Ultrasensitive C-reactive protein as a prognostic biomarker for cardiovascular disease

PICHON RIVIERE, A.; AUGUSTOVSKI, F.; GARCIA MARTI, S.; ALCARAZ, A.; GLUJOVSKY, D.; LOPEZ, A; REY-ARES, L.; BARDACH, A.; CIAPPONI, A; URTASUN M; COMANDE D; BRITO, V

Documento de Evaluación de Tecnologías Sanitarias

IECS

Año: 2013 p. 1 - 30

1668-2793

Cardiovascular disease (CVD) is the main cause of death worldwide both in developed and developing countries; it accounts for 32% of total deaths in Argentina. In the last decade, several markers have been assessed as potential cardiovascular event (CVE) risk factors in the asymptomatic population, as well as prognostic factors in patients with CVD. Ultra- sensitive C-reactive protein (US-CRP) is one of the most studied biomarkers due to its role in the inflammatory stage of the vascular lesion process that accompanies atherosclerosis. Technology CRP is one hepatic synthesis plasma globulin generally present at levels less than 1 mg/dL, which during the acute phase of inflammatory processes may increase up to more than one hundred times its baseline value. The techniques which are conventionally used to measure plasma CRP levels during inflammatory or infectious processes have a detection limit in the range of 3 to 5 mg/L. The levels observed in healthy subjects are well below the above mentioned range. This has led to the development of high-sensitivity techniques such as the laser nephelometry and immunoturbidimetry to increase the detectable signal of the marker known as US-CRP. Purpose To assess the available evidence on the usefulness of ultra-sensitive C-reactive protein as an independent risk factor in the asymptomatic population and as prognostic factor in the population with cardiovascular disease. Methods A bibliographic search was carried out on the main databases (MEDLINE, Cochrane, DARE, NHS EED), on Internet general search engines, in health technology evaluation agencies and health sponsors. This search was initially restricted to systematic reviews (SR), clinical practice guidelines (CPG), health technology assessments (HTA) and coverage policies (CP). In a second stage, the search was focused on observational studies and randomized clinical trials (RCT) published after the search dates when the last SR were found. Results The association of US-CRP as an independent risk factor or prognostic marker was assessed in the following three population groups: Population with unknown coronary disease Two SR published in 2009 and eight more recent cohort studies were found and, most of them, showed a significant association between high levels of US-CRP and the risk of CVE and death, even after adjusting by the conventional cardiovascular risk factors (CRF). Adding US-CRP to the conventional risk estimate models, e.g., the Framingham score, could re-categorize the risk in 11% to 28% of the subjects with intermediate risk of events at 10 years. Population with stable coronary disease (SCD) One SR published in 2010 and four cohort studies published at a later date were found; they showed an association between the US-CRP level and general mortality, fatal and non fatal CVE. However, including US-CRP in a risk assessment model of events consisting of conventional CRF could slightly increase its power for CVE risk discrimination. Population with acute coronary syndrome (ACS) Two SR published in 2011 and three cohort studies of later date were found; these allowed identifying an association between high levels of US-CRP and general mortality, cardiovascular mortality (CVM) and CVE, even after adjusting by conventional CRF. However, including US-CRP to the TIMI risk score, one of the classic models in prognostic assessment after an ACS, did not improve the individual risk estimate after adding markers such as troponin I and ejection fraction. Clinical Practice Guidelines and Coverage Policies Four clinical practice guidelines (CPG), three health technology assessments (HTA) and four coverage policies (CP) from other countries were found. For the population without CVD, most of them considered that US-CRP- was a coronary disease (CHD)-independent predictor of moderate sensitivity when compared with conventional CRF. There was consensus on recommending US-CRP determination to define statin therapy in men of or over 50 years old or women of 60 years old or older with LDL-c < 130 mg/dL levels, who did not receive hormone therapy replacement, hypolipidemic or immune-suppressive drugs or who did not have chronic inflammatory diseases, diabetes or CVE. US-CRP level determination was considered acceptable when estimating the individual risk of CVE in asymptomatic 50-year-old men or younger and in 60-year-old women or younger with intermediate risk of CVE at 10 years based on the conventional models. It is not recommended to assess US-CRP levels in asymptomatic individuals at high cardiovascular risk and in males below 50 years old or women 60 years old or younger at low risk of CVE at 10 years. Conclusions There is large high quality evidence about the usefulness of US-CRP as CVE risk predictor in all the populations assessed. However, despite there is no direct evidence whether the US-CRP determination can improve patient prognosis, there is consensus on the fact that by re-categorizing the risk and altering the therapeutic approaches, there may be benefits for the group of patients without known coronary disease and with intermediate risk of events at 10 years. The existing evidence to date highlights the usefulness of US-CRP at the time of defining statin therapy in men and women without CHD, 50 years old or older and 60 years old respectively, in the presence of LDL-