Recent studies have shown that the decrease in ventricular septal rupture (VSR) incidence after acute myocardial infarction is related to the improvement of reperfusion strategies. Our main objective was to explore the influence of therapeutic management changes on post-infarct VSR patient outcomes in a single reference center over a period of 30 years. We analyzed therapeutic management strategies and mortality rates in 228 patients with VSR after acute myocardial infarction admitted from 1981 to 2010. Patients were classified in 3 successive decades. There were no significant differences in clinical characteristics of patients with VSR at admission among those decades. Overall, surgery was performed in 159 patients (71.9%), primary transcatheter VSR closure was attempted in 5 patients (2.2%), and 64 patients (27.6%) were managed medically. Independent predictors of in-hospital mortality were VSR surgical repair (odds ratio [OR] 0.22, 95% confidence interval [CI] 0.1 to 0.7, p = 0.008), cardiogenic shock (OR 6.06, 95% CI 2.8 to 13.1, p <0.0001), and Killip class on admission (OR 1.75, 95% CI 1.1 to 9.9, p = 0.02). We found a significant 1-year mortality reduction between the first and second decades (hazard ratio 0.48, 95% CI 0.28 to 0.80; p = 0.005), with no significant change in the last decade (p = 0.2). This change was related to a systematic referral to surgical repair and shorter delays to VSR surgery (5.2 ± 6.3 vs 1.9 ± 3.2 days from first to second decade; p = 0.012). In conclusion, surgical repair remains the only significant efficient therapy to reduce mortality in patients with VSR (p <10 −3 ). In-hospital prognosis remains disappointing. This contrasts with the favorable long-term outcome of patients who survive the perioperative period and are discharged from hospital.
Ventricular septal rupture (VSR) is a severe and rare complication of acute myocardial infarction (AMI). Reperfusion therapies have significantly reduced the incidence of VSR. However, mortality rates have remained unchanged. In patients without surgical treatment, this disease is associated with a 90% in-hospital mortality. Current guidelines recommend that patients with post-AMI VSR should be considered for urgent surgical repair, unless further support is futile. The timing of surgery and perioperative therapeutic management of patients with VSR remain controversial. Therapeutic management strategies of patients with post-AMI VSR and effects of changes in these strategies in the previous 3 decades remain poorly explored. This report assesses therapeutic management strategies as well as trends and predictors in mortality rates of patients with post-AMI VSR over 3 decades in a single tertiary referral center.
Methods
Data for this study were obtained from our institutional database, containing discharge records of all patients admitted with a diagnosis of AMI. This retrospective study was approved by our Institutional Review Board. The study cohort was constituted by patients admitted to our institution, a tertiary referral university hospital, with a diagnosis of acute ST-segment elevation myocardial infarction (STEMI) complicated with VSR from 1981 to 2010. Patients with VSR complicating STEMI during this period were identified by the diagnosis code for acquired VSR (International Classification of Disease code 429.71; n = 171) or a procedure code indicating VSR surgical repair within 30 days after initial AMI admission (International Classification of Disease codes 35.53, 35.62, or 35.72; n = 57).
Patients surviving the initial hospital period were contacted by telephone in 2010 to inquire about cardiovascular diagnoses and death. Next-of-kin interviews for out-of-hospital cardiovascular deaths were obtained. To verify self-reported diagnoses, copies of all death certificates and medical records were requested for all outpatient cardiovascular diagnoses. We were successful in getting medical records in 99% of patients. Follow-up telephonic interviews were completed in 95% of living patients.
Demographic, medical history, clinical and hemodynamic data at admission and at the time of surgery, echocardiographic data, and therapeutic management procedures were recorded as covariates. Cardiogenic shock was defined as a systolic blood pressure of <90 mm Hg for 30 minutes, not responsive to fluid administration alone, and associated with clinical signs of hypoperfusion (urine output of <30 ml/h and cold extremities).
