The ATS/ERS have published standards for most Pulmonary Function tests. These standards include the criteria that are used for assessing test quality. What these standards don’t address however, is how poor a test’s quality can be yet still be acceptable for reporting.
We’ve all had patients that aren’t able to perform tests correctly. There can be many reasons why this happens. Lack of comprehension due to cultural or language barriers. Dementia. Extreme shortness of breath. Fatigue. Neuromuscular disease. Whatever the reason, you’ve done the tests with a patient and gotten the best you think you can get, how bad do they have to be before you don’t report them?It could be argued that any test results that do not meet ATS/ERS criteria should not be accepted or reported. If this was the case, however, many of our patients would never have any test results reported at all so there has to be some leeway. An opposing argument could be that any test result should be accepted because it at least shows the minimum a patient is capable of. But because poor quality results can imply the presence of a serious lung disease when it may not actually be present there has to be a point at which results should not be accepted.
I am sure we all overlook minor faults where tests don’t meet strict ATS/ERS criteria. Spirometry efforts that aren’t six seconds long. Diffusing capacity tests with a slightly low inspired volume. Lung volume tests where the SVC is lower than the FVC.
So where do we draw the line? At one extreme there is a monthly ALS clinic that usually sends over a half dozen patients in a single morning for spirometry (upright and supine) and MIPS/MEPS to the PFT Lab I am associated with. Once ALS starts to advance patients have a great deal of difficulty performing spirometry but as long as they are willing to try we will accept whatever they can do no matter how poor the quality is.
At the other extreme should be disability evaluations or their equivalent. I haven’t been involved in these for a while but at the first hospital I worked there were probably about a dozen a year. I think this is an area where any inability to meet strict criteria for both test quality and reproducibility should carefully documented and adhered to.
But in-between there is a lot of gray. In general I think that the guidelines should be to accept and report results when for physical or neurological reasons the result is the best the patient can do; when the error does not significantly alter the interpretation (i.e. still within normal limits); or when despite the error the results rule something out that is relevant to the patient.
For example a new patient with poor quality spirometry and the FEV1 or FVC is high enough (usually within normal limits) to rule out significant airway obstruction or restriction may be acceptable. For a returning patient when the FVC or FEV1 results are (significantly) better than they were at their last visit may also be a good reason for acceptance.
For the diffusing capacity test the most common error is probably a low inspired volume. VA is dependent both on inspired volume and how well the inhaled gas mixture is diluted in the lung. If the inspired volume is low but the VA is either near the TLC or within normal limits then the DLCO can probably be accepted.
I don’t think there is a lot of wiggle room for helium dilution or nitrogen washout lung volumes though. Most errors with these tests lead to an overestimation of FRC or RV and the degree of this overestimation is always unknown. On the other hand if the FRC or RV measurement is okay but the problem is the SVC component it may be worthwhile to report the results.
Testing system software can pinpoint many quality issues but I think it is a mistake to rely on it for anything more than simple checks. For this reason anybody that performs pulmonary function testing really needs to know the quality criteria for all tests they perform and should comment on test quality in the technician notes.
The policy in the PFT Lab I am associated with is that if it can’t be reported it can’t be billed. This is not an incentive to report bad results and is in fact a reason for the technicians to be cautious. They know that once a test has been billed it is very difficult to get it un-billed so they are more likely to bring questionable results to the lab manager or to the ordering physician’s attention before reporting them.
I don’t think that it is possible to come up with firm guidelines about when to accept low quality test results. Where the line is drawn for one patient is likely not going to be in the same location as it is for another patient. At some point questionable test results will have to be passed on to a reviewer. I think that feedback from the reviewer to the technician that performed the tests is important. The more a technician knows about test quality and when results should or should not be accepted the more likely they are to be able to improve test quality in the first place.
PFT Blog by Richard Johnston is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.