a draft and was reproduced from it’s electronic form for informational purposes, only. This draft should not, under any circumstances, be relied on as a complete and final reproduction of the legal medical record unless authenticated by the responsible author.”
    ………………………………………………..
    I may call my doctor’s office to ask if a more complete report is available as, yes, it would be helpful to have those other PB values.

    From reading-up on various studies that I searched on the internet, it caught my attention when I read that, if the FEF 25/75 shows a post-bronchodilator increase of 30%, it could be considered as significant as an FEV1 post bronchdilator increase of the 12% standard. Yes, of the many articles that I’ve read, this was the only article that made particular note of the FEF25/75 with regard to its having some weight in an evaluation (although a 30% PB “reversibility” value seems to me (?), would be a particularly unusual occurrence- and mine being a close-to-it 28% had me wondering): it seemed more of an asthma test result, and I don’t have asthma.

    Upon reading this, my theory was that, if the above is the case, and it could be interpreted as “effectively” a post-bronchodilator increase in the FEV1 to some degree, (my FEV1 being @ 77% pre-bronchodilator), then perhaps my test results would qualify as exceeding the “80% predicted normal” standard under the Gold System recommendation of using the post-bronchodilator values for the final determination of the test. In other words, under the Gold System, I’d have a normal FEV1 as well as a normal FVC. However, yes, my FEV1/FVC ratio may very well still show a decrease to a below “70% normal” value. But, the overall evaluation may be viewed as a more health-wise, positive result.

    However, I do take well you’re point in explanation that the FEV1 significantly supersedes other, much lesser considerations of values (especially the FEF25/75), and my thoughts on the matter probably aren’t leading to a substantive difference in the overall interpretation of the test results.

    Thanks again,
    Ed

    Now, this would be if I have all my facts of the matter correctly understood, especially with regard to the Gold System of evaluation, etc.

    • A number of different values have been proposed to measure the response to bronchodilators. One example is the change in Inspiratory Capacity. Many individuals with COPD are hyperinflated (their FRC is elevated because their airway obstruction does not let them exhale completely before the next breath). For these patients a clinically significant increase in IC (and decrease in FRC) may occur post-bronchodilator when the FEV1 and FVC do not change significantly. IC, however, can be altered by changes in breathing pattern (tidal volume and respiratory rate) that have nothing to do with a change in hyperinflation and for this reason (like FEF25-75) it has not been considered to be a robust enough a measurement to be used to assess reversibility.

      Remember also that your percent predicted depends on the reference equations that are used by the spirometry system you took your test on. Reference equations continue to be an area of contention (for the issue of ethnicity if nothing else). The ATS-ERS recommendations state that a PFT Lab should chooses the reference equations that best match their patient population but do not give any particular guidelines on how to do this. At the moment, I think the NHANESIII and GLI reference equations are superior to most others because of the large number of subjects and the sophisticated statistical analysis. If you are interested, the GLI has a program you can download that will calculate your percent predicted FVC, FEV1, FEF25-75 and FEV1/FVC ratio (http://www.lungfunction.org/component/content/article/85-equations-and-tools/equations/94-gli-2012-desktop-software.html).

      GOLD does not recommend a specific set of reference equations and tends to depend more on absolute ratios. In the past they have used an FEV1/FVC ratio below 0.70 as a threshold but more recently they have acknowledged that that values below 0.70 may be normal in the elderly and that a threshold of 0.65 may be more appropriate for individuals over the age of 70.

  3. Hi, I thought that I’d follow-up with the post-Bronchodilator test results as discussed earlier, above.

    The FEV1 did increase for Post-Bronchdilator (from 2.48L to 2.77L) by 9%. And by 5% for the FVC (5.77L to 5.99L).

    The VC was 5.95L (127%>) Pre-Bronchdilator, on a “predicted favored range” , btw. No post-Bronchdilator was given on the “VC” (which I assume is short for “SVC”).

    Also, I now think that I understand the dynamics of the FEF25%/75% a lot better than when I originally posted. In that, because i hadn’t seen the pre-Bronchodilaor values of the test, and only the post-bronchdilator results of the FEF25%/75%, I assumed that a 28% post-Bronchodilator was at least somewhat significant. However, the pre-Bronchdilator values for all the “FEF values” were quite low, anyway, and didn’t approach normal range; even with the 28% post-bronchodilator increase.

