XSOLVD: 12-year follow-up of SOLVD trials shows continued mortality benefit of enalapril

Susan Jeffrey

September 06, 2002

Berlin, Germany - Extended follow-up of the landmark Studies of Left Ventricular Dysfunction (SOLVD) prevention and treatment trials shows treatment with enalapril improved 12-year survival for these patients with LV dysfunction.

Results of the Extended SOLVD study, dubbed XSOLVD, were presented here by Drs Philip Jong and Salim Yusuf (McMaster University, Hamilton, ON) at the European Society of Cardiology Congress 2002.

The benefits of enalapril seen in the treatment trial were sustained in the long term, and a previous trend to mortality benefit seen during the prevention trial became significant several years after termination of the study. The new results underline the need, as much as possible, to follow the patients seen in treatment trials over the long term, Yusuf said.

"The SOLVD data indicate that the result at the end of the trial was an underestimate of the true benefits of lifelong ACE inhibitor therapy," he concluded. Long-term follow-up of most trials should be considered for at least several years, "to give us the best estimate on long-term life expectancy."

Dr Salim Yusuf

XSOLVD = SOLVD prevention + SOLVD treatment

The SOLVD studies compared enalapril with placebo in 2 cohorts: 2569 patients with heart failure were enrolled in the SOLVD treatment trial, and 4228 patients with asymptomatic LV dysfunction were enrolled in the SOLVD prevention trial. At the conclusion of both trials, after an average follow-up of 3.2 years, heart failure hospitalization and MI were reduced with enalapril, but mortality was reduced significantly only in the treatment trial, with a nonsignificant trend in the prevention trial, Jong said.

The aim of XSOLVD was to complete a 12-year follow-up of the patients in the SOLVD trials to establish whether the mortality reduction in the treatment trial was sustained and whether the reductions in morbidity seen with enalapril in the prevention arm would translate into a significant improvement in survival.

The investigators were able to establish vital status in 99.8% of patients through various means, including telephone contact in the Belgian cohort and through national databases for US and Canadian patients. Mean duration of follow-up was 11.2 years in the prevention trial and 12.1 years in the treatment trial, Jong noted.

In the prevention trial, cumulative survival curves continued to diverge over time in favor of the enalapril group, so that the absolute difference in survival increased to 6%, which was highly significant (p=0.001).

Cumulative 12-year survival in the SOLVD prevention trial

Time point Enalapril (%) Placebo (%)
Termination of trial 86 84
5 years 77 73
12 years 47 41

In the treatment trial, survival benefit of enalapril persisted for 4.9 years after the trial had terminated before the 2 curves started to converge, Jong said. At 12 years, the survival rates were similar. "Nevertheless, the difference in the overall survival experience between enalapril and placebo was significant, with a p value of 0.01," he said.

Cumulative 12-year survival in the SOLVD treatment trial

Time point Enalapril (%) Placebo (%)
Termination of trial 64 60
5 years 53 49
12 years 20 20

Because there was no statistical evidence of a difference in treatment effect between the trials, Jong said, they were able to combine patients and derive a single hazard ratio. The overall hazard ratio was 0.90 (0.84-0.95) for the combined trials, a 10% relative risk reduction in favor of enalapril. The finding was highly significant (p=0.0003).

Increase in life expectancy

The long-term follow-up also allowed them to calculate the magnitude of the increase in life expectancy from treatment. "As you know, this is very, very rare in the trials we do," noted Yusuf, who did the second half of the presentation. Enalapril treatment extended median survival by 9.2 months in the prevention trial and by 8.6 months in the treatment trial.

All-cause mortality was reduced from 65.7% to 61.8%, for a hazard ratio of 0.90, which was highly significant (p=0.0003), Yusuf said. The difference was driven almost entirely by a decrease in cardiovascular mortality, mainly cardiac mortality. There was also some difference between groups in noncardiac vascular mortality but no impact of treatment on noncardiovascular mortality, identical at 12.4%.

Subgroup analysis showed consistent results, at least in direction, across all subgroups. Yusuf pointed particularly to race data. "There has been a previous publication, not by SOLVD investigators but using SOLVD data, claiming ACE inhibitors do not work in blacks," Yusuf said. "There is no reason not to expect a benefit of ACE inhibitors in nonwhite individuals."

Finally, they examined survival among patients in the prevention trial by whether or not they had experienced heart failure or MI during the trial. It appeared that survival was positively affected by treatment, even among patients who did not have a morbid event, he said. "What that means is that the mechanism of benefit long term is only partly related to the prevention of morbid events," Yusuf noted, and other factors such as LV, vascular remodeling, or other unknown factors may contribute to the long-term sustained benefit of treatment.


Discussant for the trial was Dr Karl Swedburg (Gothenburg, Sweden), who was a co-principal investigator on the CONSENSUS trial, the first trial to show a mortality benefit of ACE inhibitor treatment in patients with heart failure.

A 10-year follow-up of the CONSENSUS cohort, patients with severe heart failure, had shown treatment with enalapril was associated with improved survival compared with placebo even after the initial 6-month treatment duration, before the curves eventually came together.

The SOLVD trial was similar, he said, "but it was 27 times bigger in asymptomatic patients, and the treatment effect was not attenuated for many years, but there seems to have been an increase after the controlled study period."

The importance of these observations suggests the translation of treatment effect from short-term trials of 2 to 3 years to long-term effect is difficult and is still more difficult for life-long therapy, he said.

"I would recommend that in the design and planning of outcome trials an extended follow-up should be considered," Swedburg concluded.

Yusuf expanded on this idea at a press conference later. "What is now increasingly happening is that we're persuading our sponsors at the end of the study to say please give us a bit of extra funding so that an academic group can independently continue," he noted. Some countries have national mortality databases that allow passive follow-up, and they are planning this type of analysis in every trial they are associated with, he said.

A 3-year extension of the HOPE study is under way, for example, and results will be presented, possibly at this meeting next year, he said. They are planning a 3-year follow-up for the CURE study and, for the ongoing CHARM trial, have proposed an extended 3-year follow-up by mail and an "indefinite" follow-up by national mortality data.

"The most important message is that we are probably underestimating the full benefits of our treatments," Yusuf said.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.