|
|
||||||||
Ann Thorac Surg 2002;74:384-389
© 2002 The Society of Thoracic Surgeons
a Department of Cardiothoracic and Vascular Surgery, Deutsches Herzzentrum Berlin, Germany
b Department of Hematology and Oncology, Humboldt University of Berlin, Charité, Campus Virchow Klinikum, Berlin, Germany
Accepted for publication April 10, 2002.
* Address reprint requests to Dr Potapov, Deutsches Herzzentrum Berlin, Augustenburger Platz 1, 13353 Berlin, Germany
e-mail: potapov{at}dhzb.de
| Abstract |
|---|
|
|
|---|
Methods. From 1992 to 2000, a total of 28 patients with CLL underwent heart surgery using CPB at our institution (group I). These patients were compared with 25 patients from the CLL register who were retrospectively matched with regard to preoperative administration of chemotherapy, Binet classification, age, and sex (group II). A time-point was selected for each patient in group II so that the variables for the two groups corresponded in relation to the time of operation of the patients in group I. Midterm follow-up data in both groups were analyzed.
Results. There were no differences between groups regarding matched variables. The mean follow-up time was similar in both groups (2.6 ± 2.2 vs 2.3 ± 1.3 years, p > 0.5). The 30-day mortality in group I was 14.3%. The mean stay in the intensive care unit was 4.2 ± 7.5 days; the median number of units of packed red blood cells transfused was three (range 0 to 17). Compared with group II, in group I significantly fewer patients (11 vs 17, p = 0.049) required chemotherapy significantly later (1.98 ± 2.06 vs 0.84 ± 1.18 years, p = 0.018). During follow-up, no difference was found between groups regarding severe infections (10 vs 14, p = 0.14). Despite postoperative mortality in group I, the long-term mortality was similar in both groups (p = 0.3).
Conclusions. Cardiac surgery using CPB did not have a negative impact on the natural course of CLL. Moreover, this procedure seems to be associated with a decrease in the number of postoperative chemotherapy administrations and with an increase of chemotherapy-free survival time. Although CLL may be a risk factor in the early postoperative period, it is not a contraindication for cardiac surgery using CPB.
| Introduction |
|---|
|
|
|---|
Cardiopulmonary bypass (CPB) and surgical trauma have long been known to cause altered activation of the inflammatory response, in part related to blood exposure to the large artificial surfaces [5, 6]. This includes complement activation and consumption along with elevation in levels of interleukin, tumor necrosis factor-
, and procalcitonin [7]. This results in changes in counts and properties of granulocytes and lymphocytes, as well as their subsets [8]. Furthermore, CPB decreases serum gamma globulin levels and also platelets and red blood cell counts. All of these changes lead to significant immunodepression. Despite significant progress in technique [5, 6], CPB remains a significant factor contributing to morbidity after cardiac surgery.
The aging of the population and the increase in life expectancy in developed countries [9], as well as progress in CLL treatment [3, 10, 11], have led to an increased number of older patients with CLL-related immune deficiency who may require operations using CPB. On the other hand, the improvements in CPB technique [12, 13] and the experience gained with older patients have led to increased acceptance of these patients for cardiac surgery [14]. Some authors have reported good postoperative outcome in CLL patients [1517]. However, the postoperative course of CLL in these patients is unknown.
The aim of the present study was to evaluate the impact of cardiac surgery using CPB on the mid-term course of CLL.
| Material and methods |
|---|
|
|
|---|
As a control group (group II), 25 patients from the CLL Register of Humboldt University, Virchow Klinikum, were retrospectively matched to patients from group I. The matching critera (in order of priority) are as follows: administration of chemotherapy preoperatively (group I) and before the matching time-point (group II), Binet classification, age, and sex, based on information from the database and medical records. The modus of chemotherapy (short- or long-term) and type (single drug or multiple drug administration) were taken into consideration. The order of priority was chosen because the need for chemotherapy and the Binet classification are the two variables that correlate with quality of life and prognosis for CLL patients, respectively. The time-point for the beginning of follow-up investigation for each patient in group II was selected so that the variables for paired patients corresponded, taking the time of operation of the patient in group I as the beginning of the follow-up time.
The medical records of patients in both groups were retrospectively analyzed. Follow-up was completed by means of a questionnaire or by telephone contact with general practitioners, hematologists, and, if necessary, the patients themselves.
The indications for chemotherapy were similar in both groups and remained constant throughout the study period; these included symptoms of bone marrow failure, progressive lymphadenopathy or hepatosplenomegaly, progressive lymphocytosis, and development of constitutional symptoms [1, 10, 11].
