I don’t recommend or use PCR testing. The results of PCR tests are prone to false-positive and false-negative results. PCR is more prone to false reporting than traditional methods of air and surface sampling. To learn the test methods I recommend, read one of my E-Guides or book a Consultation with me.
Using PCR, one of the causes for false-positive results is that the laboratory only looks for partial gene sequences. It’s faster and cheaper than looking at the entire genome. It may be concluded that there's a match, when in fact a different organism is present that has the same partial gene sequence. An example is with Covid-19 testing. In the example that follows, one of the laboratories performing PCR specifies known-overlaps. Although it is a DNA test, the types of mold listed in the report are not necessarily those present, rather they belong to a group of molds that share the same gene sequence:
Genetically closed-related species may be detected in the indicator assay: Aspergillus fumigatus covers Neosartorya fischeri. Aspergillus restrictus covers A. caesillus and A. conicus. Aspergillus niger covers A. awamori, A. foetidus and A. phoenicis. Aspergillus ochraceus covers A. ostianus. Eurotium (Asp.) amstelodami covers E. chevalieri, E. herbariorum, E. rubrum and E. repens. Mucor amphibiorum covers M. circinelloides, M. hiemalis, M. indicus, M. mucedo, M. racemosus, M. ramosissimus and Rhizopus azygosporus, R.homothalicus, R. microsporus, R. oligosporus, R. oryzae. Penicillium brevicompactum covers P. stoloniferum. Penicillium crustosum covers P. camembertii, P. commune, P. echinulatum and P. solitum. Penicillium spinulosum covers P. glabrum, P. lividum, P. pupurescens and P. thomii. Trichoderma viride covers T. atroviride and T. koningii.
False positives may result when primers and probes (chemicals used by the lab as part of the PCR procedure) bind to homologous sequences. Homologous sequences are sequences of genes shared by a common ancestor. All life shares common sequences of genes. All of the possible organisms that share similar genetic sequences to the mold the lab is looking for are not known. Therefore, the lab results may be in error, even if all the QC/QA procedures are followed.
A common reason for false-negatives (mold is present, but not detected) is what are referred to as “inhibitors.” House-hold chemicals are inhibitors. Laboratories state that they have QA in place to detect this, and will notify clients when it occurs. They have not disclosed a warning or a list of know chemicals that cause this. I have a client who received a report for testing done using PCR for which the analyst hand-wrote on the test results:
“Results are skewed due to chemical found in dust sample -- chemical interferes w/ test results. Can’t say whether the mold results should be higher or lower than what’s reported --just that the #’s are skewed.”
The laboratory none-the-less provided an otherwise normal looking report listing the types and levels of mold detected.
I have a client that did several rounds of PCR testing of their home with three different laboratories analyzing the samples. Two out of three laboratories reported relatively normal/low levels of mold; the other extraordinary high levels of Stachybotrys and Aspergillus flavus. I subsequently called the lab that reported high levels of mold, spoke with the analyst, and asked what kind of QC is in place to prevent false-positives. We were not able to determine what went wrong. One thought was the dust in the home. The cause for the false-positives might be low amounts of Stachybotrys and Aspergillus flavus. The error may have occurred in the initial stages of the PCR processing when the lab grinds the dust to isolate the DNA inside the spores. (The DNA inside a spore can not be detected unless the spores is crushed open with a bead grinder). Perhaps the fragments of DNA were of low quality or low quantity.
The following is the result by the first laboratory:
The following are the PCR results from the second laboratory. They show high levels of the Stachybotrys and Aspergillus flavus:
The home owner, concerned about the high levels of mold reported by the second laboratory, collected two more samples, including an outdoor control. These were sent to a third laboratory for PCR. The level of Stachybotrys and Aspergillus indoors is similar to outside and normal compared to the other types of mold detected:
Outdoors / Indoors
After receiving results from the third lab, the home owner collected more samples and sent them to the second lab (the one that reported high levels of mold). This time, Stachybotrys was not detected in the indoor sample. Stachybotrys was detected in the outdoor sample and at a relatively high level (17% of the total). In Whole House sample, there was a high level of Scopulariopsis chatarum (140,000 spore equivalents, 26% of the total). This type of mold was barely detected in the outdoor sample. It therefore seemed an anomaly. The following are the results before the data correction:
I spoke with the lab manager. He found an error had been made in the data entry. The Scopulariopsis chatarum should have been reported as Stachybotrys chatarum. The lab manager corrected the error.
