ERS and ATS diagnostic guidelines for primary ciliary dyskinesia: similarities and differences in approach to diagnosis

Methodology and application of the grade system in PCD guideline development

Both ERS and ATS developed guidelines based on pre-defined questions relevant for clinical care, a systematic review of the literature, and assessment of the evidence using the GRADE approach to decide the strength of recommendations (Grading of Recommendations, Assessment, Development and Evaluation) [8–11]. Both committees comprised international multidisciplinary PCD experts, expert methodologists and patient representatives.

For the ERS guideline, the predefined “PICO” question for each diagnostic test was: with Patients suspected of having PCD, Investigated by nasal nitric oxide (nNO), transmission electron microscopy (TEM), high-speed video microscopy analysis (HSVMA), genetics or immunofluorescence labelling, when Comparing patients with a final positive or negative outcome, what was the diagnostic accuracy (Outcome) of the test? To determine the value of each diagnostic test, the task force required studies that mimic the clinical situation. They therefore only included studies with patients referred for PCD testing, in whom the diagnosis was initially uncertain. Case–control studies that compared previously diagnosed typical PCD patients with healthy controls or disease controls were not considered because the situation does not occur in clinical practice.

The ATS objective determined the accuracy of four ancillary diagnostic tests (an extended genetic panel test including >12 PCD genes, nNO, HSVMA and ciliary beat frequency) for diagnosing PCD in patients with a high pre-test probability of having PCD. Studies that recruited patients with at least two of four clinical features (early onset chronic wet cough, early onset chronic nasal congestion, unexplained neonatal respiratory distress, or organ laterality defects) met criteria for high pre-test likelihood of PCD. Committee members agreed on a reference standard diagnosis for PCD: either a classic cilia ultrastructural defect detected on TEM and/or two pathogenic variants in one disease causing PCD gene. Four PICO questions were developed which compared the four ancillary PCD tests as replacements of the reference standard diagnosis. For the nNO PICO question, the population was limited to patients ≥5 years of age with cystic fibrosis ruled out.

Only a small number of published studies met the strict GRADE criteria for inclusion in the guidelines. For example, the ERS systematic search identified 98 manuscripts about nNO, but only four met the inclusion criteria for study design. None of the 462 studies on genetics, nor 276 studies reporting immunofluorescence met the criteria. There is therefore an urgent need for well-designed studies in a diagnostic setting to better inform future revisions of the guidelines.

Summary of the ERS and ATS guidelines

The ERS and ATS guidelines are summarised in figure 1 and comparisons between the two guidelines detailed in table 1.

• Download figure
• Open in new tab
• Download powerpoint

FIGURE 1

a) Adapted from the European Respiratory Society primary ciliary dyskinesia (PCD) diagnostic algorithm [6]. Patients do not necessarily need to go through all steps, and the order of investigations may vary depending on local resources [6]. The clinical importance of the algorithm is the summary of how the combination of tests contributes to the diagnostic outcome: positive, highly likely and highly unlikely. b) Adapted from the American Thoracic Society PCD diagnostic algorithm [7]. ^#: cystic fibrosis (CF) should be ruled out before performing nasal nitric oxide (nNO) measurement, as roughly one-third of CF patients can have nNO values below PCD diagnostic cut-offs. nNO measurements should only be performed with chemiluminescence analysers using standardised protocols at centres with specific expertise in nNO measurements. Some nNO analysers have not received approval from federal agencies worldwide (US Food and Drug Administration and Health Canada), which may have implications for clinical implementation. ^¶: genetic panel testing for mutations in >12 disease-associated PCD genes, including deletion/duplication analysis. ⁺: as nNO levels can be significantly decreased by viral respiratory tract infections, a repeat nNO measurement, at least 2 weeks after the initial low value (expert opinion), is recommended to ensure the initial low value is not secondary to a viral process. A normal nNO value upon repeat testing is not suggestive of PCD, as nNO values remain consistently low in PCD. ^§: most forms of PCD resulting in normal nNO levels have normal or non-diagnostic electron microscopy studies. Thus, genetic testing is recommended in these cases. ^ƒ: or presence of variants of unknown significance. ^##: additional corroborative testing may provide information on clinical prognosis, further disease understanding, and potential future therapeutic considerations. ^¶¶: known disease-associated transmission electron microscopy (TEM) ultrastructural defects include outer dynein arm (ODA) defects, ODA plus inner dynein arm (IDA) defects, IDA defect with microtubular disorganisation, and absent central pair, identified using established criteria [1, 6, 13]. Of note, the presence of IDA defects alone is rarely diagnostic for PCD. ⁺⁺: up to 30% of PCD cases can have normal ciliary ultrastructure on electron microscopy. Consider referral to PCD specialty centre if there is a strong clinical phenotype but all electron microscopy and genetic testing are negative. HSVMA: high-speed video microscopy analysis.

