(Chest. 2003;124:102-107.)
© 2003
American College of Chest Physicians
Correlation Between Perioperative Blood Transfusion and Prognosis of Patients Subjected to Surgery for Stage I Lung Cancer*
Mario Nosotti, MD;
Paolo Rebulla, MD;
Donatella Riccardi, MD;
Alessandro Baisi, MD;
Nadia Bellaviti, MD;
Lorenzo Rosso, MD and
Luigi Santambrogio, MD
* From the Thoracic Surgery Unit (Drs. Nosotti, Baisi, Bellaviti, Rosso, and Santambrogio), Centro Trasfusionale e di Immunologia dei Trapianti (Drs. Rebulla and Riccardi), I.R.C.C.S. Ospedale Maggiore Policlinico, Milan, Italy.
Correspondence to: Mario Nosotti, MD, Thoracic Surgery Unit, Ospedale Maggiore Policlinico, I.R.C.C.S. Via F. Sforza, 35, 20122 Milan, Italy; e-mail: marionosotti{at}libero.it
 |
Abstract
|
|---|
Background: It has been reported, but not proven, that perioperative blood transfusions have a detrimental effect on the survival of patients undergoing surgery for lung cancer.
Study design and methods: A prospective study was carried out on the patients undergoing lobectomy for stage I lung cancer at our department from 1995 to 2000. The criteria for exclusion included previous cases of malignancy, autoimmune diseases, and any other relevant comorbidity.
Results: Two hundred eighty-one patients were observed, 24.6% of whom received transfusions. The only significant difference between the transfused and nontransfused patients was their preoperative hemoglobin (Hb) concentration (12.5 ± 1.20 g/dL vs 13.3 ± 1.22 g/dL, p < 0.001). The disease-free interval of the transfused patients was significantly lower than that of the nontransfused patients (53% vs 78% at 73 months, p < 0.005), as was also the case for actuarial survival (52% vs 71% at 73 months, p < 0.02). Blood transfusion was significantly predictive of tumor relapse according to the Cox model adjusted for the T state, preoperative Hb concentration, sex, age, histologic type, and grading (hazard ratio, 2.3; p = 0.017).
Conclusions: Our data show that perioperative blood transfusion is significantly correlated to worse prognosis in patients undergoing surgery for stage I lung cancer.
Key Words: blood transfusion • lung neoplasms • recurrence
|
Introduction
|
|---|
Between 1986 and 1994, numerous observational studies investigated the effects of blood transfusion on the survival of patients undergoing pulmonary resection for cancer. Nonetheless, the attractive hypothesis that the nonspecific immunosuppression caused by transfusion1
may support tumor growth and lower survival has not been definitively proven.2
3
4
5
6
7
8
9
The immune disorder caused by blood transfusion is nonetheless a clinical phenomenon that is well described in the medical literature and goes by the name allogeneic blood transfusion-associated immunomodulation (TRIM).10
In addition to the field of oncology, TRIM also appears to be involved in the improved survival of transplanted kidneys, the reduction of relapses in Crohn disease, the higher incidence of postoperative bacterial infections, and the activation of cytomegalovirus infections.
The majority of retrospective (or partially retrospective9
) observational studies that have analyzed the relationship between TRIM and lung cancer tend to demonstrate that patients who receive transfusions during the perioperative period have a higher mortality rate and/or higher incidence of tumor relapse than patients who do not receive transfusions. However, in these studies it is possible to observe that the transfused patients differ from the nontransfused patients in terms of several important prognostic factors, such as clinical stage, frequency of systemic diseases, and extent of surgery. Consequently, despite the use of multivariate statistical models of regression, the reported adverse effect of perioperative blood transfusion may be influenced by known and unknown confounding factors.11
In 1995, we commenced a prospective observational study on a selected sample of patients undergoing lobectomy for stage I lung cancer with the purpose of providing further data on the effect of blood transfusion on patients undergoing surgery for lung cancer.
|
Materials and Methods
|
|---|
Data were prospectively gathered for all of the 324 patients who underwent pulmonary lobectomy for stage I non-small cell lung cancer between 1995 and 2000. In particular, we recorded their age, sex, histologic type, T state, and preoperative hemoglobin (Hb) values. The T state was defined according to the International System for Staging Lung Cancer.12
Thirty-eight patients with histories of other tumors, previous cycles of chemotherapy or radiotherapy, autoimmune diseases, chronic obstructive bronchitis (FEV1 < 70% of predicted value13
), hepatic or renal insufficiency, and clinical factors of intermediate or higher cardiac risk14
were excluded. Finally, five patients with preoperative Hb levels of 
11 g/dL or affected by qualitative or quantitative platelet deficiencies were also excluded.