Correcting that error got us back to having a report with high levels of Stachybotrys indoors and outdoors, similar to previous results. The lab manager did not find a similar data entry error for Aspergillus flavus, leaving the cause for the high levels reported for Aspergillus flavus and Stachybotrys chatarum, a mystery.
The lab investigated further and found that the “calibrator” had been set too high, resulting in the levels of all of the molds being reported higher than they actually are. The lab corrected the “calibrator.” After spending days checking the equipment and process, the lab manager wrote back: “I want to start by apologizing. The calibrator was off by about 6 meaning the results were off by ~2^6 when the computer ran the calculations. A huge error. I manually fixed them.”
Correcting the calibrator and redoing the calculations reduced the levels of mold detected across the board. It did not, however, correct the issue with high levels of Stachybotrys and Aspergillus flavus compared to results from the other two laboratories. Contamination is not thought to be an issue because the analyst who performed the PCR checked the results of samples analyzed the prior run and found those did not have high levels of Stachybotrys and Aspergillus.
The following are the revised results. The numbers are lower. I had not noticed before that the levels outdoors for nearly all of the spore types detected, including species of Aspergillus, are lower outdoors than indoors. Without the outdoor reference one might think there is a source of mold contamination indoors. The results are still not in agreement with the findings of the other two laboratories. The lab manager agreed that the level of Stachybotrys detected outside is unusually high. The Stachybotrys detected indoors is less than outdoors, but the level (22% of the total) is a concern. Typically if there is a source of mold contamination indoors, there is more than one type of mold associated with growth. It appears that Stachybotrys is the only type of mold indoors, that indicates there might be growth:
I asked what this lab might be doing differently than the other two. The lab manager told me all of the labs use the same process and chemicals and place the mold detected into groups, meaning the types reported may be groups of mold that share similar gene sequences. (I am not certain that all of the labs used the same primers. An analyst I spoke with at the first lab said they place an order with the supplier based on the “cocktail” of types of mold they want to test for. It’s possible there are variances in stocks, the reference species used. Variances in production are unknown). I found a paper “Amplification of Stachybotrys Gene Using PCR” which found certain stock isolates of Stachybotrys chatarum have sequences similar to species of Penicillium.
I asked if it’s possible there might be another species present that’s being reported as Stachybotrys. This, the lab manager told me, can be checked by doing a “blast search.” The Basic Local Alignment Search Tool (BLAST) https://blast.ncbi.nlm.nih.gov/Blast.cgi. finds regions of similarity between biological sequences. The program compares nucleotide or protein sequences to sequence databases and calculates the statistical significance. I found it too complex to use. I was reminded that a mold can have a different appearance depending on its sexual state, and some species are placed in completely different genera depending on sexual state (example Eurotium / Aspergillus).
I asked the lab manager if there was anything else he knew of that could explain why relatively low levels of Stachybotrys were detected by the other two labs. He said he would “check the curve.”
Meanwhile, furthering my study, I learned that “cross-amplification” and “co-amplification” can occur when an organism such as a mold, pollen, plant DNA, yeasts (a type of fungi), mice or rat cells, and DNA of other organisms. In order to know the effect on the outcome (how these effect the lab results) experiments must be run with and without the contamination along with the target organism (the mold the lab is trying to detect). Is it possible a cross-amplificator and co-amplificator was present? If so why were the results of only the one laboratory affected?
