The algorithms that are used to calculate the future risk of a heart attack and to decide which patients need clinical interventions may underestimate risk in people with HIV, the European AIDS Conference was told recently.
Dr Giovanni Guaraldi of the University of Modena in Italy said that 41% of a group of patients he studied qualified for medical interventions such as statin therapy to prevent heart disease, on the basis of a direct electron-beam tomography (CAT) scan which detected calcification and hardening of the arteries.
In contrast two different heart disease algorithms, the European Society for Hypertension Guidelines (ESH) and the Framingham Risk Score (FRS), suggested medical intervention for 33% and 35% of patients respectively, thus omitting 6%-8% of high risk patients.
The European and Framingham guidelines were particularly likely to miss younger and/or female patients with hardening of the arteries, Dr Guaraldi said.
Guaraldi said: “At present we have to stratify patients according to the Framingham algorithm, but this is not validated in HIV patients and may underestimate disease.”
What should be done for patients with an ‘intermediate’ risk (which means a 10-20% chance of a cardiovascular disease event in the next ten years) is particularly unclear, he added, and imaging studies could help to identify members of this group who needed interventions.
Guaraldi’s team studied 724 people with HIV. Seventy-two per cent were male and their average age was 47. He calculated ESH and FRC scores for them and gave all of them CAT scans. The overall Framingham risk, that is the risk of a heart attack in the next ten years per individual, averaged over the whole group, was 7.5%.
The FRS score was low in 71% of patients, intermediate in 22% and high in 7% of patients: in contrast the CAT scan predicted low risk in 66%, intermediate in 24% and high risk in 10% of patients.
In only 83% of cases did the FRS and CAT risk categories agree with each other and the ESH guidelines were even worse, with only 63% agreement.
The discrepancies were particularly pronounced in female and younger patients (‘younger’ in this context means under 55 if male and under 65 if female).
Ten per cent of people with a low FRS had a high CAT scan score, he added. If the CAT score was turned into an age equivalent, a high score would mean that a man of 30 had a biological artery age of 56. If this was used as the age in algorithms, then the proportion of people not qualifying for prevention interventions despite a high artery calcification score was 3.5% according to FRS but only 1% if CAT scan results were used to determine medication.
The EACS Treatment Guidelines, just reissued, recommend that the FRS score be the standard used to decide on medical interventions for people at risk of heart disease, and Dr Guaraldi suggested at the guidelines session that imaging should be used wherever possible to complement this.
In his own session he was asked if the technique was applicable in clinical practice.
He said: “Maybe not everyday but it is important to gain more data, as we are probably under-treating our HIV population for [cardiovascular disease] prevention.”
Guaraldi G et al. Moving from risk factor assessment to atherosclerosis imaging to select the most appropriate patient for primary prevention. 12th European AIDS Conference, Cologne. Abstract BPD 2/2. 2009.