During the whole study period, a senior cardiac surgeon, a senior anesthesiologist, and a senior cardiologist in the intensive cardiac care unit were on call. Based on our internal institution protocol, each VSR case was collegially discussed for global therapeutic management and the indication and timing of VSR surgical repair. Depending on the period, various protocols within our institution were applied. From 1981 to 1990, if the patient presented with a stable hemodynamic status, VSR surgical repair was not considered in emergency and was performed from 1 to 3 weeks after index AMI hospitalization. Patients were therefore initially managed with medical treatment with or without intra-aortic balloon pump (IABP) support and close monitoring within our intensive care unit. Urgent VSR surgical repair was performed only in case of significant alterations of the hemodynamic clinical status. A systematic change in therapeutic management strategy occurred after 1991 when an institutional consensus decision was made to attempt VSR surgical repair for all patients as early as possible after diagnosis or hospital admission. At the same time, systematic support with IABP for all patients with VSR was applied while waiting for surgery.
Various surgical repair techniques were used, depending on the location and pathologic characteristics of VSR and at the surgeon’s discretion. All open-chest surgeries were performed with cardiopulmonary bypass, crystalloid cardioplegia, and moderate systemic hypothermia. Concomitant coronary artery bypass was performed at the surgeon’s discretion.
From 2001 to 2010, with the advent of systematic primary percutaneous coronary intervention for patients with STEMI referred to our institution, the rates of efficient reperfusion significantly improved. Therapeutic management strategy of patients with VSR remained the same as the previous decade. However, from 2006 to 2010, a percutaneous interventional strategy was attempted. This strategy was based on immediate primary transcatheter closure of VSR in a selected group of patients presenting with acute anteroseptal myocardial infarction. Transcatheter closure was attempted according to VSR location and its anatomical characteristics. Also, in this period, extracorporeal membrane oxygenation was proposed and applied in case of severe right ventricular dysfunction after VSR surgical repair.
We compared demographic, clinical and cardiac procedures data over the 3 successive decades of our study period. Chi-square tests were performed for categorical variables and Student t test or Wilcoxon rank sum test for continuous variables as appropriate to compare clinical and demographic characteristics.
Overall in-hospital and 1-year mortalities were the 2 studied outcomes. Overall in-hospital mortality was defined as mortality occurring at the hospital or within the month after admission, and 1-year mortality included all deaths from any cause that occurred within the first year of follow-up after admission.
For both outcomes, Kaplan-Meier survival curves were estimated per decade. The 3 survival curves were compared using log-rank test. We used univariate logistic regression models to analyze potential determinants of in-hospital mortality with systematic adjustment for time decade and surgical intervention. Variables that were tested as in-hospital mortality determinants were time decade, surgical intervention, age, gender, diabetes, Killip class, infarct location, coronary artery bypass graft surgery, IABP implantation, right ventricle dilation, coronary angiography, time delay between infarction and VSR diagnosis, cardiogenic shock, and urine output in the first 24 hours after VSR diagnosis. All these variables were then included in a full multivariate model. To study overall 1-year survival, nonadjusted and adjusted Cox proportional hazard models were built including relevant demographic and clinical variables aforementioned. A 2-tailed p value <0.05 defined statistical significance. Analyses were performed using R, version 2.15-2 (The R Foundation for Statistical Computing, Vienna, Austria).
