| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
a Department of Cardiac Surgery, IRCCS Policlinico San Donato, Milan, Italy
b Department of Cardiothoracic and Vascular Anesthesia & ICU, IRCCS Policlinico San Donato, Milan, Italy
c Department of Critical Care Medicine, University of Florence, Florence, Italy
Accepted for publication July 2, 2009.
* Address correspondence to Dr Castelvecchio, Department of Cardiac Surgery, IRCCS, Policlinico San Donato, Via Morandi 30, San Donato Milanese, Milan, 20097, Italy (Email: castelvecchio.serenella{at}gmail.com).
 |
Abstract |
|---|
 Top Abstract IntroductionMaterial and MethodsResultsCommentReferences |
|---|
Methods: A total of 329 patients (234 nondiabetic and 95 diabetic) who survived the surgical ventricular restoration operation were admitted to this study. Cardiac mortality follow-up data were collected. Actuarial survival curves were calculated for the two groups; differences between groups and the impact of other comorbidities were established using a log-rank test and a Cox regression analysis.
Results: The mean follow-up time was 44 months. Diabetic patients had a significantly worse survival rate: at 5 years, their survival rate was 81%, versus 89% for nondiabetic patients (p = 0.019). Other comorbidities significantly associated with the survival rate were chronic renal failure, New York Heart Association class, and liver dysfunction. Diabetic patients without comorbidities had a survival rate similar to that of nondiabetic patients. Diabetic patients with at least one comorbidity had a significantly worse outcome. Diabetic patients with chronic renal failure had a 5-year survival rate of 40%, versus 85% for nondiabetic patients (p = 0.002).
Conclusions: Noncomplicated diabetes has no negative impact on long-term survival after surgical ventricular restoration. Conversely, complicated diabetes, namely the presence of chronic renal failure, carries a long-term cardiac mortality risk that is four times higher than the risk for nondiabetic patients.
|
Introduction |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsCommentReferences |
|---|
Data from previous, observational, nonrandomized studies report an acceptable 30-day mortality after SVR and good long-term survival [11–14]. Reported prognostic predictors for late morbidity and mortality are predominantly related to cardiac conditions [11, 13, 16], whereas few studies have assessed the prognostic impact of comorbidities [12]. Diabetes is a complex disease determining macrovascular and microvascular alterations. As a consequence, diabetic patients suffer from an increased morbidity due to either coronary artery disease, cardiovascular dysfunction, and congestive HF or hypertension or renal failure, which are secondary to microvascular disease. The impact of diabetes and related comorbidities, including chronic renal failure (CRF), low ejection fraction (EF), cerebrovascular accident, or peripheral vascular disease, has been investigated to a limited extent and restricted to patients surviving coronary artery bypass graft surgery [17, 18]. However, it is well known that diabetes has a greater impact on the incidence and prognosis of congestive HF than on that of coronary artery disease. We hypothesized that HF patients with diabetes and diabetes-related complications are at greater risk of cardiac death after SVR surgery.
Thus, the objective of the present study was to identify the impact of diabetes and related comorbidities on long-term survival of diabetic HF patients surviving SVR.
|
Material and Methods |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsCommentReferences |
|---|
All patients received an SVR operation; coronary artery bypass graft surgery was associated in 310 patients (94.2% of the overall population). The mitral valve was repaired when indicated. Indications for surgery were heart failure or angina or a combination of both.
Patients with diabetes were identified as those receiving oral antidiabetic treatment or insulin, or both, at the time of surgery. Patients receiving nutritional modifications as the sole treatment for hyperglycemia were not considered as diabetic patients. Insulin treatment of transient postoperative hyperglycemia was not considered a criterion for being included in the diabetic group. During and after the operation, blood glucose levels were controlled by insulin infusion to maintain a level below 180 mg/dL. At the time of surgery, all patients were on state-of-the-art optimized medical therapy.
Preoperative laboratory assays included serum creatinine value (mg · dL–1); serum bilirubin value (mg · dL–1), and hematocrit (%).
