Surgical lung biopsies in elderly patients with interstitial lung disease: Weighing up the pros and cons

The process of obtaining a confident interstitial lung disease (ILD) diagnosis has become increasingly nuanced in recent times. Updated guidelines and breakthroughs in therapy have placed more importance on distinguishing idiopathic pulmonary fibrosis (IPF) from non-IPF diagnoses.1-3 The discovery of effective antifibrotic therapy for IPF, and the parallel finding of increased harm with immunosuppression in this same patient subgroup, have emphasized the need for more robust means of phenotyping ILD patients.4-6 The multidisciplinary discussion (MDD) has emerged as the primary tool for amalgamating clinical, radiological and sometimes histopathological data to derive a diagnostic label. This process has been adopted widely and advocated in guidelines, given that no single modality consistently yields a precise diagnosis on its own.

Frequently and frustratingly, the available clinical information is insufficient to enable a confident diagnosis, despite MDD involvement. It is noteworthy that approximately 10% of patients referred to specialist clinics have unclassifiable disease, chiefly due to their inability to undergo invasive surgical lung biopsy (SLB).7, 8 With a 10–15% risk of complications, the decision to proceed to SLB is informed by the patient’s premorbid condition and the value in obtaining such information. In the situation where a patient is elderly and frail, the algorithm is straightforward—a biopsy must be foregone in favour of a pragmatic management approach. In the patient older than 75 years who is otherwise well, however, greater consideration for SLB is needed when high-resolution computed tomography (HRCT) and other clinical findings are inconclusive.

Patient characteristics known to increase the risk of morbidity and mortality of SLB include male gender, worsening respiratory physiology and advancing age.9, 10 In a large retrospective sample of ILD patients undergoing SLB, a steadily increasing risk of in-hospital death was observed with every decade for those over the age of 45 years.10 In the group with patients of age 75–84 years, the adjusted OR for death was 4.46 (95% CI: 2.69–7.40) compared with the under 45-year-old group. In that series, the overall in-hospital mortality was 1.7% for elective procedures, compared with 16.0% for non-elective cases.10 Historical series report 30-day mortality rates between 3% and 16%, with worse outcomes seen in the surgical cohorts that included mechanically ventilated patients and those suffering acute exacerbations just prior to biopsy.11, 12 Considering these findings and the progressive, sometimes unpredictable nature of the diseases at hand, the need for careful patient selection is clear.

In this issue of Respirology, Vaszar et al. report on outcomes in 55 patients aged between 76 and 80 years, who had previously undergone thoracoscopic SLB for ILD diagnosis at two centres.13 Overall diagnostic yield was high at over 96%, with two-thirds of the population given a final diagnosis of IPF. Subjects were further analysed according to HRCT patterns, as defined in the 2011 American Thoracic Society/European Respiratory Society (ATS/ERS) IPF Guidelines: ‘inconsistent with the usual interstitial pneumonia (UIP) pattern’ (seen in nearly 70% of participants), ‘possible UIP’ (in 24%) and ‘definite UIP’ in the remainder.3 Unsurprisingly, all patients with definite UIP on HRCT had concordant histopathology, leading to an ultimate label of IPF. In the group with possible UIP on HRCT, 77% were eventually diagnosed with IPF and in those with inconsistent with UIP HRCT patterns, just over 60% had IPF.

The reported overall 30- and 90-day mortality rates were sobering at 9.7% and 15.4%, respectively. These figures were worse in the IPF subgroup at 14.7% and 20.7%, respectively. Most deaths were attributed to acute exacerbations of IPF, a known risk of thoracic surgery.

In light of these data, the question of whether invasive lung biopsy should ever be performed in patients older than 75 years needs to be asked. As might be expected, in this elderly, predominantly male population irrespective of the pre-surgical label, subjects were statistically most likely to have IPF. Furthermore, this study reports at least a 1 in 10 risk of death within 30 days of surgery. Post-operative morbidity is not evaluated but would likely be substantial. Certainly in practice, there has been a general trend away from SLB since 2011 ATS/ERS IPF Guidelines were published. These recommend against biopsy in those with definite radiological UIP, given the high pre-test probability of IPF.3 The authors emphasize that patients with definite UIP in this group underwent biopsy prior to the 2011 publication. Some commentators less in favour of biopsy have recently proposed that the definition of IPF be expanded, given that disease morphology and/or behaviour pattern may better predict prognosis and response to therapy than a specific disease label. Indeed, within the INPULSIS Phase 3 nintedanib studies, one-third of participants had possible UIP on HRCT without confirmatory biopsies.14 These subjects experienced the same magnitude of therapeutic benefit as those with radiological definite UIP or biopsy-proven disease.

But what of the patients with HRCT patterns inconsistent with UIP? In Vaszar et al.’s cohort, a substantial proportion (40%) of these had non-IPF diagnoses requiring tailored therapies. This finding suggests that the SLB should not be abandoned entirely in the older patient group. In essence, the risk of pursuing a specific diagnosis through invasive means may be acceptable in a select group of patients older than 75 years.

A related issue not addressed in this study is the role of transbronchial lung cryobiopsy (TBLC), a less-invasive procedure gaining traction across many centres. Without any published head-to-head comparisons of TBLC against SLB, it is unclear if the new technique offers a similar degree of diagnostic accuracy. With pooled estimates of moderate-to-severe bleeding in 39% (95% CI: 3–76) and pneumothorax in 12% (95% CI: 3–21%), it would seem unwise to consider TBLC as the “safe alternative” for all elderly ILD patients requiring biopsy.15

In time, we may move away from lung biopsy altogether, as microimaging, biomarkers and techniques of genetic interrogation become more refined. For now, however, histopathology serves as one of the major tools for ILD diagnosis, as long as the patient is well-chosen.

Disclosure statement

L.K.T. has received funding from Covidien Pty Ltd. (a Medtronic Company) towards the Cryobiopsy versus Open Lung biopsy in the Diagnosis of Interstitial Lung Disease (COLDICE) study.