    I think that I now realize one of your main points about the lack of usefulness of the FEF25%/75%, where: if the FAV is high, it drives down the “FEF values” proportionately. Or rather, as you put it, if the FVC is low, it drives up the FEF25%/75% proportionately. I’m assuming that it “works both ways”.

    One other value on the “full test results” that got my attention was that the FEV3 value was normal: 4.00L @ 88%, and post-Bronchdilator up 12% @ 4.48L (% ?) of a predicted normal basis of 4.53L. The “FEV values” seem to respond to post-Bronchodilator; if not technically, clinically significant, then notable given the somewhat favorable level of FEV1 pre-Bronchodilator value of 77% (post-Bronchdilator increases it to 79% ?).

    The lung volumes (pre-bronchodilator; no post bronchodilator was done for the volumes):

    TLC “mildly increased” @ 8.72L, 121%> on a “predicted favor range” of 7.18 ; RV normal @ 110%; RV/TLC ratio normal @ 91%. so far so good, but I did read your entry about FRC indication of hyperinflation, and the value of FRC was 5.79L @ 142%> of a of 4.07L “predicted favor range.”

    So, i thought that I’d fill-in some of the blanks from, above, and thanks again.

    • It is customary to express post-bronchodilator change in measured values as percent of the baseline value. The use of percentages is an idiosyncrasy peculiar to respiratory medicine. Consider the bronchodilator effects on FEV1. There is ample evidence that the post-BD change in FEV1 is independent of the baseline value (see references 1-7). This implies that, as the baseline value is more pathological, change expressed as % baseline increases. So we would be led to believe that, as airways obstruction worsens there is greater reversibility. This is a paradoxical conclusion, as severe airways obstruction is almost invariably associated with structural changes of peripheral airways and lung tissue which preclude large changes in airway diameter (Ref. 8). It is for that reason that it is to be preferred to express change as a percentage of the predicted value, because this avoids the bias associated with a poor baseline value.

      1. Brand PL et al. Interpretation of bronchodilator response in patients with obstructive airways disease. The Dutch Chronic Non-Specific Lung Disease (CNSLD) Study Group. Thorax. 1992;47(6):429-436.
      2. Report of the Committee on Emphysema, American College of Chest Physicians. Criteria for the assessment of reversibility in airways obstruction. Chest. 1974;65(5):552-553.
      3. Ries AL. Response to bronchodilators. In: Clausen JL, ed. Pulmonary function testing guidelines and controversies, New York: Academic Press, Inc. 1982;215-222.
      4. Sourk RL, Nugent KM. Bronchodilator testing: confidence intervals derived from placebo inhalations. Am Rev Respir Dis. 1983;128(1):153-157.
      5. Tan WC et al. Worldwide patterns of bronchodilator responsiveness: results from the Burden of Obstructive Lung Disease study. Thorax. 2012;67(8):718–726.
      6. Anthonisen NR, Wright EC. Bronchodilator response in chronic obstructive pulmonary disease. Am Rev Respir Dis. 1986;133(5):814-819.
      7. Ward H, Cooper BG, Miller MR. Improved criterion for assessing lung function reversibility. Chest. 2015;148(4):877-886.
      8. Jeffery PK. Remodeling in asthma and chronic obstructive lung disease. Am J Respir Crit Care Med. 2001;164(10 Pt 2):S28-38.

  4. I fully agree with the conclusion that FEF25-75% (just like other instantaneous flows) is an index that should be abandoned, and that it is a myth that it is a specific indicator of “small airways disease”. I have analysed this using data from real patients and showed that FEF25-75% did not contribute information that was not available from FEV1, FVC and FEV1/FVC [1]. These findings have since been corroborated [2-3]. See also [4].

    By the way, it is also high time that we scuttle the habit of expressing a measurement as percent of predicted. As Sobol wrote [5]: “It implies that all functions in pulmonary physiology have a variance around the predicted, which is a fixed per cent of predicted. Nowhere else in medicine is such a naive view taken of the limit of normal.”