Perioperative transfusion was defined as the administration of any type of blood product during surgery and the following 7 days. The units of packed red cells without buffy coat administered to the patients contained 140 to 150 mL of erythrocytes (50 to 55 g of Hb) suspended in an additive solution (physiologic solution, adenine, glucose, and mannitol) and 10 to 20 mL of plasma. These units also contained approximately 1 x 109 leukocytes. The units of fresh frozen plasma contained an average of 250 mL of plasma and approximately 500,000 leukocytes.
The following transfusion criteria were observed for the infusion of packed red cells: estimated postoperative Hb levels < 10 g/dL for intraoperative transfusions, and < 8.5 g/dL for postoperative transfusions. For hemorrhagic patients, transfusions of fresh frozen plasma were performed when the prothrombin and/or partial thromboplastin time values, expressed as a patient/control ratio, were 
1.4. All of the patients underwent pulmonary lobectomy and mediastinal lymphadenectomy by means of posterolateral thoracotomy with muscular preservation, following orotracheal intubation with a double-lumen tube and invasive monitoring of the central venous and arterial pressure. Only patients with stage I cancer, according to the pathologic staging, were enrolled in the study; none of them underwent postoperative adjuvant therapy until the eventual appearance of tumor relapse.
Following surgery, our oncologic surveillance program monitored the patients in order to register the occurrence of tumor relapse or death. The main statistical point of the study consisted of the time interval of tumor relapse, followed by survival. We used the 
2 test, t test, and Mann-Whitney rank-sum test to relate the individual clinical variables to the transfusion state. The disease-free interval and survival were examined using the Kaplan-Meier and log-rank methods for each clinical variable, while the multivariate analysis employed the Cox proportional risk regression model.
|
Results
|
|---|
Two hundred eighty-one patients undergoing pulmonary lobectomy for non-small cell lung cancer were enrolled in the study. Sixty-nine patients (24.6%) received blood transfusions and constituted the group of transfused patients. They received a total of 163 U of packed red cells (28 U during the operation and 135 U in the postoperative period) and 15 U of fresh frozen plasma. The remaining 212 patients (75.4%) constituted the group of nontransfused patients.
Age, sex, histology, grading, and T factor were not found to be significantly different between the two groups (Table 1
). The preoperative Hb level was the only value that was significantly lower in the group of transfused patients (p < 0.001).
The median follow-up time was 34 months; 38 patients were unavailable for follow-up. Among the parameters examined, only the transfusion state and the T factor (T1 vs T2) were found to be significantly correlated to the disease-free interval and survival. Figures 1 , 2
show the actuarial survival curves in relation to the dimensions of the tumor; the patients with T2 tumors had a significantly lower disease-free interval and survival rate than the T1 patients (p < 0.05). Figures 3
, 4
show the actuarial curves in relation to the transfusion state; the transfused patients had a significantly lower disease-free interval and survival rate than the nontransfused patients (p < 0.005 and p < 0.02, respectively).
The Cox model enabled us to establish that the administration of blood products constituted a significantly predictive factor of tumor relapse when adjusted according to the T state, grading, histologic type, preoperative Hb level, age, and sex (p = 0.017). The T state was also significantly predictive of relapse (p = 0.028). Similarly, blood transfusion was significantly correlated to mortality (p = 0.045) [Table 2
]. Among the transfused patients, the number of blood transfusions was not correlated with disease-free interval and survival rate. The type of recurrence was not correlated to transfusion state.
|
Discussion
|
|---|
Thoracic oncologic surgery requires blood transfusion in a considerable percentage of patients (25 to 55%15
). Since the transfusion of blood components during the perioperative period does not constitute a definitive therapy but the correction of a temporary deficit, the objective of the transfusion must not be to restore preoperative Hb levels, but rather to ensure sufficient levels to permit adequate tissue oxygenation. In order to achieve this, Shoemaker16
suggests a Hb concentration > 9.2 g/dL along with a level of arterial saturation > 95%, a cardiac index > 4.5 L/min/m2, an oxygen extraction level < 31%, and an adequate circulating volume. The Italian guidelines on blood transfusion in surgery state that blood transfusion is not advisable when the Hb level is > 10 g/dL.17
Although we achieved the objective of staying below this threshold in transfused patients at discharge, our greatest preoccupation concerning blood transfusions in oncology patients, apart from the problems of viral diseases, was the occurrence of TRIM, with its supposed impact on tumor development.