Meanwhile, my client had a sample of the house dust analyzed by culturing it. Some mold inspectors use cultured dust to assess a house for mold contamination and claim to have levels for they can make conclusions. I avoid this practice. I find using dust samples to assess if a house is contaminated with mold isn’t reliable, whether done using cultures, PCR, or other methods. A baseline outdoor reference is required and the surfaces inside and outside must have been cleaned prior to letting the dust collect. We can, however, learn something by comparing the results of the cultured dust to the PCR results since the laboratory cultured the dust sample using three different agars.
Regarding types of agars: Malt Extract Agar (MEA) is used for most fungi. It's what I use to test the air in homes when collecting cultured air samples. It has inhibitors to slow down the growth of fast-growers, so slow growers like Stachybotrys have a chance. DG18 is normally used to isolate Penicillium and types of mold that prefer less water. Aspergillus grows so well in DG18 it can overgrow the plate, preventing other types of mold from sporulating. CMA (Corn Meal Agar) is used to isolate Stachybotrys.
Using PCR to test a house for mold
The EPA tried to develop a method to test homes for mold using PCR. It is known as The Environmental Relative Mold Index (ERMI). After it was introduced, mold remediators used it to convince homeowners remediation was necessary. Homeowners failed to ask what level their house should be to conclude there was no mold contamination or the mold was gone after remediation. After spending tens of thousands of dollars, but not achieving as low of reading as they desired, people complained to the Office of the Inspector General (OIG). In 2013 the OIG required the EPA to release a statement reminding people that it was a research project and not intended for public use. They substantiated allegations that many of the homes tested did not need mold remediation.
Reference: “Public May Be Making Indoor Mold Cleanup Decisions Based on EPA Tool Developed Only for Research Applications.” U.S Environmental Protection Agency, Office of the Inspector General. 13-P-0356. August 22, 2013.
Considering the types of agar used and which molds grow best in each type of agar: Cladosporium was the dominant type of mold that grew in the DG18 agar. Therefore, it seems there must not be a lot of Penicillium or Aspergillus in the house dust. Zero Stachybotrys grew in the CA agar. Therefore, there must not be a lot (or any) Stachybotrys in the dust, as the PCR results suggest. This supports the idea that the PCR results are in error. It also seems there is relatively little Aspergillus flavus reported in the cultured dust. Aspergillus flavus is the other type of mold for which the high levels in the PCR did not seem credible. This supports the idea that the PCR results are in error.
Since PCR detects both viable and non-viable cells, it seems PCR should detect more total mold than cultured samples. However, it’s not credible to suggest that the explanation for the discrepancy between PCR and cultures is that no viable spores are present. The PCR results suggest 21% of the total mold present is of the spore type Stachybotrys. That suggests all of the mold for that spore type is “dead” which would suggest all of the mold is dead. This is not true. Plenty of other types of mold were detected in the culture.
The lab manager did not find an error with the “curve.” He seemed to think the presence of the Stachybotrys (and Aspergillus flavus) was acceptable, as the levels outside are higher than indoors, so outdoors would be the source.
I asked what could be variable between labs. He said it's likely the other labs are using “EPA primers and probes” where as they gets theirs from their equipment manufacturer whom they have an agreement. The significance of this is the lab might be using different stocks (species/strains) compared to the other labs. If that's the case there's a chance something in the dust shares a similar gene sequence, i.e, what's being detected is not what they think it is. This is just one possible explanation. I can't say that it's likely. We are continuing to investigate.
To summarize, PCR is prone to false-positive and false-negative results. The Quality Assurance (QA) and Quality Control (QC) in place at laboratories is not sufficient to detect when a false result occurs and when it does, the reason for it. When one does notify the lab that the results are questionable, the lab will defend the results if they are unable to identify a cause for error.
To understand more about the PCR process in regard to QA and QC, see Quality Assurance/Quality Control Guidance for Laboratories Performing PCR Analyses on Environmental Samples published by the EPA.