Infections were classified and defined according to severity. They were included in the analysis if the patient required hospital admission and antibiotic therapy or died due to infection.
To assess the impact of cardiac surgery using CPB on blood cells, the hemoglobin level and platelets and white blood cell (WBC) counts were recorded both preoperatively and on the third postoperative day.
The number of units of packed red blood cells, units of fresh frozen plasma, and platelet concentrates transfused during the hospital stay of patients in group I was recorded. For comparison, the use of blood products during hospital stay in consecutive patients who were more 18 years of age and were undergoing operation with CPB in our institution during the arbitrarily selected period from January 1 through December 31, 1999 (n = 3,078) was calculated using the hospital database. A hemoglobin level of 8 to 9 g/dL, depending on the hemodynamic and clinical situation, was the transfusion trigger point both in these patients and in the CLL patients.
Statistical analysis
Statistical analysis of the data was performed with SPSS software, version 9.0.0 for Windows (SPSS Inc, Chicago, IL). Normally distributed data are presented as means and standard deviations, whereas the non-normally distributed date are given as median, 25th percentile, and 75th percentile. The unpaired t test or Wilcoxon test was used to test for group differences in the case of quantitative data, as appropriate. For qualitative data, frequencies were calculated and the
2 test was used to test for group differences. Freedom from chemotherapy initiated after surgery (group I) or after the matching time-point (group II) and cumulative survival were examined by Kaplan-Meier analysis and compared using a log-rank test. A reference value of p of less than 0.05 was considered statistically significant.
| Results |
|---|
|
|
|---|
|
The blood cell counts and values for hemoglobin in group I measured preoperatively and early postoperatively (ie, on the third postoperative day) are presented in Table 2. There was a significant decrease in hemoglobin level and platelet number on the third postoperative day. The WBC increased postoperatively, but the difference did not reach a statistically significant level.
|
During follow-up, significantly fewer patients in group I required chemotherapy compared with those in group II. The mean period between operation and initiation of chemotherapy in group I or between the selected time-point and the initiation of chemotherapy in group II was also significantly greater in group I (Table 1). Kaplan-Meier analysis results are shown in Figure 1. Despite the inclusion in the analysis of the postoperative mortality rate of 14.3% in group I, the long-term survival was similar in both groups. The Kaplan-Meier analysis is presented in Figure 2.
|
|
|
| Comment |
|---|
|
|
|---|
Although the patients in group I had cardiac disease as an additional risk factor, their midterm survival was similar to that of patients in group II as well as that reported from the US National Cancer Data Base for CLL patients between 60 and 80 years of age [19]. This fact could be explained by improvement in the natural course of CLL in patients who underwent cardiac surgery using CPB (Fig 1). Diehl and colleagues [19] showed that, in CLL patients, CLL is still a major cause of death with increasing age. Thus, the improved course of CLL may lead to improved survival in CLL patients with additional heart disease who require cardiac surgery.
Despite well-balanced matching, it is possible that preselection of patients in group I as being fit enough to withstand cardiac surgery could be considered to affect our study results; but, in fact, the patients in group I were in general sicker due to the existence of cardiac disease.
Some effects of CPB and cardiac surgery on immunologic response have been described [58]. In summary, the proinflammatory role of CPB, surgical trauma and cardioplegic arrest has been demonstrated by an increase of proinflammatory complement factors and cytokine levels in serum [5, 6, 20]. Both CPB and allogenic blood transfusion lead to an increase in the antiinflammatory interleukin-10 serum level, which suppresses T-cell activity, especially that of the cytotoxic T-cells and Th1 cells [8, 21].
Increased apoptosis of CLL cells related to CPB may also have an effect on the course of CLL. The microenvironment plays a prominent role in the apoptotic behavior of CLL cells, as the malignant cells accumulate in vivo with a life span of up to 4 years, whereas they rapidly undergo apoptosis when cultured in vitro [22].
Recently CD4+ T-cells have been identified as interacting with CLL cells by means of the CD40 ligand. Through the stimulation of CD40, survivin is expressed. The stimulated B-CLL cells are also Bcl-2+, thus showing a tendency to survive [23]. As CD4+ lymphocytes are significantly reduced after CPB [24], the reduced costimulatory effects of CD4+ T-lymphocytes may lead to higher sensitivity towards proapoptotic signals.
There is some clinical and biochemical evidence that the impact of CPB on immune response is greater than that of major surgical trauma [5, 25, 26]. However, the interactions between the variables of immune response during and after cardiac surgery remain not clearly understood, and there are still more questions than answers [27].