    1. Quanjer PH, Weiner DJ, Pretto JJ, Brazzale DJ, Boros PW. (2014) Measurement of FEF25-75% and FEF75% does not contribute to clinical decision making. Eur Respir J 43: 1051-1058.
    2. Boutin B, Koskas M, Guillo H, Maingot L, La Rocca MC, Boule et al. (2015). Forced expiratory flows’ contribution to lung function interpretation in schoolchildren. Eur Respir J, 45(1), 107-115.
    3. Lukic KZ, Coates AL. (2015). Does the FEF25-75 or the FEF75 have any value in assessing lung disease in children with cystic fibrosis or asthma? Pediatr Pulmonol 50(9), 863-868.
    4. Pellegrino R, Brusasco V, Miller MR. (2014). Question everything. Eur Respir J, 43(4), 947-948.
    5. Sobol BJ, Sobol PG. 1979 Per cent of predicted as the limit of normal in pulmonary function testing: a statistically valid approach. Thorax 34, 1-3

  5. Allow me to convince you that FEF 25-75 does have some clinical value. I’ve had PFTs where my FVC, FEV1, FVC/FEV1 ratio were completely normal but my FEF 25-75 was pretty low at 60%. This contradicts the study you noted about the FEF 25-75 always being normal if the FVC/FEV1 ratio is normal. I have had instances were I inhaled two puffs of Xopenex and my FEF 25-75 improved by 19% post bronchodilator, but my FVC and FEV1 stayed the same. Due to guidelines stating that asthma can only be considered if FEF 25-75 improves by 30% or more, we couldn’t say that I had reverseable small airways disease (asthma) but we could see I may have a mild case of COPD that was not seen using the traditional method of focusing solely on FVC. I will say that it can be confusing to interpret FEF 25-75 because I’ve read some data that states you should be above 80% and other data that states as long as you’re above 65% you’re considered “normal”, especially if your FVC and FEV1 are normal as well. The highest I’ve hit on the PFT for FEF 25-75 is 75% so I would be comfortable saying that I do believe I have some level of small airways disease.

    • Michael –

      I pretty much stand by my original statements. For the vast majority of people the FEF25-75 is measured inside the same volume that is measured by the FEV1. The way in which FEF25-75 is measured causes it to be disproportionally affected by small changes in the FVC. This is one reason why FEF25-75 is more variable than FEV1 and FVC on a test-to-test basis (and why its standard deviation in reference equations is so much higher). Post-BD FEF25-75 changes are often measured by the isovolume technique (i.e. at exactly the same expiratory volumes as the pre-BD FEF25-75) which begs the question as to what’s being measured (and smacks of adjusting the results to meet the expectations). Reported FEF25-75 measurements are supposed to come from a single spirometry effort with the best combined FVC+FEV1 (ATS/ERS standard) but there is not a single study that says this is the “best” FEF25-75. There is also no study that says the highest (or lowest or the average) value of FEF25-75 is correct either. Since most lab software can select the FVC and FEV1 from separate efforts (again ATS/ERS standard) this brings into question which effort the reported FEF25-75 was taken from (if from an effort with the highest FEV1 and a lower FVC, FEF25-75 will be “overestimated” and if from an effort with the highest FVC and a lower FEV1 it will be “underestimated”). For these and other reasons, measurement of terminal air flows is likely more accurate when done by the FEV3 and FEV3/FVC ratio.

      Regards, Richard

      • Richard, after reading your reply I can understand how one can be skeptical of the value of FEF 25-75. You bring up a valid point – During my PFTs, as many as four different expirations are recorded, so which FVC and FEV1 is being chosen to determine FEF 25-75? This can create underestimated and overestimated values when it comes to calculating FEF 25-75. You have to understand, I am merely a patient. Since you are a pulmonary tech who has been doing this for decades, some of what you are saying is difficult for me to interpret, having no training in the pulmonary field. Regardless, thank you for your reply and I will be sure to bring up some of what you said at my next meeting with my pulmonologist on Tuesday.