The mechanisms governing the occurrence of TRIM are still unknown, although experimental studies suggest that a central role is played by the transfusion of leukocytes.18
A careful meta-analysis of the three randomized studies19
20
21
that have compared the risk of rectal-colic tumor relapse in patients receiving transfusions of autologous blood, leukocyte reduced blood, and allogeneic blood without buffy coat demonstrated that it is not possible to establish a statistical correlation between the transfusion of leukocytes and tumor relapse.22
Furthermore, the results of the observational part of the same randomized studies are contradictory regarding the incidence of tumor relapse in transfused patients vs nontransfused patients.
More than 150 observational studies have investigated the relationship between blood transfusion and tumor progression and/or survival, and the majority of studies fundamentally report a negative effect of transfusion. Meta-analysis studies23
24
25
have attributed the lower survival of transfused patients to a possible negative effect of uncontrolled factors of confusion. For example, the lower survival of transfused patients could be due to the greater prevalence of comorbidity. Or, as patients at different stages are almost always observed, the advanced stage of the oncologic disease could itself increase the need for allogeneic blood transfusion (due to greater surgical difficulties and/or decline of general conditions), and consequently the lower survival could simply be linked to an indirect effect of the advanced stage rather than to TRIM. The same criticisms are true for the studies performed on lung cancer, and this led us to carry out our own observational study. We enlisted 281 patients with non-small cell lung cancer, all at the same stage (stage I) and subjected to the same operation (pulmonary lobectomy and mediastinal lymphadenectomy) performed by the same surgical team. Great care was taken to exclude patients with comorbidity. Patients who quite definitely needed perioperative blood transfusion (low preoperative Hb level or platelet deficiency) were excluded in order to eliminate confounding factors. The transfusion criteria established earlier were scrupulously observed, as demonstrated by the Hb concentration of the transfused patients at discharge.
A statistical analysis of the results of our study shows a higher incidence of tumor relapse in patients with T2 tumors. This result was to be expected, and indeed it is true that as early as the 1997 revision of the International System for Staging Lung Cancer12
subdivided stage I into stages IA and IB, precisely on the basis of the T factor. We considered it appropriate to retain the original recruitment criteria for our study because stages IA and IB nonetheless comprise the patients with the best prognosis in respect to those who show signs of lymph nodes or remote metastasis. Furthermore, those stage IB patients who would have caused a modification of the surgical strategy were automatically excluded, as pneumonectomy and bilobectomy were not comprised in the criteria for inclusion. An examination of the actuarial survival curves relating to the disease-free interval of transfused and nontransfused patients reveals a correlation between blood transfusion and tumor relapse (Fig 3) , and this correlation is also confirmed by the Cox proportional risk regression model, adjusted for the variables mentioned, including the T state (Table 2)
.
In our opinion, this result provides strong evidence for the existence of a correlation between blood transfusion and the progression of lung cancer, particularly as it was obtained in such a select sample of patients as to minimize the influence of confounding factors. It remains to be ascertained whether the transfusion of blood without leukocytes eliminates or considerably reduces the correlation between transfusion and tumor relapse, also in consideration of the increase in costs that would be associated with the generalized use of such preparations.
|
Footnotes
|
|---|
Abbreviations: Hb = hemoglobin; TRIM = allogeneic blood transfusion-associated immunomodulation
Received for publication July 8, 2002.
Accepted for publication October 24, 2002.