Another factor that may contribute to the immunosupressive effect of cardiac surgery using CPB is the postoperative transfusions of stored allogenic blood components that were performed in the majority of patients in group I. Alloantigens that are expressed in various blood products can elicit a suppression of the immune response [28]. The immunologic mechanism leading to downregulation of the alloimmune response is not clear. One possible explanation is the induction of a Th2 response by nonprofessional antigen presentation by the transfused blood cells with subsequent increase of the serum interleukin-10 level [21]. On the other hand, evidence is accumulating that the degree of HLA compatibility between donor and recipient is a determining factor [28].
The fact that chemotherapy was continued perioperatively in 2 patients in group I suggests that the postoperative status did not lead to a delay in the initiation of chemotherapy. However, further investigations are necessary to establish whether the significantly longer chemotherapy-free period in group I was related to alterations of the inflammatory response system due to CPB, major surgical trauma, or allogenic blood transfusion.
Chronic lymphatic leukemia remains a risk factor in the early postoperative period after cardiac surgery. The operative mortality in CLL patients is higher [1517] than in a comparable population as reported in the Registry of The Society of Thoracic Surgeons. In our patients, the early postoperative mortality was strongly related to the need for emergency surgery (all patients who required emergency surgery died), whereas 3 of the 4 patients with advanced disease (Binet stage C) survived over the early postoperative period.
Some risk factors in patients with CLL may contribute to increased postoperative mortality. Thrombocytopenia and abnormal platelet function are risk factors for postoperative bleeding. The normal preoperative median platelet count and thrombocytopenia in only 1 patient in group I may explain the similar frequency of the postoperative use of blood products (with the exception of a slight increase in use of units of packed red blood cells) in CLL patients compared with that calculated from our hospital database for 1999 (Table 3). The small number of patients with advanced disease who underwent cardiac surgery, and the possible preoperative selection of patients without thrombocytopenia for elective procedures, may also contribute to this effect. However, the significant decrease in median postoperative platelet count in our study may enhance the risk for bleeding, especially in the thrombocytopenic patients. This risk may be reduced if the off-pump technique is used [13, 29].
The decreased immune competence of CLL patients is another risk factor contributing to increased incidence of postoperative infections, which remains the major postoperative complication [15, 16]. The immune suppression related to CPB, surgical trauma, and blood transfusion is also a risk factor and may contribute to this phenomenon. Two patients in our series died from sepsis early in the postoperative period. However, both of these patients required emergency surgery, which is known to be associated with increased postoperative mortality. Based on our experience, we use the first generation of cephalosporins for routine perioperative prophylaxis also in CLL patients, but change to therapy with broad spectrum antibiotics early if signs of systemic infection occur, and then to antibiotics specific to the antibiogram. Although all patients in group I underwent major surgery (which is per se a risk factor for infection [5]), whereas only 1 patient in group II had major surgery (in this case, bowel resection) during follow-up, no differences with regard to severe infection during the follow-up period were found between the groups, even when the early postoperative period was included in the analysis (Table 1).
Limitations of the study
The descriptive nature of the study and the relatively small number of patients preclude a definitive answer about the impact on CLL of cardiac surgery using CPB. Another limitation is that the study could not differentiate between the impact of CPB itself and the transfusion of blood components on the natural course of CLL. There was a slightly different distribution of clinical stages in both groups. Nevertheless, stage distribution was similar to that reported by Karmiris and colleagues [4]. Although our study reports the largest population of CLL patients who underwent cardiac surgery using CPB, the relatively small number of patients precluded better matching or differentiation into subgroups. In larger studies tests for paired variables could be used to improve the statistical analysis. The chemotherapy was initiated in both groups based on published recommendations [1, 10, 11]. However, a long period of data collection precluded more rigorous definition. Cellular morphology, patterns of bone marrow involvement, lymphocyte doubling time, drug resistance, cytogenic variables, Ig VH mutational status, and serum markers were not considered as matching criteria because these data were not available for most patients, especially in the early period.
In conclusion, cardiac surgery using CPB probably did not have a negative impact on the natural course of CLL in this study. Moreover, this procedure seems to be associated with a decrease in the number of postoperative chemotherapy administrations and with an increase of chemotherapy-free survival time. Although CLL may be a risk factor in the early postoperative period, especially with regard to infection, it is not a contraindication for cardiac surgery using CPB.
| Acknowledgments |
|---|
|
|
|---|
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
A. M. Fecher, T. J. Birdas, D. Haybron, P. K. Papasavas, D. Evers, and P. F. Caushaj Cardiac operations in patients with hematologic malignancies Eur. J. Cardiothorac. Surg., April 1, 2004; 25(4): 537 - 540. [Abstract] [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| 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 |