  6. Hi Michael,
    No-one would express the glucose, creatinine or whatever other concentration of a substance in blood or plasma as per cent predicted. Yet it is ingrained in respiratory medicine unlike in other medical disciplines to express a measured value as per cent of predicted. As explained above this is a bad habit which leads to clinically important erroneous conclusions. The underlying assumption that the natural variability is proportional to the mean value in a healthy subject is not valid. Let us consider a white male of average height (175 cm). At age 20, 40, 60 and 80 years the predicted FEF25-75 is 4.92, 4.00, 2.91 and 1.97 L/s, respectively; the corresponding lower limits or normal are 3.21, 2.34, 1.43 and 0.75 L/s, respectively. Therefore the lower limit of normal declines from 65.2% at age 20, 58.5% at age 40, 49.1% at age 60 to 38.1% at age 80 years. Clearly the scatter around predicted in healthy subjects is not proportional to the predicted value, and expressing measured values as per cent of predicted introduces an serious age bias.
    It is also a myth that a low FEF25-75 is indicative of small airways obstruction, unlike the FEV1, FVC and FEV1/FVC. Do by all means read the above publications as well as:
    1. Pellegrino R et al.. (2014). Question everything. Eur Respir J 2014;43(4), 947-948.
    2. Stanojevic S et al. The Global Lung Function Initiative: dispelling some myths of lung function test interpretation. Breathe 2013; 9; 463-474.

    • Hello, Philip. I agree with you that percentages can be misleading. For example sometimes a patient’s CBC might show the percentage of Neutrophils to be a little high but when you look at the absolute Neutrophils they are really not that elevated and do not have clinical significance. So allow me to show you my absolute values, which were recorded a few weeks ago at my pulmonologist’s office. Pre-bronchodilator, my FEF 25-75 was 3.45 Liters. Post-bronchodilator, my FEF 25-75 was 4.13. Again, although it didn’t quite meet the criteria for reverseable airway disease, I blew more than half a liter of air compared to pre-bronchodilator readings. How would you explain that? My FVC only changed from 5.53 to 5.64, and my FEV1 only changed from 4.23 to 4.47 so it’s hard to imagine such a minuscule change in FVC and FEV1 would overestimate the significant change in FEF 25-75. Btw, I looked at the first study that you posted and found it to be very interesting. I had no idea that FEF 25-75 was established to look into airway disease associated with smoking. I googled your name and see you work at the hospital in Rotterdam as a pulmonologist. Just out of curiosity, do you mostly do research or do you also treat patients?

  7. Michael,
    I appreciate your reasoning: the relative change in FEF25-75 is bigger than that in FEV1, so you conclude that it is the better index of bronchodilatation. However, bigger is a observational finding, better is a value judgement, and they are not equivalent. There is post-bronchodilator improvement in all indices, whether it is significant or one change is more significant than another hinges on the definition of “significant”. The reproducibility of a measurement matters when comparing pre- and post-BD measurements. Therefore in bronchodilator testing one performs measurements in a large panel of healthy subjects to gauge the range of responses, and then decides a reference range for what constitutes a normal response. The FEF25-75 has much greater variability than FEV1 and FVC; so the reference range for a response is likely to be much larger, but I am not aware of a study quantifying this, whereas there are many studies documenting this for FEV1 and FVC. Therefore, yes, there appears to be an effect of the bronchodilator drug in your case, and numerically the change in FEF25-75 looks more impressive. Obviously the improved airway potency is expressed in both FEF25-75 and FEV1, and the 240 mL increase in FEV1 is by no means trivial. An improvement in FVC is usually next to nil in healthy subjects; FVC starts to increase in subjects with overt airway obstruction, and the improvement is larger the more severe the obstruction; it reflects opening up of lung compartments behind hitherto obstructed airways (air trapping). I do not know your age and height, but judging from the values that you listed your pulmonary function is near or possibly within the normal range.
    Finally, the more indices one measures, the greater the chance that one of them might be out of the reference range by poor chance.
    Bottom line: do not equate “bigger” to “better”.

  8. I found this blog post while researching an unrelated topic, and wanted to add a few thoughts.

    Regarding Ed Townsend’s comments, errors in FVC measurement are only errors of understatement, and I believe this is the only measurement in pulmonary diagnostics where this is true. If the FVC (or SVC) measurement is maximally and correctly performed, and the equipment is functioning correctly, if you achieved a FVC of 124% of predicted, THAT is your FVC. It’s not overstated.