Results
Two hundred twenty-eight consecutive patients with post-AMI VSR were hospitalized at our institution from 1981 to 2010. Sixty-nine patients (30%) were transferred from other centers. As listed in Table 1 , 92 patients were hospitalized in 1981 to 1990, 83 in 1991 to 2000, and 53 in 2001 to 2010. Mean age was 71 ± 8 years, with 51% women. Infarct location was anterior in 55%, and mean time interval between onset of infarction and VSR diagnosis was 4.5 ± 7 days. Cardiogenic shock was present at presentation in 24 patients (13.3%). Cardiac catheterization was performed in 111 patients (49%). Cardiac procedures by study decade are listed in Table 2 . Overall, 159 patients (71.9%) were treated surgically. The proportion of patients who received VSR repair significantly increased from 63% in 1981 to 1990 to 83% in 1991 to 2000 (p = 0.003). It decreased to 70% in 2001 to 2010.
| Variable | Time Period | All Patients (n = 228) | ||
|---|---|---|---|---|
| 1981–1990 (n = 92) | 1991–2000 (n = 83) | 2001–2010 (n = 53) | ||
| Age (yrs) | 70 ± 8 | 72 ± 9 | 78 ± 8 | 71 ± 8 |
| Women | 40 (43.5) | 41 (49.4) | 35 (66.0) | 116 (50.9) |
| Diabetes mellitus | 15 (23.4) | 19 (22.9) | 8 (15.7) | 42 (21.2) |
| Smoker | 28 (30.4) | 18 (21.7) | 17 (32.1) | 63 (27.6) |
| Hypertension | 40 (43.5) | 33 (39.8) | 25 (47.2) | 98 (43.0) |
| Preinfarct angina pectoris | 18 (19.6) | 16 (19.3) | 3 (5.7) | 37 (16.2) |
| Location of infarct | ||||
| Anterior wall | 45 (48.9) | 43 (52.4) | 36 (67.9) | 124 (54.6) |
| Inferior/lateral wall | 47 (51.1) | 39 (47.6) | 17 (32.1) | 104 (45.4) |
| Right ventricular dysfunction | 39 (50.0) | 24 (35.8) | 20 (38.5) | 83 (42.1) |
| Killip class at admission | ||||
| I | 23 (41.8) | 41 (55.4) | 31 (60.8) | 95 (52.8) |
| II | 8 (14.5) | 23 (27.7) | 11 (31.1) | 42 (23.3) |
| III | 13 (23.6) | 3 (4) | 3 (5.9) | 19 (10.6) |
| IV | 11 (20.0) | 7 (9.9) | 6 (11.8) | 24 (13.3) |
| Cardiogenic shock during hospitalization | 60 (65.2) | 42 (50.6) | 33 (62.3) | 135 (59.2) |
| Time interval between MI and VSR (days) | 3.8 ± 4.4 | 5.9 ± 10.1 | 3.7 ± 5.0 | 4.5 ± 7.1 |
| GRACE score | NA | 232.8 ± 34.1 | 227.4 ± 36.5 | 228.7 ± 35.7 |
| Cardiac Procedure | Time Period | All Patients | ||
|---|---|---|---|---|
| 1981–1990 (n = 92) | 1991–2000 (n = 83) | 2001–2010 (n = 53) | ||
| Diagnostic catheterization | 18 (19.6) | 56 (68.3) | 37 (69.8) | 111 (48.9) |
| Coronary bypass | 1 (1.8) | 14 (19.7) | 9 (20.5) | 24 (14.0) |
| IABP | 11 (12.0) | 50 (60.2) | 41 (77.4) | 102 (44.7) |
| Inotropic agents | 75 (81.5) | 53 (63.9) | 33 (62.3) | 161 (70.6) |
| VSR repair surgery | 58 (63.0) | 69 (83.1) | 37 (69.8) | 164 (71.9) |
| VSR diagnostic to VSR surgery time delay (days) | 5.2 ± 6.3 | 1.9 ± 3.2 | 2.2 ± 4.0 | 3.2 ± 4.9 |
| Percutaneous transcatheter VSR closure | — | — | 5 (9.4) | 5 (2.1) |
Surgical approach to repair VSR was through the left atrium in 156 patients and right atrium in 3 patients. In addition to VSR repair, 24 patients (14%) received combined coronary artery bypass graft surgery. One patient was treated with a heart transplant.
