Computer Algorithm Highly Accurate at Detecting PAH in Scleroderma

Forest Ray PhD avatar

by Forest Ray PhD |

Share this article:

Share article via email
computer algorithm showed screening potential

A computer algorithm called DETECT correctly identified all cases of a potentially life-threatening type of high blood pressure in a group of people with scleroderma in a recent study.

The algorithm outperformed standard methods, showing potential as an effective means of screening SSc patients for developing pulmonary arterial hypertension (PAH), according to researchers.

The investigators noted that PAH — which is often not identified using standard methods such as ultrasounds of the heart — benefits most from early treatment.

“These ultrasounds miss around one in three patients who may have pulmonary arterial hypertension,” Dinesh Khanna, MD, the study’s senior author and the director of Michigan Medicine’s Scleroderma Program, said in a university press release.

Khanna and his colleagues at the University of Michigan created and tested the computer algorithm — appropriately called DETECT — to determine earlier a patient’s risk for developing PAH.

The results of their study, “Performance of the DETECT Algorithm for Pulmonary Hypertension Screening in a Systemic Sclerosis Cohort,” were published in the journal Arthritis & Rheumatology.

Scleroderma, or systemic sclerosis (SSc), involves the fibrosis or scarring of multiple organs, inflammation, and damage to blood vessels. These effects contribute to PAH, a rare type of pulmonary hypertension, in some 10% to 12% of patients, according to researchers.

PAH is a leading cause of mortality in scleroderma and must be caught and treated early for best results. Echocardiograms — ultrasounds of the heart — are often used to screen patients for PAH but can be inaccurate when evaluating asymptomatic individuals or when early in the condition’s progression. Indeed, according to Khanna, such ultrasounds can miss evidence of PAH in about one-third of scleroderma patients.

The DETECT algorithm is based on evidence from a multicenter study comparing various clinical variables to right heart catheterization or RHC — considered the “gold standard” for PAH diagnosis. A number of medical societies have recommended DETECT for early screening of PAH in scleroderma, and it was included in the 2015 screening guidelines from the European Society of Cardiology/European Respiratory Society (ESC/ERS).

The algorithm performs its analysis in two steps. First, it analyzes six clinical variables to determine whether a patient needs an echocardiogram. Then, the second step incorporates that result into an evaluation of whether that patient should be referred for RHC.

In the current study, the investigators identified 68 adults (including 58 women) with a mean age of 60, who had undergone RHC prior to March 14, 2019, and had enough clinical information to calculate a DETECT score. Of these patients, 10 had been diagnosed with PAH.

Based on the group’s data, DETECT accurately selected all 10 individuals with PAH, with no false negatives (determining that someone does not have PAH, when in fact, they do). By comparison, the echocardiogram resulted in two false negatives.

“It didn’t miss a single patient; it can’t get better than that,” Khanna said. “This is a highly sensitive screening tool and can be very useful.”

The algorithm also showed superior performance among participants with a predicted diffusing capacity for carbon monoxide — a lung function assessment of how much oxygen is transferred from the lungs to the bloodstream — of at least 60%.

However, among those with signs of pulmonary hypertension according to DETECT, only 20% actually had the condition as analyzed by RHC.

“That’s the trade-off of having such a sensitive test,” Khanna said. “The right heart catheterization is invasive, but because the mortality of [pulmonary arterial hypertension] is so high, and the prevalence is so high, the benefits outweigh the risks.”

The researchers said this was the first group study to assess the performance of the DETECT algorithm.

“The DETECT algorithm is a better screening tool for SSc-PAH than [echocardiograms],” the investigators concluded.

Khanna noted that PAH is a leading cause of death for scleroderma patients.

“Early diagnosis and treatment of pulmonary arterial hypertension will lead to better outcomes, including improved quality of life and survival in people with scleroderma,” he said.