Cardiac function was assessed by the left ventricular EF, measured before the operation with an echocardiographic assessment. In case of repeated, different measurements, the lowest value was used; an EF value less than 0.30 was defined as a low EF. Cardiac-related conditions assessed were New York Heart Association (NYHA) functional class; hypertension (under medical treatment at the time of surgery); dyslipidemia (under medical treatment at the time of surgery); unstable angina (intravenous use of nitrates at the time of the operation); and previous myocardial infarction. The following comorbid conditions were recorded: obesity (body mass index greater than 30 kg · m–2); chronic obstructive pulmonary disease (on medication at the time of surgery); previous cerebrovascular accident; CRF (serum creatinine value greater than 150 µmol · dL–1); chronic liver dysfunction (serum bilirubin value more than 2 mg · dL–1); anemia (hematocrit less than 36%); previous vascular surgery; and previous cardiac surgery.
Operative data recorded were associated mitral valve procedure; use of internal thoracic artery; number of distal anastomoses; cardiopulmonary bypass duration (minutes); aortic cross-clamp duration (minutes); and lowest temperature on cardiopulmonary bypass (°C).
Surgical Technique
Details of the surgical technique have been previously reported [13–15]. Briefly, the procedure was conducted on an arrested heart with antegrade blood cold cardioplegia. Complete coronary revascularization was performed first, almost always with the left internal thoracic artery on the left anterior descending artery and the sequential venous grafts on right and circumflex arteries when needed. When needed, the mitral valve was repaired through the ventricular opening with a posterior annuloplasty. Surgical vascular restoration was performed using a mannequin (TRISVR TM; Chase Medical, Richardson, TX) filled at 50 to 60 mL/m2 to optimize the size and shape of the new ventricle.
Operative Mortality and Follow-Up
Operative mortality was defined as death occurring during hospitalization or within 30 days from the operation. Long-term survival was assessed on the basis of the occurrence of all-cause or cardiac death during the follow-up period. Cardiac death was defined as any cardiac-related, sudden, or unknown cause of death. The follow-up was conducted during January 2009 by direct telephone contact with the patients or their relatives; date and cause of death were checked using data from the National Statistical Institute.
Statistical Analysis
Data are expressed as mean and standard deviation for continuous normally distributed variables, median and range for nonnormally distributed variables, and number and percentage for binary categorical variables. Differences between groups were tested using Student's t test, Mann-Whitney U test, and Fisher's exact test when appropriate. All tests were two-sided.
Dichotomization of continuous variables (EF, serum creatinine, serum bilirubin) was obtained using a locally weighted scatterplot smoothing technique (a method based on iterative weighted least squares applied to nonlinear regressions) or a receiver operating characteristics analysis, where adequate cut-off points were identified at the point where the sum of sensitivity and specificity was the highest, according to the Youden's index: (sensitivity + specificity) – 1.
Operative mortality was explored using a multivariable logistic regression analysis where all the factors being univariately associated with the outcome (p level < 0.1) were entered, producing a final model including only the independent predictors, with an odds ratio and 95% confidence interval. Late mortality was explored separately for all-cause or cardiac mortality.
Patients were censored at the time of last complete information collection. Kaplan-Meier analysis was performed to assess freedom from death. Differences between groups were tested using a log-rank test. Diabetes and the other comorbid conditions were explored separately for association with freedom from cardiac death using a Mantel-Cox analysis, producing hazard ratios (HR) and 95% confidence intervals.
All the factors that were associated (p < 0.1) with cardiac death during the follow-up period were entered into a multivariable Cox proportional hazard analysis (stepwise forward), and a model including the independent predictors of cardiac death was produced, with HR and 95% confidence interval.
Proportionality assumption was checked by graphical representation using a plot of log-minus-log survival function. In the presence of a constant (vertical) difference between the curves, the hazards of subjects with different values for the independent variable were considered proportional over time.
For all statistical tests, a p value less than 0.05 was considered significant. Statistical calculations were performed using a computerized statistical program (SPSS 13.0; SPSS, Chicago, IL).
|
Results |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsCommentReferences |
|---|
|
|
|
|
|
|
Therefore, after adjustment for other potential confounders, diabetes carries a significantly increased risk for long-term cardiac mortality, and the presence of one comorbid condition may result in a four times increase of long-term cardiovascular mortality risk.
|
Comment |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsCommentReferences |
|---|
The impact of diabetes on long-term outcome after cardiac surgery is still controversial. Some studies have shown no deleterious effect of diabetes on intermediate and long-term mortality [18, 19]; others reported a significant negative effect of insulin requirement on long-term survival [20, 21]; and finally, others identified specific diabetes-related complications such as CRF, low EF, and peripheral vascular disease as independent risk factors for late mortality [17, 18]. Recently, Sartipy and associates [12] reported that diabetes, along with increasing age and mitral regurgitation grade III or IV, were associated with an increased risk for late mortality among HF patients after SVR surgery. Similarly, our diabetic HF patients treated with SVR had a significantly worse long-term outcome compared with nondiabetic patients in terms of cardiac mortality. It should be noted that the late mortality was acceptable for both diabetic and nondiabetic patients, considering the high-risk profile of this patient population.
