The accuracy of idiopathic pulmonary fibrosis (IPF) diagnoses is improving with the use of multidisciplinary team meetings and updated guidelines, based on the findings of a study that compared diagnostic agreement of individual clinicians and teams evaluating patients with interstitial lung disease.
Pulmonologists who participate in multidisciplinary team meetings said the findings validate the team approach.
“The [study’s] data confirm what we see in clinical practice … it takes a multidisciplinary – and perhaps often multiple pulmonologists – to review these cases,” Marilyn K. Glassberg, MD, professor of medicine and surgery and director of the interstitial lung disease program at the University of Miami Health System, said in an interview.“This study demonstrates the importance of multiple perspectives when evaluating a patient and coming to a diagnosis at a time when reliable biomarkers are not available.”
The study , published in The Lancet Respiratory Medicine (2016;4[7]:557-65), is the first evaluation of multidisciplinary team agreement on diagnosis of interstitial lung disease since updated guidelines were published, according to Simon L. F. Walsh, MD, of Kings College Hospital NHS Foundation Trust, London, and his colleagues.
In 2015, the American Thoracic Society (ATS), European Respiratory Society (ERS), Japanese Respiratory Society (JRS), and Latin American Thoracic Association (ALTA) adopted joint guidelines for the treatment of IPF . In 2013, the ATS and ERS updated guidelines for the classification and terminology for idiopathic interstitial pneumonias.
“Our study shows … in [IPF], MDTMs (multidisciplinary team meetings) have a higher level of agreement on diagnoses, assign diagnoses with higher confidence more frequently, and provide diagnoses that have non-significant greater prognostic separation than do clinicians or radiologists in most cases,” the researchers wrote.
Before MDTMs were initiated, the clinicians, radiologists, and pathologists who would be participating in them independently reviewed each patient’s case without consulting other specialists and provided up to five diagnoses with diagnostic likelihoods for each patient.
For the study, 70 patients were evaluated and the level of diagnostic agreement was assessed at seven international centers for the diagnosis of interstitial lung disease (diffuse parenchymal lung disease). Following independent reviews of the 70 cases, the clinician, radiologist, and pathologist from each center met as a multidisciplinary team to review the same cases together and give up to five diagnoses with diagnostic likelihoods.
All clinical information supplied in the first stage of the study, including pulmonary function test results, high-resolution CT at presentation, and digitalized surgical lung biopsy slides, were available to the multidisciplinary team. The patients’ outcomes were used to validate the diagnoses. The survival period for each patient was calculated based on the date of referral to the host institution to the minimum date of death, date patient was last known to be alive, or June 1, 2015 – the end of the study period.
The inter-MDTM agreement was better than interobserver agreement for all diagnoses (unweighted kappa value (K) = 0.50), and inter-MDTM agreement was highest for IPF (K = 0.60) and connective tissue disease-related interstitial lung disease (K = 0.64).
“We have shown an acceptable level [based on a K of greater than 0.40 being deemed clinically acceptable] of diagnostic agreement exists between multidisciplinary teams in the setting of diffuse parenchymal lung disease. Additionally, we showed that this agreement was validated by the nonsignificant increases toward greater prognostic separation of an IPF diagnosis made by multidisciplinary teams than by individual clinicians or radiologists,” the researchers wrote.
The weighted kappa (KW) values for estimation of diagnostic likelihood for diagnoses of IPF were 0.72 (0.67-0.76) for clinicians, 0.60 (0.46-0.66) for radiologists, 0.58 (0.45-0.66) for pathologists and 0.71 (0.64-0.77) for MDTMs.
For connective tissue disease–related interstitial lung diseases, the KW for estimation of diagnostic likelihood for diagnoses for MDTMs were 0.73 (0.68-0.78), compared with 0.76 (0.70-0.78) for clinicians, 0.17 (0.08-0.31) for radiologists, and 0.21 (0.06-0.36) for pathologists.
Krishna Thavarajah, MD, , who sees patients with interstitial lung disease within the Henry Ford Health System in Detroit, has been participating in MDTMs for nearly 6 years.
“The accuracies of diagnoses for patients with IPF are much better than even 10 years ago,” she said in an interview. “I think this is because of the improvement in consistency in diagnostic criteria based on the updated guidelines in IPF. Among the MDTMs that participated in the study, the agreement about diagnoses was highest for IPF. The interobserver agreement for clinicians was also pretty high for IPF.”
In her work within an academic center, Dr. Glassberg sees patients in an IPF clinic and in a separate autoimmune disorders clinic. For each clinic, there is a multidisciplinary team. In the IPF clinic, there are three pulmonologists and a radiologist, and when there is a biopsy, there are two pathologists. Dr. Glassberg’s IPF team also includes four pulmonary radiologists.
During her MDTMs, Dr. Thavarajah, a radiologist, and a pathologist will examine a patient’s chest imaging and pathology slides. They sit together until they become confident of their diagnosis in the absence of a biopsy.
There are times when the team tells a patient the probable diagnosis and acknowledges the small chance of an alternative diagnosis. “It was comforting to me that, in the Lancet study, there was a good level of agreement in diagnosis of IPF among multidisciplinary teams, whether the patients had undergone a biopsy or not,” said Dr. Thavarajah. “The mortality of patients given a diagnosis of IPF was worse than those given a diagnosis of non-IPF to validate the IPF diagnosis.”
Establishing and implementing MDTMs is challenging, though, said Dr. Glassberg.
“[We] need to address how multidisciplinary teams could work for doctors who are in smaller cities or who are not in academic centers. We need to utilize existing channels to create new avenues for these colleagues to present their cases – particularly challenging ones or patients who need to be referred – to be evaluated by an interdisciplinary team. The Internet may offer these opportunities for networking and decision making, said Dr. Glassberg.
The study was funded by the National Institute of Health Research, Imperial College London. Several of the study’s authors declared receiving personal fees, grants, or research support from a variety of sources, but had no financial disclosures relevant to this study.