    Richard, you addressed an issue with FEF25-75% reporting and the ATS/ERS guidelines in your response to Michael Iqbal. The ATS/ERS standards are made by a committee, and these standards are intended for a group with large variations in clinical ability (and motivation). By no means is this a criticism of the guideline committees, and overall, they work as intended, but there are areas that aren’t clearly addressed (DLCO BHT: what should be reported – this isn’t addressed at all), or there are recommendations such as reporting everything from the trial with the best effort, defined as the trial with the largest sum of FVC and FEV-1. I feel that the ATS/ERS definition of “best effort” doesn’t work well with more than moderate obstruction. First and foremost, spirometry is performed to assess the presence of airway obstruction, and airway obstruction is best revealed when the patient sharply and suddenly presses down on these airways. I think Vext and PEFR need to be included in the “best effort” criteria, because the trial with the lowest Vext and highest PEFR always best show airway obstruction. If the FVC and FEV-1 are lower in the trial with the lowest Vext and highest PEFR, these are the most accurate data. The highest FVC and FEV-1 will be overstated and misleading if the Vext and PEFR in that trial do not demonstrate maximal performance. To borrow from Philip Quanjer (in a different context), do not equate “bigger” to “better”.

    Regarding FEF25-75%, I’ve often felt that much of the variability could be attributed to variable FVC performance, much of which could be attributed to variable maneuver instruction and correction by technologists. I’ve always gotten something of a Gestalt sense from the FEF25-75%; it confirmed what I saw in FEV-1 and on the FVL. That said, because it is so variable, I’m in agreement with discontinuing the reporting of this value, and I look forward to the results of further investigation of what can be obtained from FEV-3, FEV-6, etc.

    • You make some valid points, Jim. The only thing I can add is that I don’t think we should fully abandon FEF 25-75 because it does serve a clinical purpose in rare circumstances. For example, if someone’s FEV1, FVC, and FEV1/FVC ratio were completely normal but his or her FEF 25-75 was let’s say 30%, that is significant and illustrates proof that small airway obstruction doesn’t always show up on FEV1. In addition, if that same person completed a bronchodilator challenge and his or her FEF 25-75 went from 30% to 70%, but all the other values remained the same, I would argue many pulmonologists would diagnose this person with having some level of reverseable small airways obstruction. Not to mention at this point in time, most of the machines do not report FEV3 and I suspect it’s going to be some time before they do. Btw, there’s a great study that I will try to post here which explains that low FEF 25-75 with normal FEV1 is associated with an increased level of uncontrolled asthma and morbidity in children.

      http://www.ncbi.nlm.nih.gov/pmc/articles/PMC3398223/

      • Thanks for providing the link to the article. I agree with Michael. I don’t think the FEF 25-75 should be abandon, instead we need more research & data in this areas. I have a medical background and have studied my PFT’s and other respiratory tests for years and I have had consistently low readings for the FEF 25-75 for the last 15 years. I also have had huge response in the post-bronchodilator measurements of the FEF 25-75, especially in FEF 50 & FEF 75 by 36-49% and higher in some tests. Surely this type of consistency in some patients must tell us something. I have had 10% improvement on and off in the FEV1, but most often not, and yet the FEF 25-75 have always been below 70 and most often below 60. This has to mean something in my clinical case, at least. I believe it is significant and is a tell tale sign of the severity of the asthma. When one looks at my measurements in FVC & FEV1 you would think I was exaggerating my asthma symptoms because those measurements appear normal and therefore I am not taken seriously. I would bet that prior to developing asthma as an adult, my FVC and FEV1 would have been much higher than they are now. Evidence on a good breathing day when I have been tested shows that I can get well above 100% of predicted in FVC and FEV1. Why Pulmonologists would ignore my mid range flows measurements is beyond me as they are consistently low and tell a story. I think we have to start thinking outside the box and not follow along like a herd of sheep. Any new discoveries in medicine occurred because people were able to think for themselves and challenge the status quo. If any Pulmonologist wants to see my test results I would be happy to share them. Any feedback would be appreciated.

        • Teri –

          The FEF25-75 has been studied for close to 50 years so I’m not sure how much, if any, additional study is warranted. The reason that most pulmonologists are disenchanted with the FEF25-75 is that it is usually abnormal only when the FEV1 is abnormal; that the normal range is excessively broad; that it is actually part of the FEV1; that it is excessively affected by the FVC volume and expiratory time; and that there has been no clear proof that it is associated with small airways disease (as opposed to any other form of airway obstruction). I personally think that the FEV3/FVC ratio presents a clearer signal of terminal expiratory flows than does the FEF25-75.