The increased late cardiac mortality among diabetic HF patients may be related to a more severe preoperative profile, characterized by a lower hematocrit count and worse renal function along with more severe and diffuse coronary artery disease (as demonstrated by the more extensive coronary artery revascularization performed). A number of factors independently associated with late cardiac death included CRF, chronic liver dysfunction, and previous cerebrovascular accident, and these comorbidities are frequently related to diabetes. In the presence of one or more of these factors, the long-term outcome becomes seriously worse; in particular, CRF is associated with a mortality risk at 5 years that is more than twice the value in patients without CRF. This increased risk is a highly clinically relevant finding, supporting the important concept of "complicated diabetes," as introduced by Senni and associates [22]. These authors proposed a new risk stratification model to predict the 1-year outcome in HF patients. Their cardiovascular medicine heart failure index included cardiac conditions and comorbidities. Within the variables assessed in the derivation cohort at the univariate analysis, uncomplicated diabetes mellitus was discarded, whereas complicated diabetes mellitus (defined as diabetes with retinopathy, neuropathy, or nephropathy) entered into a logistic regression model and was included in the cardiovascular medicine heart failure index, with an odds ratio for 1-year mortality of 2.41. Our findings for 5-year mortality confirm this value. The relationship between renal function, diabetes, and mortality is explained by the fact that the kidney is the most specific target organ in diabetic patients. When the damage is advanced, renal failure patients are immune-compromised, with an increased risk for postoperative infection [23], are chronically anemic because of reduced erythropoietin production by kidneys [24], and are more predisposed to thromboembolic events and stroke [25]. Along with the progression of the organ disease, the lack of the metabolic effect of the insulin, either for decreased release or for insulin resistance, leads to an increased concentration of lipids in the circulation and subsequent chronic liver dysfunction. In our study, chronic liver dysfunction was independently associated with late cardiac death (HR 5.7).
In a previous study [14], we could identify a low EF as a determinant of operative mortality in patients undergone a SVR operation. Conversely, this factor was not associated with a late cardiac mortality in the present study. This finding suggests that the well-known anatomical changes in left ventricle shape and the increased EF that is generally a result of SVR operation may deprive the preoperative EF of a predictive role for late cardiac mortality. More important seems to be the functional class: patients in NYHA class III or IV at the time of the operation demonstrated a worse long-term outcome, with an HR of 2.0 for cardiac mortality.
The present study does not compare the long-term survival after SVR operations with other possible strategies for treating diabetic HF patients. A recent large randomized controlled trial [26] did not show any long-term outcome difference between HF patients receiving coronary artery bypass graft surgery either alone or with SVR. However, data analyzing specific subgroups of patients within this trial have not yet been reported.
Study Limitations
This study has some limitations. First, diabetes was not systematically assessed using standardized diagnostic criteria. Mean duration of diabetes was not available, and glucose tolerance, insulin resistance, or glycosylated hemoglobin levels were not evaluated, limiting the accuracy of our analysis. However, all the patients were followed by the referring clinician before being scheduled for the SVR operation, and the diagnosis of diabetes was known before the hospital admittance. Finally, the relatively small number of diabetic patient did not allow us to consider the study population separately regarding insulin- and noninsulin-dependent diabetic patients.
In conclusion, our study suggests that diabetes without comorbidities should not be considered a negative prognostic factor within the surgical decision process when considering SVR operations; conversely, for diabetic HF patients with preoperative CRF, surgeons should be aware of the poor long-term outcome of this operation.
|
References |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsCommentReferences |
|---|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| ANN THORAC SURG | ASIAN CARDIOVASC THORAC ANN | EUR J CARDIOTHORAC SURG |
| J THORAC CARDIOVASC SURG | ICVTS | ALL CTSNet JOURNALS |
