高血压论坛 坛主  朱鼎良 高平进 
     高血压是心脑血管疾病最重要的危险因素之一，如何通过对高血压人群进行有效干预来降低心脑血管疾病的发病率、致残率及致死率，是我国目前面临的重大的公共卫生问题。本论坛将从基础研究到临床诊断全面综述高血压诊治进展，精彩内容，值得期待。

     In this presentation， we will focus on the prevention of stroke recurrence and new-onset diabetes mellitus.
Prevention of stroke recurrence
    The PRoFESS trial involved 20，332 patients with ischemic stroke. The median interval from stroke to randomization was 15 days. During a mean follow-up of 2.5 years， BP was 3.8/2.0 mm Hg lower on telmisartan than on placebo. 880 patients (8.7%) on telmisartan and 934 patients (9.2%) on placebo had a subsequent stroke (hazard ratio [HR]， 0.95; P=0.23). Major cardiovascular events occurred in 1367 patients (13.5%) in telmisartan group and 1463 patients (14.4%) in placebo group (HR， 0.94; P=0.11). In an analysis involving 1141 strokes that occurred 6 months after randomization， stroke recurrence decreased by 12% (P=0.04). The PRoFESS investigators proposed that randomization of patients soon after the qualifying event might explain the null results of their trial. However， the odds of stroke recurrence were similar in patients randomized within 10 days of the qualifying event (0.92; P=0.19) and in those randomized later (0.93; P=0.18). The P-value for interaction was 0.84. 
     For a better interpretation of the PRoFESS results， we combined the results of 10 trials. Overall， the odds ratio (OR) for the prevention of stroke recurrence by BP lowering therapy was 0.78 (P=0.0007). There was significant heterogeneity between studies. The pooled OR was 0.63 (P<0.0001) for trials involving diuretics as a component of therapy， but only 0.93 (P=0.086) for trials，in which the mainstay of treatment consisted of inhibition of the renin system by atenolol， perindopril， ramipril， candesartan， or telmisartan. The weighted reduction in systolic BP averaged 9.6 mm Hg in 4 studies of diuretics， 4.0 mm Hg in 5 studies of renin system inhibitors， and 5.1 mm Hg in all studies combined. In metaregression， the weighted correlation coefficient between the OR for stroke recurrence and the BP reduction was –0.57 (p=0.067). The significant heterogeneity (P<0.0001) between diuretics and renin system inhibitors in the prevention of stroke recurrence might therefore be explained by the greater BP reduction on treatments including diuretics.  
     In the PRoFESS trial， 6822 patients with a SBP at entry of 135 mm Hg or less had an OR for stroke recurrence on active treatment vs placebo of 1.04 (P=0.63).  Similarly， in the TRANSCEND trial， 1955 patients with a SBP at entry of 133 mm Hg or less did not experience benefit from BP lowering therapy in terms of the primary and secondary composite endpoints. In contrast， the 913 PATS patients， whose BPat randomization was less than 140 mm Hg systolic and 90 mm Hg diastolic， had on treatment with indapamide a 50% lower risk of stroke recurrence (P=0.03).
Prevention of diabetes mellitus 
     A network meta-analysis evaluated the incidence of new-onset diabetes according to the drug class used to initiate BP lowering therapy. With diuretics as the comparator， the OR were 0.57 (P<0.0001) for ARBs; 0.67 (P<0.0001) for ACEIs; 0.75 (P=0.002) for CCBs; 0.77 (P=0.009) for placebo; and 0.90 (P=0.30) for βblockers. In a prospective observational study of hypertensive patients (median follow-up， 6 years)， not confounded by previous treatment， new-onset diabetes and having diabetes already at baseline carried similar cardiovascular risk.  However， ALLHAT (follow-up， 4.9 years)， VALUE (4.2 years) and ASCOT (5.5 years) did not report a significantly elevated risk in patients with new-onset diabetes. Assuming an absolute risk of 10% over 5 years on older drugs and a relative benefit on the newer drugs of ~30%， about 1000 patients would have to be treated for 1 year with the newer agents to avoid ~6 new cases of diabetes. An important caveat in the interpretation of this estimate is that in most trials included in the network meta-analysis new-onset diabetes was not a predefined endpoint， that follow-up was relatively short， and that in some trials the comparator was a diuretic or βblocker. 
 In PRoFESS and TRANSCEND， diabetes was a predefined secondary endpoint and the comparator was placebo. In PRoFESS， the number of patients who had new-onset diabetes after randomization was 125 of 7360 (1.70%) in the telmisartan group， as compared with 151 of 7283 (2.08%) in the placebo group (HR， 0.82; P=0.10). In TRANSCEND， the number of patients who had a clinical diagnosis of new-onset diabetes was 209 of 1895 (11.0%) in the telmisartan group， as compared with 245 of 1913 (12.8%) in the placebo group (HR， 0.85; P=0.081).  Estimates indicate that 1000 patients would need to be treated for 1 year with an ARB instead of a metabolically neutral drug (placebo) to prevent just 2 cases of new-onset diabetes. However， in both trials， the HRs for new-onset diabetes included unity and the absence of any beneficial effect of telmisartan on new-onset diabetes cannot be excluded. 
Conclusions
     In contrast to the recommendations in some guidelines， the above results do not support the use of renin system inhibitors for the prevention of stroke recurrence. Furthermore， lowering BP in patients with a history of stroke and a normal BP cannot be recommended， because the published results are contradictory. The evidence from recent trials does not change the conclusion from a systematic review published in 2005 that currently no single (antihypertensive) agent can be recommended for the prevention of impaired glucose tolerance or diabetes.

(References available upon request)

黄河厅 10:30-11:00