Mean time from VSR diagnosis to VSR repair was 3.2 ± 4.9 days. Mean delays between AMI and surgery significantly decreased from 9.2 ± 8.3 days in the first decade to 6.2 ± 7.6 in the last decade (p = 0.001). This was because of a sharp decrease in the VSR surgery delay from 5.2 ± 6.3 days in the first decade to 2.2 ± 4.0 days in the second decade (p <0.0001).
From 2001 to 2010, 5 patients (9.4%) were selected for percutaneous transcatheter VSR closure. Device implantation was successful in 3 patients. Left-to-right ventricle shunt persisted in all cases. All 3 patients underwent subsequent surgical VSR repair and died during the postoperative period. In the other 2 patients, device implantation failed.
Overall, 64 patients (27.9%) were treated medically. All but 5 died within 2 months after hospitalization. In the first decade, 13 patients (14%) died while waiting for surgery and 21 (22%) were denied surgical treatment. In the second decade, 2 patients (2.4%) died while waiting for surgery and 11 (13%) were denied surgical treatment. During the last decade, no patient died while waiting for surgery and 13 patients (24%) were denied surgical treatment.
In-hospital survival rate increased significantly from 23% in the first decade to 40% in the second and 38% in the last decade. The test for trend was significant over the 3 periods (p = 0.03). Patients with cardiogenic shock had significantly greater in-hospital mortality than patients without (84% vs 43%, respectively, p <0.05). There was a nonsignificant trend toward a better in-hospital survival after surgery from 34% in 1981 to 1990 to 45% in 1991 to 2000 and 49% in 2001 to 2010 (p = 0.31). Determinants of in-hospital mortality analyses are listed in Table 3 . The only independent predictors of in-hospital mortality were VSR surgical repair, presence or evolution toward cardiogenic shock, and Killip class on admission ( Table 3 ).
| Risk Factors in Logistic Models | Unadjusted Regression | Adjusted Regression | ||||
|---|---|---|---|---|---|---|
| OR | 95% Confidence Interval | p | OR | 95% Confidence Interval | p | |
| Admission decade | ||||||
| 1981–1990 | Reference | Reference | ||||
| 1991–2000 | 0.56 | 0.28–1.12 | 0.10 | 0.53 | 0.25–1.16 | 0.114 |
| 2001–2010 | 0.50 | 0.22–1.10 | 0.09 | 0.37 | 0.15–0.89 | 0.027 |
| Surgical repair | 0.12 | 0.05–0.33 | <0.0001 | 0.22 | 0.08–0.62 | 0.004 |
| Age (yrs) | 1.04 | 1.00–1.08 | 0.04 | 1.06 | 1.02–1.10 | 0.008 |
| Time interval between MI and VSR (days) | 0.96 | 0.91–1.00 | 0.07 | |||
| Gender (female) | 1.12 | 0.61–2.06 | 0.714 | |||
| Cardiogenic shock | 5.95 | 3.08–11.47 | <0.0001 | 6.95 | 3.48–13.88 | <0.0001 |
| Oliguria at admission | 8.94 | 1.71–46.86 | 0.01 | |||
| Diabetes | 2.99 | 1.17–7.65 | 0.02 | |||
| Infarct location | ||||||
| Anterior wall | Reference | |||||
| Inferior/lateral wall | 1.46 | 0.79–2.70 | 0.23 | |||
| Right ventricle dysfunction | 0.99 | 0.52–1.92 | 0.99 | |||
| Diagnostic catheterization | 1.16 | 0.58–2.33 | 0.68 | |||
| Killip class at admission | ||||||
| I | Reference | |||||
| II | 1.09 | 0.52–2.28 | 0.82 | |||
| III | 3.95 | 1.08–14.48 | 0.04 | |||
| IV | — | — | — | |||
| CABG | 0.99 | 0.39–2.49 | 0.98 | |||
| IABP | 2.26 | 1.05–4.84 | 0.04 | |||
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
