CHICAGO – Researchers have derived and validated a 10-item formula to estimate a patient’s risk for developing a major adverse event while on NSAID treatment.
The calculator could “help guide use of NSAIDs in clinical practice,” said, at the annual meeting of the American College of Rheumatology. Although he called for further validation of the risk-score formula using other databases, he noted that it uses readily available data and could easily be calculated with standard inputs in an electronic medical record. The formula predicts the risk for a major adverse effect during 1 year of daily NSAID use.
Dr. Solomon and his associates devised the risk-score calculator with data collected in the(Prospective Randomized Evaluation of Celecoxib Integrated Safety vs Ibuprofen or Naproxen) trial, a safety study designed to test whether treatment with celecoxib was noninferior to treatment with naproxen or ibuprofen for producing cardiovascular adverse events, a hypothesis proven by the study’s results (N Engl J Med. 2016 Dec 29; ). They had full data available for 23,950 of the more than 24,000 enrolled patients. The patients averaged 63 years old, just over a third were men, their average body mass index was 31 kg/m2, and 90% had osteoarthritis and 10% had rheumatoid arthritis. The study enrolled patients with an elevated risk for a cardiovascular event, so 63% had hypertension and 36% had diabetes.
The adverse events included as possible outcomes estimated by the formula were all-cause death, major adverse cardiovascular events, clinically significant GI events, or renal insufficiency or failure. The investigators used data from more than 15,000 patients enrolled during the first 4 years of the study to derive the risk-score formula, and data from the nearly 9,000 patients enrolled during the next 5 years to validate it.
The analysis identified and validated 10 baseline items that, when plugged into the formula, calculated a predicted rate for the subsequent development of a major averse event during 1 year of NSAID treatment. The 10 parameters are: age, sex, known cardiovascular disease, hypertension, diabetes, current cigarette use, on treatment with a statin, baseline serum creatinine level, rheumatoid arthritis, and hematocrit.
As examples of the accuracy of the prediction score, Dr. Solomon reported that, among the patients with a predicted risk for a major adverse event of less than 1%, the observed rate was 0.4%; among people with a predicted rate of 1%-4%, the observed rate was 1.7%; and among those with a predicted risk of more than 4% the observed rate was 5.6%. Major cardiovascular events were the most common type of adverse events observed among the nearly 24,000 patients enrolled in PRECISION. A total of 5% of the patients fell into the lowest risk category, with a risk of less than 1%; 70% were in the intermediate risk category, with a predicted risk of 1%-4%; and 25% had a predicted risk of more than 4%, reported Dr. Solomon, a professor of medicine at Harvard Medical School and a rheumatologist at Brigham and Women’s Hospital in Boston.
Age is a major driver of risk, he noted. A patient who is at least 65 years old would have a greater than 1% risk for an adverse event regardless of the other nine risk factors in the scoring formula.
PRECISION was funded by Pfizer. Dr. Solomon has received research funding from AbbVie, Amgen, Bristol-Myers Squibb, Genentech, and Pfizer.
SOURCE: Solomon D et al. ACR Annual Meeting, Abstract . Arthritis Rheumatol. 2018;70(Suppl 10).