Using PCR to Test Air Samples for Mold
There appears to be very little mold reported in outdoor air samples collected and analyzed using PCR compared to traditional spore traps and cultured samples. In one case a PCR air sample showed a count a total of 7 CEs for 7000 liter of air collected compressor to raw counts of 23 spores in the outdoor air for a spore trap. Consider spore traps only collect air for only 5 minutes (75 liters of air). How could sample ten times longer and detect less mold? It seems the transfer of particles from the media (filter paper) used in the PCR process is inefficient.
Regarding the level of Aspergillus fumigatus in samples collected from house dust and analyzed using PCR, notice the subscript in the PCR laboratory reports - there is superscript (b) next to Aspergillus fumigatus. At the bottom of the laboratory report it states: “Includes A. fumigatus and Neosartorva fischeri. The values reported may be either A. fumigatus or Neosartorva fischeri. “ Note: There is a typographical error on the part of this laboratory. The correct spelling of the genus is with a y: Neosartorya.
In the paper “Reproduction in Aspergillus fumigatus: sexuality in a supposedly asexual species?” the authors write: “A. fumigatus hasclose taxonomic relatives with described sexual states within the genus Noesartorya.” (Dyer and Paoletti. Medical Mycology, Volume 43, Issue Supplement 1, January 2005, pages S7-S14.) The significance of this is that a PCR report can not be certain there is Aspergillus fumigatus in samples vs other species.
PCR testing does not fall under that category. PCR is licensed to laboratories by the EPA. Laboratories authorized to use the technology must pay a royalty to the EPA for each sample analyzed. A license to use the technology does not mean the laboratory is proficient (accredited) using it. At the bottom of each of the PCR lab report, the laboratory states the following: Method ID: US Environmental Protection Agency licensed technology for mold specific quantitative polymerase chain reaction (MSQPRC) analysis. Not within the AIHA scope.
If you need help knowing how to test for mold, book a consultation.
REFERENCES
“(NAHA) as a Marker of Fungal Cell Biomass” Ragnar Rylander. International Journal of Environmental Monitoring and Analysis. 2015 3(4) 205-209.
“Mycotoxins.” J.W. Bennett and M. Klich. Clinical Microbiology Reviews, July 2003, p. 497-516.
“Public May Be Making Indoor Mold Cleanup Decisions Based on EPA Tool Developed Only for Research Applications.” U.S Environmental Protection Agency, Office of the Inspector General. 13-P-0356. August 22, 2013.
Quality Assurance/Quality Control Guidance for Laboratories Performing PCR Analyses on Environmental Samples. The United States Envionmental Protection Agency (EPA), 2004.
“Airborne enzyme measurements to detect indoor mould exposure.” Ragnar Rylander, Morten Reeslevb and Thomas Hulander. Journal of Environmental Monitoring. 2010. DOI: 10.1039/c0em00336k.
Quantitative PCR for the Detection and Quantitation of Environmental Microorganisms Basics and Applications. Chin S Yang. EMLab P&K. 2004.
“The MIQE Guidelines: Minimum Information for Publication of Quantitative Real-Time PCR Experiments.” Stephen A. Bustin, Et al. Clinical Chemistry. 55:4 611–622 (2009).
THE ENVIRONMENTAL TECHNOLOGY VERIFICATION PROGRAM, ETV Verification Statement, TECHNOLOGY NAME: Mycometer®-test. https://archive.epa.gov/nrmrl/archive-etv/web/pdf/p100dzkz.pdf.
“Partial N Gene Sequencing for SARS-CoV-2 Verification and Pathway Tracing.” Sin Hang Lee, Jonathan McGrath, Stephen P Connolly, and John Lambert. https://pdfs.semanticscholar.org/148b/1b242c7a36ba3313be88d7e4a16fe52c30e3.pdf?_ga=2.19754106.313842569.1645838281-738510753.1644813797
Bioaerosols: Assessment and Control, Janet Macher, Ed., American Conference of Governmental Industrial Hygienists, Cincinnati, OH (1999).