          – Richard

          • Hi Richard,

            Thanks for the explanation. I have not heard much about the FEV3/FVC ratio until this post. Can you get this ratio from the existing tests or only in the research lab? My FEV1 was within normal range in early Feb. 2017, but I had a 10% improvement post bronchodilator, while there was little change in FVC which was 110% vs 114% post. I am puzzled why the mid flow range is consistently out of proportion and lower compared to my other lung measurements.

          • Teri –

            I can’t say the FEV3 and FEV3/FVC ratio is an option on all test systems, but it’s on all of the test systems I’ve seen. The issue is whether or not it’s added to the report (and whether or not the selected set of reference equations includes FEV3). There is little consensus on how terminal expiratory flow rates should be measured and what these terminal flow rates mean. One major problem with measuring terminal flow rates is that they depend far more than most of the other measurements made from the FVC maneuver on there truly being a maximal exhalation. My personal opinion is that the FEV3/FVC ratio is as accurate of measurement of these flow rates as the FEV1/FVC ratio, but given the range of test quality from routine spirometry there are limits to the accuracy and pertinence of the FEV1/FVC ratio too.

            As far as your mid-flow range being low you haven’t mentioned your FEV1 %predicted other than to say it was WNL. Your FEV1 may be WNL but when paired with an FVC that is 110% of predicted what is your FEV1/FVC ratio %predicted? Another point is that (as with all predicteds) there is a wide range of predicted FEF25-75 values so to some extent “lowness” depends on which reference equations are being used to report your results.

            One final point is that the ATS/ERS guidelines state to report the largest FVC and FEV1 even when they come from different tests. The FEF25-75 taken from a test with a high FEV1 and a low FVC will have an high FEF25-75 (relatively speaking, of course) whereas when it’s taken from a test with a low FEV1 and high FVC the FEF25-75 will be low. In my lab’s system it is sort-of possible to select FEF25-75 from a specific test but FEF25-75 is lumped together with other values such as Peak Flow and expiratory time and our policy is to always report the Peak Flow (which means the FEF25-75 and expiratory time too) from the effort with the best FEV1 but this may differ for other labs. I’ve also seen test software that does not allow the technician to select separate values from tests; the software makes the selection instead (usually based on best combined FVC + FEV1). For these reasons your reported FEF25-75 may be in part an artifact of this selection process.

            – Richard

          • Thanks Richard for your reply. My FVC, FEV1, FEV1/FVC were at least 100% of predicted or slightly over and with 10% improvement in FEV1. It is only the range between FEF 25-75 that is low and out of proportion.

  9. Michel, You start from a very hypothetical example, i.e. that FEV1, FVC and FEV1/FVC are in the normal range but that FEF25-75 is 30% predicted. FEF25-75 forms part of the FEV1, so it is extremely difficult to imagine how flow over the middle half of the FVC can be severely limited without affecting the FEV1. I also reiterate that the use of precent of predicted should be a thing of the past, simply because it is based on the assumption that the scatter around predicted is a fixed percentage of the predicted value and does not vary with age. But that is not the case at all, the predicted value declines much faster with age than the scatter. I have checked this for healthy non-smoking women age range 4-95 years using the GLI-2012 prediction equations. In white females at ages 20, 30, 40, 50, 60, 70 and 80 years respectively the lower limit of normal is at 64.8%, 61.4%, 57.0%, 50.9%, 41.5%, 34.4% and 25.1%, respectively. In males the lower limits are even somewhat smaller.
    I looked at 22,767 clinical records (49% females), ages 4-95 years. Measurements were classified as normal/abnormal based on the lower limit of normal according to GLI-2012 predicted values. Among 12,544 subjects with a normal FEV1, FVC and FEV1/FVC ratio I identified only 3 male subjects whose FEF25-75 was 30% of predicted or less. So my first reflex would be to look vary carefully at the recordings and make very sure that the manoeuvres had been performed correctly. However, in this case the ages were 91.7, 90.8 and 92 years, implying that these results were still within the normal range for age.
    When looking at change it is also important to take into account the within-person reproducibility of the test. There are no valid data on this in the literature because the few available data is expressed in % and do not take into account that the percentage response thus expressed will vary with age.
    It is really high time to relegate % change and the FEF25-75 to the museum.

