Review of the GISSI-3 Trial

Lancet 1994;343:1115-22.Background By the time GISSI-3 was undertaken, evidence had emerged suggesting afterload reduction with angiotensin-converting-enzyme (ACE) inhibitors improved morbidity and mortality in patients with chronic systolic heart failure and these agents were already an established treatment for acute, severe heart failure based on theoretical grounds, despite lack of evidence from large scale RCTs. There was also lower level evidence that nitrates reduced mortality in patients with acute MI and these drugs were broadly used in coronary care units at the time. The GISSI-3 trial was undertaken to test the hypotheses that 6 weeks of treatment with lisinopril alone or transdermal glyceryl trinitrate (GTN) alone reduced the combined endpoint of mortality and severe LV dysfunction at 6 weeks compared to controls.Cardiology Trial’s Substack is a reader-supported publication. To receive new posts and support our work, consider becoming a free or paid subscriber.Patients Patients were eligible if they presented with chest pain and ST segment elevations of ≥1 mm in any limb lead of the electrocardiogram and/or ≥2 mm in any precordial leads and were admitted to the cardiac intensive care unit (CCU) within 24 hr from the onset of symptoms. Absolute contraindications included: severe heart failure requiring any of the study treatments, Killip class 4, high risk of hemodynamic deterioration after treatment with vasodilators (SBP Baseline characteristics The number of patients who were admitted to CCUs over the study period was 43,047 and 19,394 (45%) were randomized. Reasons for exclusion included: more than 24 hr from onset of symptoms (33%), contraindications to study treatment (23%), persistent hemodynamic instability (15%), and administrative reasons (28%). Similar to other GISSI trials, which also provided information on patients enrolled versus those who were not, women were more likely to be excluded (31% of excluded vs 22% of enrolled) as were patients >70 years of age (41% of excluded vs 27% of enrolled). As expected, excluded patients had higher in-hospital mortality (12% of excluded vs 6% of enrolled).Seventy-eight percent of patients enrolled were men and nearly three quarters were under the age of 70. Patients with anterior STEMI accounted for approximately 27%, those with inferior 32% and non-Q-wave MI’s approximately 19%. Time from symptom onset was between 12-24 hours in 40%, between 6-12 hours in 25% and less than 6 hours in 35%. The vast majority of patients had a Killip score of 1 (85%) with the remainder having a Killip score of 2 (14%). Procedures The study used a 2x2 factorial design, resulting in 4 treatment groups: lisinopril alone, transdermal GTN alone, combination therapy with lisinopril and transdermal GTN, or no trial therapy.Patients randomized to oral lisinopril received 5 mg at the time of randomization, followed by 5 mg 24 hours later and 10 mg 48 hours later to be continued at a dose of 10 mg daily for 6 weeks. Patients with SBP Those assigned to transdermal GTN were administered a continuous infusion over the first 24 hours. It was started at 5 mcg/min and increased by 5-20 mcg/min every 5 minutes for the first 30 minutes until SBP fell by at least 10%, provided it remained over 90 mmHg. After 24 hours, the IV infusion was replaced with a transdermal patch providing 10 mg daily. The patch was applied every morning and removed at bedtime, to provide a 10 hour nitrate-free interval, so as to keep tolerance to treatment to a minimum. If the patch was not tolerated, then 50 mg isosorbide mononitrate was given orally once daily. Those randomized to control received no treatment.Endpoints The combined primary endpoint was the same as GISSI-2 and consisted of all-cause mortality plus the number of patients who had late (day 4 or later) clinical congestive heart failure, or extensive LV damage (EF 70 years of age). A sample size of 20,000 patients was determined to detect a 20% reduction in the primary endpoint with incidence rates of mortality and severe LV dysfunction based on the GISSI-2 trial (10% and 23%, respectively). The primary analyses were lisinopril vs control (half of patients in each group received GTN) and transdermal GTN vs control (half of patients in each group received lisinopril).Results 19,394 patients were randomized from 200 CCU’s in Italy. Lisinopril significantly reduced the composite primary endpoint compared to control (OR 0.90; 17.0% vs 15.6%; 95% CI 0.84-0.98) as well as the individual component of all-cause death (OR 0.88; 6.3% vs 7.1%; 95% CI 0.79-0.99). Lisinopril significantly increased the rates of persistent hypotension (OR 2.44; 9.0% vs 3.0%; 95% CI 2.17-2.74) and renal dysfunction (OR 2.09; 2.4% vs 1.15; 95% CI 1.68-2.60).Lisinopril’s effects on the composite endpoint and death were preserved in the pre-specified higher risk subgroups. In patients >70 years of age, lisinopril significantly decreased the combined endpoint (OR 0.83; 24.8% vs 28.3%; 95% CI 0.74-0.94) and death (14.0% vs 15.6%). Odds ratios were not provided on individual components of the endpoints in pre-specified subgroup analyses. In women, lisinopril significantly decreased the combined endpoint (OR 0.86; 20.8% vs 23.4%; 95% CI 0.74-0.99) and death (10.7% vs 12.9%). Adverse event outcomes are not provided for these subgroups.Unlike lisinopril, transdermal GTN did not significantly reduce the combined endpoint (OR 0.94; 15.9% vs 16.7%; 95% CI 0.87-1.02) or death (OR 0.94; 6.5% vs 6.9%; 95% CI 0.84-1.05). There were no significant differences in any other clinical events. In pre-specified subgroup analyses, transdermal GTN did not significantly reduce the combined endpoint; however, absolute differences in death were very close to those seen with lisinopril.Additional results from this trial can be gleaned from analyses of patients based on whether they were in 1 of 4 treatment groups based on the factorial design 1) lisinopril + GTN control, 2) GTN + lisinopril control, 3) lisinopril control + GTN control (i.e., double control), or 4) lisinopril + GTN (i.e., double treatment); however, caution was urged in interpreting these underpowered comparisons that included sample sizes half as large as the primary comparison groups.Compared to double controls, neither lisinopril alone or GTN alone significantly reduced the combined endpoint or death but the absolute differences in events mirrored the primary analyses; however, the combination of lisinopril and GTN significantly reduced both the combined endpoint (OR 0.85; 14.8% vs 17.0%; 95% CI 0.76-0.94) and death (0.83; 6.0% vs 7.2%; 95% CI 0.70-0.97). These treatment effects were slightly larger than those for lisinopril alone.Conclusions GISSI-3 is the first trial to show that immediate treatment with ACE inhibition can reduce death and severe LV dysfunction in patients with acute myocardial infarction and this benefit is maintained in women and elderly patients who are at higher risk for experiencing adverse events. The NNT for the combined endpoint and death at 6 weeks was 71 and 125, respectively. The NNTs to prevent these outcomes in women and elderly patients were considerably lower (e.g., relative risk reductions were preserved resulting in higher absolute treatment benefits). The same effect was not observed for transdermal GTN; however, there was some hint from the trial that GTN may have a synergistic effect when used in combination with lisinopril.It must be pointed out that, like the earlier beta blocker trials, patients in GISSI-3 were hemodynamically stable and severe heart failure was an absolute contraindication. Thus, this trial does not say anything about the efficacy of ACE inhibition or nitrate therapy for patients with acute myocardial infarction complicated by acute heart failure and/or hemodynamic instability. Yet both treatments were frequently used for severe heart failure based on pathophysiologic reasoning at the time (e.g., effects of reducing preload and afterload on congestion and cardiac performance).Thank you for reading Cardiology Trial’s Substack. This post is public so feel free to share it. Get full access to Cardiology Trial’s Substack at cardiologytrials.substack.com/subscribe