|
References
|
|---|
- Gascon, P, Zoumbos, NC, Young, NS (1984) Immunologic abnormalities in patients receiving multiple blood transfusion. Ann Intern Med 100,173-177[CrossRef][ISI][Medline]
- Tartter, PI, Burrows, L, Kirschner, P Perioperative blood transfusion adversely affects prognosis after resection of stage I (subset N0) non-oat cell lung cancer. J Thorac Cardiovasc Surg 1984;88,659-662[Abstract]
- Hyman, NH, Foster, RS, DeMeules, JE, et al Blood transfusion and survival after lung cancer resection. Am J Surg 1985;149,502-507[CrossRef][ISI][Medline]
- Pastorino, U, Valente, M, Cataldo, I, et al Perioperative blood transfusion and prognosis of resected stage Ia lung cancer. Eur J Clin Oncol 1986;22,1375-1378
- Keller, SM, Groshen, S, Martini, N, et al Blood transfusion and lung cancer recurrence. Cancer 1988;62,606-610[CrossRef][ISI][Medline]
- Moores, DWO, Piantadosi, S, McKneally, MF Effect of perioperative blood transfusion on outcome in patients with surgically resected lung cancer. Ann Thorac Surg 1989;47,346-351[Abstract]
- Little, AG, Wu, HS, Ferguson, MK, et al Perioperative blood transfusion adversely affects prognosis of patients with stage I non-small cell lung cancer. Am J Surg 1990;160,630-633[CrossRef][ISI][Medline]
- Pena, CM, Rice, TW, Ahmad, M, et al Significance of perioperative blood transfusion in patients undergoing resection of stage I and II non-small-cell lung cancer. Chest 1992;102,84-88[Abstract/Free Full Text]
- Piantadosi, S, Moores, DWO, McKneally, MF The adverse effect of perioperative blood transfusion in lung cancer. Chest 1994;106(suppl),382-384
- Blajchman, MA Allogeneic blood transfusion, immunomodulation and postoperative bacterial infection: do we have the answers yet? Transfusion 1997;37,121-125[CrossRef][ISI][Medline]
- Vamvakas, E, Moore, SB Blood transfusions and postoperative septic complications. Transfusion 1994;34,714-727[CrossRef][ISI][Medline]
- Mountain, CF Revisions in the International System for Staging Lung Cancer. Chest 1997;111,1710-1717[Abstract/Free Full Text]
- Siafakas, MN, Vermiere, P, Pride, NB, et al Optimal assessment and management of chronic obstructive pulmonary disease (COPD): The European Respiratory Society Task Force. Eur Respir J 1995;8,1398-1420[CrossRef][ISI][Medline]
- Eagle, KA, Brundage, BH, Chaitman, BR, et al Guidelines for perioperative cardiovascular evaluation for non cardiac surgery: report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines. J Am Coll Cardiol 1996;27,910-948[CrossRef][ISI][Medline]
- Gwin, JL, Keller, SM Blood transfusion practices after resection of intrathoracic neoplasm. J Surg Oncol 1993;54,34-37[ISI][Medline]
- Shoemaker, WC Pathophysiology and therapy of shock syndromes. Shoemaker, WC Thompson, WL Holbrook, PR eds. Textbook of critical care 1984,52-72 W.B. Saunders. Philadelphia, PA:
- Il buon uso del sangue. Rome, Italy: Ministero della Sanità, 1993
- Bordin, JO, Heddle, NM, Blajchman, MA Biologic effects of leukocytes present in transfused cellular blood products. Blood 1994;84,1705-1721
- Busch, ORC, Hop, WCJ, van Papendrecht, MAWH, et al Blood transfusions and prognosis in colorectal cancer. N Engl J Med 1993;328,1372-1376[Abstract/Free Full Text]
- Houbiers, JGA, Brand, A, van de Watering, LMG, et al Randomized controlled trial comparing transfusion of leucocyte-depleted or buffy-coat-depleted blood in surgery for colorectal cancer. Lancet 1994;344,573-578[CrossRef][ISI][Medline]
- Heiss, MN, Mempel, W, Delanoff, C, et al Blood transfusion-modulated tumor recurrence: first results of randomized study of autologous versus allogeneic blood transfusion in colorectal cancer surgery. J Clin Oncol 1994;12,1859-1867[Abstract/Free Full Text]
- Vamvakas, EC, Blajchman, MA Deleterious clinical effects of transfusion-associated immunomodulation: fact or fiction? Blood 2001;97,1180-1195[Free Full Text]
- Chung, M, Steinmetz, OK, Gordon, PH Perioperative blood transfusion and outcome after resection for colorectal cancer. Br J Surg 1993;80,427-432[ISI][Medline]
- Vamvakas, EC Perioperative blood transfusion and cancer recurrence: meta-analysis for explanation. Transfusion 1995;35,760-768[CrossRef][ISI][Medline]
- Brand, A, Houbiers, JGA Clinical studies of blood transfusion and cancer. Vamvakas, EC Blajchman, MA eds. Immunomodulatory effects of blood transfusion 1999,145-190 AABB Press. Bethesda, MD:
TOP
Abstract
Introduction
