FROM THE JOURNAL OF THORACIC AND CARDIOVASCULAR SURGERY
Surgical lung biopsy performs well and is relatively safe for evaluating suspected interstitial lung diseases, but may be especially helpful in confirming the diagnosis and directing the treatment of patients with idiopathic pulmonary fibrosis with atypical signs and symptoms, researchers have found.
In patients with immune disorders or severe respiratory dysfunction, or on mechanical ventilation, clinicians should weigh the diagnostic benefits of surgical lung biopsy (SLB) against its potential risks, according to a systematic review and meta-analysis of 23 studies published between 2000 and 2014, comprising 2,148 patients. Dr. Qian Han of the Guangzhou Institute of Respiratory Disease in China led the investigative team. The findings were published in the May issue of the Journal of Thoracic and Cardiovascular Surgery.
The meta-analysis focused on diagnostic yield of biopsy samples and postbiopsy mortality within 90 days of surgery (J. Thorac. Cardiovasc. Surg. 2014 [doi:10.1016/j.jtcvs.2014.12.057). The mean age of patients across the studies ranged from 36 to 62 years. The population of the meta-analysis included 1,632 (76%) who had undergone video-assisted thoracic surgery (VATS) and 268 (12.5%) who had open-lung biopsy.
Slightly more than one third (33.5%) of diagnoses involved idiopathic pulmonary fibrosis, followed by nonspecific interstitial pneumonia (12%), hypersensitivity pneumonitis (9.6%), cryptogenic organizing pneumonia (7.5%), sarcoidosis (6.8%), and connective tissue disease related to interstitial lung disease (4%).
The median diagnostic yield across all studies was 95%, ranging from 42% to 100% depending on the study. One study showed a diagnostic yield below 70%. Eight studies showed that the biopsy influenced a change in the treatment plan 42%-90% of the time. In the entire meta-analysis, treatment plans were altered for 59.5% of patients who received a specific diagnosis and in 55.2% of those without a definitive diagnosis.
“These results suggested that an alteration in treatment may not be directed by a definitive histological diagnosis and nonspecific histological results could also be useful in clinical practice,” Dr. Han said.
Eleven of the studies used CT guidance to obtain biopsies without a preference to lobe, but two studies predisposed to the right lobes had diagnostic yields of 84% and 94%. One study avoided the lingual or middle lobe, with a diagnostic yield of 97%, and another focused on the lingular lobe only, with a 100% yield. Two studies showed that biopsy samples from lingual or middle lobes had the same diagnostic yield as did those from other lobes.
With regard to diagnostic performance based on biopsy numbers, one study showed that multiple biopsies may increase the diagnostic yield. Twelve studies obtained one to three samples, and among them eight studies suggested one sample containing both lesion and normal tissue was sufficient to represent pathological changes. One study showed that multiple biopsies may increase diagnostic yield.
Of the 16 studies that provided sufficient data on mortality rates, the pooled 30- and 90-day mortality rates were 2.2% (95% CI 1.0-4.0%) and 3.4% (95% CI 1.8-5.5%), respectively. The composite postoperative mortality rate was 3.6% (95% CI 2.1%-5.5%).
In their discussion, Dr. Han and colleagues addressed the controversy surrounding lingual vs. middle lobe biopsy by noting that high-resolution CT (HRCT) can be valuable in identifying the appropriate biopsy location. While multiple studies supported the effectiveness of only one biopsy as long as it contained both normal and abnormal tissue, the researchers pointed out that future studies evaluating biopsy samples would do well to evaluate biopsy number combined with sample size.
HRCT, while highly specific, may be less sensitive than SLB in the diagnosis of idiopathic pulmonary fibrosis, they found. Two studies the meta-analysis looked at compared the diagnostic yield between SLB and HRCT; SLB finally diagnosed idiopathic pulmonary fibrosis in 75%-91% of suspected cases and in 19%-74% of cases when HRCT did not raise suspicion of the disease. “These findings suggested that HRCT, albeit highly specific, is less sensitive in the diagnosis of IPF, therefore necessitating the utility of SLB in the diagnosis of these HRCT-omitted cases,” Dr. Han said.
On the safety issue, while studies that excluded patients on mechanical ventilation reported lower mortality rates and two studies identified ventilator dependence as an independent risk factor for mortality, the investigators reported that the higher mortality rates were probably the result of a sicker patient population rather than the SLB procedure itself. They wrote that to “indiscreetly refuse” to perform SLB in these patients is “overcautious and inappropriate” given the benefits of SLB in validating diagnoses and influencing treatment plans.
Dr. Han and his colleagues reported having no relevant disclosures.