The decision of whether a patient has a definitive diagnosis of PCD is not always straightforward, and both the ERS and ATS task forces concluded that there is no stand-alone test or combination of tests that can exclude PCD with 100% certainty in a patient with strong clinical phenotype. The ERS guidelines determine three diagnostic categories: “PCD confirmed”, “PCD highly likely” and “PCD highly unlikely”. If the diagnosis is “highly likely” but not conclusive, patients should be told that the diagnosis is likely, but given the limitations of diagnostic tests, patients should have other causes for their symptoms investigated whilst receiving treatment for PCD until a definite cause can be established. As new diagnostic tests become available, further investigations should be offered. ATS guideline diagnoses are more binary with “PCD positive” and “PCD negative”.

Explanation of some differences between ATS and ERS guidelines

In box 1 we propose reasons why the use of the same evidence has led to different recommendations. First, the ATS guidelines had a strict clinical criteria compared to the ERS, thus altering pre-test probability of a false-positive or false-negative result. The evidence has also been evaluated on a background of different geographical challenges and healthcare systems, and therefore accessibility, availability and feasibility of testing has impacted the recommendations. For example, in Europe, where HSVMA and TEM equipment is established and expertise is concentrated in national referral centres, these tests are feasible, whereas in North America, access to these tests is more limited as PCD infrastructure is less centralised and spread across a vast and geographically dispersed medical system, where feasibility is challenged through lack of experience across institutions. Conversely, in North America, clinical genotyping is already recognised through insurance systems as a regulated and reliable test. Interestingly, both guideline committees included patient representatives and their different experiences may also have influenced the committee. Possibly, the biggest differences are in the definition of the reference standard and the years that these guidelines were developed. The ATS had an additional year of peer reviewed publications which included a key cohort publication for genetic testing. Also, the ATS used updated GRADE guidelines that have a binary classification of recommendations (strong or conditional), and they were all conditional due to the lack of evidence that a diagnosis of PCD changes long-term health outcomes.

Suggested research required to develop an internationally agreed diagnostic pathway for PCD

Although there are many points of agreement, the disparities between the two guideline approaches are confusing for the end user, and in a rare disease such as PCD, an internationally agreed upon diagnostic testing guideline would be preferable. There are two purposes for the requirements of a clear PCD diagnostic definition: 1) to inform patients of their diagnosis and (where applicable) for insurance purposes to access treatment, and 2) for recruitment into research and drug development trials. Evidence for PCD treatment is severely lacking and usually borrowed from studies in cystic fibrosis or bronchiectasis. Whilst it is reasonable to have a relatively low threshold for starting clinical management for PCD based on low nNO, it must be recognised that there will be false-positive and false-negative cases, and that further confirmatory testing should be explored with genetics and TEM, as recommended in both guidelines. It is paramount for clinical studies that the correct patients are recruited, and only patients with definitive PCD diagnoses are included.

Box 1 highlights some suggestions for steps required to further develop international agreement. Significant research and development is required to identify new testing modalities and improve those already in use. New modalities currently being investigated include radiolabelled mucociliary clearance testing and use of immunofluorescence [21]. Funding for PCD genetic testing across European national health care programmes should also be strongly advocated.

Finally, PCD is a heterogenous group of genetic conditions and it may be that one test or one algorithm will never fit all cases. Large gene discovery programmes and cohort studies such as the European iPCD study and the Genetic Disorders of Mucociliary Clearance Consortium (GDMCC) will in future allow improved understanding of molecular causes of disease and consequently PCD diagnosis [22].

Shareable PDF