    • Philip, I actually agree with most of what you said – like the fact that there is some variability in FEF 25-75, due to the patient not always being able to reproduce the same values on multiple expirations. I saw a second pulmonologist who looked at my spirometry tests and he said I’m pretty much normal at this time. He said he usually doesn’t pay attention to FEF 25-75 unless the post bronchodilator change is upwards of 30% or more, compared to pre bronchodilator readings. He said many patients get misdiagnosed as having asthma due to doctors misinterpreting FEF 25-75. In essence, he did indirectly agree with what you and Richard have been saying. My post FEF 25-75 readings were 19%, so he concluded that I don’t have active asthma. Another thing is that when I take Xopenex, I don’t see any improvement in my breathing. There are only two ICS that target the small airways, QVAR and Alvesco. I tried QVAR and it just made me worse. This is why I’m convinced that I don’t have asthma. I may have some kind of other respiratory disease, but I believe my SOB may be a side effect of Atenolol. Atenolol is supposed to be one of the beta blockers that doesn’t cause SOB but about 6% of patients still experience. I could be one of them. I also have GERD, allergies, prominent lingual tonsils – all of which can also cause SOB. Anyway, Philip, I would still like you to address this study that does show that there are children who have asthma with normal FEV1 values and low FEF 25-75. Thanks.

      http://www.ncbi.nlm.nih.gov/pmc/articles/PMC3398223/

  10. Hi Michel,
    Trust your doctor. Even doctors with all their background knowledge often do not well at self-diagnosing their disease, let alone someone who does not have this background and therefore cannot balance on piece of information against another. The article you refer to bases its assessment of FEF25-75% on an article by Knudson published in 1983, and uses a fixed percent of predicted across age ranges. We have moved on since.
    Please have a look at my message on this forum dated 24 March 2016: there are 4 references to publications that found the FEF25-75% has no added value over FEV1, FVC and FEV1/FVC. The fifth reference argues against the use of percent of predicted.
    You quote two corticosteroids as targeting specifically small airways. That is a myth, just look at the specifications provided by the manufacturer.
    I feel quite sorry about your respiratory disorder and fully understand that it greatly worries you and that you want to know more about it in the hope of finding a remedy. Do value your doctor’s expertise, however.

  11. Hello, I am a patient and I try to learn about my medical conditions. A few months ago I knew nothing of asthma except that people who are asthmatic tend to wheeze and could die without an inhaler. Over the past few months I have read a lot about FVC, FEV1, FEF 25-75%, most of which is over my head. I find it interesting that there is a lot of discussion that dismisses FEF 25-75%, especially considering my test results. I would like to know what this forum would make of my test results. Would you consider this a case of bronchial asthma? Would you prescribe an inhaler?

    Pre-Bronch
    FVC – (97%), FEV1 – (84%), FEV1/FVC – (85%),
    FEF25% – (69%), FEF 75% – (41%), FEF 25-75% – (55%)

    Post-Broch
    FVC – (96%), FEV1 – (97%), FEV1/FVC – (100%),
    FEF25% – (102%), FEF 75% – (85%), FEF 25-75% – (94%)

    Notable Changes
    FEV1 (+16), FEF 75% (+107), FEF 25-75% (+70)

    A little more about me… I am 41 female
    – casual runner but not as much as I used to, currently 1 to 3 miles a couple times a week vs 3-8miles
    – former scuba diver with scuba incident (Navy Dr advised that I over exerted my lungs, had trapped gas in pleural space
    – never smoked/ parents smoked when I was younger
    – avg rest HR high 30’s low 40s
    – avg blood pressure 90’s/40’s

    Over the past couple of years, I started to notice a trend, it seemed like I was hardly breathing when I woke up and it seemed as though I was conscientiously making myself breath when I woke up. It felt as if my lungs were getting their first real breath of air in while. During this time frame I noticed increased coughing throughout the night and I would need to take sips of water. Also, for several years I have had severe coughing after some of my runs. This was especially common on cold days. When the sickies went around, I tend to cough longer than everyone else. All of this led my primary care provider to send me to the pulmonary doctor.

    Thank you for your time and reading my post. I look forward to learn how this forum would interpret and respond to my test results.

    • Sherri –

      FYI, I am a technologist, not a physician, and am best able to comment on your test results, less so on your symptoms. Your pre-bronchodilator FEV1 and FEV1/FVC ratio show you have mild airway obstruction. The post-bronchodilator increase in FEV1 is significant and indicates you have hyperreactive airways, which is a useful synonym for asthma. None of the physicians I work with use any of the instantaneous flows (FEF25%, FEF50% or FEF75%) or the FEF25%-75% although the reasons are slightly different. There are no good reference equations for the instantaneous flows and they’ve never been used as a way of determining a significant response to bronchodilator. The FEF25%-75% was used for this purpose at one time and a significant change was usually considered to be somewhere around 35-40%. Over the last couple of decades however, most physicians have become dis-enchanted with the FEF25%-75% because it is highly dependent on the FVC, highly variable and has little or no clinical relevance not already shown by the FEV1.

      A chronic cough is often a symptom of asthma (but many other things too) and there is something called “cough-variant asthma” where spirometry test results are often normal and the only symptom is a cough that often goes away when the individual is treated with the usual asthma medications (inhaled steroids, short- and long-acting bronchodilators). There is also “exercise-induced asthma” which is relatively common among athletes, and is often diagnosed by hyperventilation challenge testing (both cold air and dry air variations). There is (or at least was since I haven’t heard it discussed in a while) something called “nocturnal asthma” which was when symptoms (cough, wheeze, SOB) only showed up at night. Any or all (or none) of these variants may apply to your symptoms. Since you’ve already shown you improve following a bronchodilator you probably don’t need further testing (methacholine challenge, cold air challenge, eucapnic voluntary hyperventilation challenge or exercise challenge) to confirm a diagnosis of hyperreactive airways, but it’s possible your pulmonary physician would order additional testing if it was important to determine what your specific triggers are. However, I have to point out that even though your symptoms seem to fall mostly along the asthma spectrum that doesn’t mean that something else isn’t going on, but that’s why you’re seeing a pulmonary physician.

      Regards, Richard

  12. Sorry it took so long to reply. Just been a bit busy lately.

    Dr. Quanjer, here is the front page of the web site of QVAR which shows that it is one of the only ICS that reaches the small airways. The reason this is significant and worth mentioning is because through years of studies we know that you are only getting 30% of the medication delivered to the distal airways when using the traditional ICS like Flovent and Pulmicort. The newer drugs like QVAR and Alvesco deliver a much higher concentration to the distal airways.

    http://qvar.com/asthma-control-inhaler-medicine/default.aspx

    One can argue that this is because asthma effects both the small and large airways and the drug was created to be more effective, as opposed to specifically designed for people with isolated small airways disease. However, if isolated small airways disease does in fact exist (normal FEV 1, low FEF 25-75), this is a huge step in the are of pulmonary medicine and may offer relief to those patients.

    I have an interesting question for Richard – I’m sure you have conversations with your patients while performing PFTS. I notice most of them are very talkative during the test. Do you find patients with mild atelectasis of the bottom portion of their lungs have lower PFT readings? Just wondering because I have this condition and always wondered if this can effect readings and may be causing my SOB. Two different pulmonologist have told me this is insignificant. Also, have you found that patients taking certain beta blockers have worse readings? Thanks.

    • Michael –

      I notice that the QVAR webpage does not specify its particle size (or range of particle sizes). I’ll agree that in general smaller particles will be inhaled further down the respiratory tree, but also have to point out that doesn’t guarantee that it’s the smaller airways where they will be deposited and that smaller particles also have less medication in them than larger particles (1/2 the diameter has 1/8 the volume). I’ll also note that the webpage does not cite any studies that show QVAR is more effective than any other inhaled steroid so for the moment I’ll take their claim with a bit of salt.

      I usually have very little information about patients before they are tested and whether or not they have atelectasis is hardly ever something I am told. In addition most patients know very little about their condition, and again atelectasis is not something they tend to know they do or don’t have. Unless there are special circumstances surrounding a patient’s tests I really don’t have the time to look up any information (other than their orders and possibly their medications) about them either. Beta blockers will affect the heart rate response during exercise (and can be a limiting factor due to chronotropic incompetence) but I wouldn’t expect them to make any particular difference on regular PFTs.

      Regards, Richard

Leave a Reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.