- Email: [email protected]
Postsplenectomy infection in patients with chronic leukemia
SCIENTIFIC PAPERS
Postsplenectomy Infection in Patients With Chronic Leukemia
Willam R. Mower, MD, Job A. Hawkins, MD, and Edward W. Nelson, MD, Salt Lake Ci,
The spleen is known to play an important role in various immunologic functions as demonstrated by the increased risk of infection due to encapsulated bacteria after splenectomy [I]. Although well documented in the pediatric age group [2], the question of an increased susceptibility to infection in otherwise normal asplenic adults has only recently been addressed [I]. The added potential risks of splenectomy in specific populations of immunocompromised patients, such as organ transplant recipients [3,4] and patients with Hodgkin’s disease, has also been reported [5,6]. This review was designed to examine the risk of postsplenectomy infection in a specific subset of immunocompromised hosts, namely patients with chronic lymphocytic leukemia and chronic myelogenous leukemia. PatiWltSandbthOdS The clinical records of all patients admitted to the University of Utah Medical Center from January 1962 until December 1984 with the diagnosis of chronic leukemia were examined, with the exception of patients who had hairy cell leukemia and erythroleukemia. All patients with the diagnosis of chronic myelogenous leukemia or chronic lymphocytic leukemia who had undergone splenectomy were then more closely reviewed for postsplenectomy infectious complications. For comparison, a group of age- and sex-matched control subjects seen during the same time period with the diagnosis of chronic myelogenous leukemia and chronic lymphocytic leukemia but not undergoing splenectomy were similarly reviewed. A serious infection was defined as one that occurred during hospitalization for the treatment of leukemia or that required rehospitalization. In the patients who underwent splenectomy, this included only those infections that occurred after both the diagnosis of leukemia and operative removal of the spleen. For the patients who did not underFrom the Dep&me~ of Surgery, University of Utah School of Medicine, Salt Lake City, IJteb. Requests for reprints shod be addressed to Edwerd W. Nelson, MD, Department of Surgery, Universtty of Uteh htedical Center, 50 North Medical Drive. salt L&e City, Uteh 84132. Preserited at the 38th Annual Meeting of the Southwestern Surgical Congress, San Francisco, California. April 21-24. 1986.
votume 152, thamber
1986
Utah
go splenectomy, serious infection included all infections occurring after the diagnosis of chronic leukemia had been made. For each patient in the study, data were collected regarding the type of chronic leukemia, age at diagnosis, sex of the patient, whether or not splenectomy had been performed, length of survival, and history of serious infection. All serious infections were categorized by date of occurrence, type of organisms, location, and relation to patient survival. Statistical analysis was performed using the chi-square test for contingency tables and the Student’s t test for comparison of means. Probabilities of less than 5 percent (p <0.05) were reported as statistically significant.
Real&S From January 1962 until December 1964, a total of 23 patients with chronic leukemia underwent splenectomy for the following indications: splenic infarction and pain, severe splenomegaly, and hypersplenism. This group included 14 patients with chronic lymphocytic leukemia and 9 patients with chronic myelogenous leukemia. A matched group of 48 patients with chronic lymphocytic leukemia and 39 patients with chronic myelogenous leukemia treated without splenectomy was also evaluated. There was no difference in age or in the sex distribution between the two groups. The mean age of patients with chronic myelogenous leukemia with and without splenectomy was 44 years; whereas the mean age of patients with chronic lymphocytic leukemia who had undergone splenectomy was 62 years, and for those with chronic lymphocytic leukemia who had not undergone splenectomy, 64 years. Mean overall survival from the time of diagnosis did not differ significantly between the patients with chronic lymphocytic leukemia who underwent splenectomy 18% months) and the patients with chronic lymphocytic leukemia without splenectomy (48 months). A similar but still nonsignificant difference in mean survival was seen between patients with chronic myelogenous leukemia who underwent splenectomy (61 months) and those patients with
583
Mower eta1
TABLE I
Types of OrganismsIsolated In Chronic Leukemia Patients Wlth and Without Splenectomy No.of isolates
CUMULATIVE PEACENT 6o
j
/--ti
0
I
2
3
4
5
6
7
8 9 IO
YFBRS FIgwe 1. Cunwlatlve mnnber ot Intectkma aftef sptenectomy In pattents wlttt chrmlc Ieukemla comparad with the cumulative ttumber after the dlagnnsts of chronic leukemia In patlents wlthout aplenectomy. Mast Infectknis occurred during the tkst 3 years atter splenectomy. 7Ra dltterence between gruups was ltt#ly s&rMlcant (p <0.007).
TABLE II
Ty~ar of lnfectlcns In Chrcnk Leukemia Patients Wlth and Wlttwut Spianectomy No.of cases
infection Pneumonia Sepsis Abscess Other
Splenectomy 23 14 5 8
No Splenectomy 33 12 1 4
chronic myelogenous leukemia who did not undergo eplenectomy (36 months). Although survival was not significantly changed for eplenectomy and nonsplenectomy patients, there was a significant influence on the incidence of serious infections. Fifteen of 23 patients undergoing splenectomy (65 percent) were found to have had at least one serious infection (10 in the chronic lymphocytic leukemia group and 5 in the chronic myelogenous leukemia group) compared with 29 of 87 patients (35 percent) who did not undergo splenectomy (15 in the chronic lymphocytic leukemia group and 14 in the chronic myelogenous leukemia group) (p
584
Organisms
1
Splenectomy
Pseudomonas aeruginosa Staphylococcus aureus Streptococcus pneumoniae Salmonella species Hemophilus influenza Escherichia coli Pneumocystis carlnii Fungal Unidentified l
11 7 4 4 3 2 0 10 13
No Splenectomy 3’ 11 3 0 5 4 3 16 14
p < 0.005.
mia patients without splenectomy (28 months) (p
The American Journal of Surgery
Postsplenectomy Infection in Chronic Leukemia
matic splenectomy in healthy adults, although previously controversial, appears to have increased slightly but significantly over that in the adult population who have undergone spelenctomy without any postoperative trauma [I]. Past reviews of specific patient groups other than children and healthy adults have concentrated on patients who are otherwise immunocompromised due to chemotherapy, radiotherapy, immunosuppressive drugs, or hematologic malignancies [4-71. For example, Corker et al [5] and Shimm et al [6] have reported that patients undergoing staging laparotomy for Hodgkin’s disease rarely have an increased incidence of serious postsplenectomy infection unless they are otherwise predisposed by extremes of age, advanced stage of disease, leukopenia, or ingestion of steroids. Others who have reviewed infectious complications after splenectomy in a wide variety of hematologic disorders have found the highest rates (17 percent to 20 percent) in patients with chronic leukemia and myeloid metaplasia [8]. Although the potentials for immunologic consequences after splenectomy are consistent with the contribution of the spleen to the body’s total lymphatic tissue and reticuloendothelial mass [9], the exact mechanism of postsplenectomy infection has not been defined. The loss of a variety of well known splenic functions, including filtration of bacteria, the production of immunoglobulin M, and loss of various opsonins (tuftsin and properdin), are all presumably of importance [2,9]. Possibly, the addition of impaired cellular and humoral immunity found during chronic leukemia and secondary chemotherapy have an additive effect on the loss of splenic function. Overall, other than hairy cell leukemia, splenectomy has a limited role in chronic leukemia [IO]. Only a minority of patients with chronic lymphocytic leukemia have development of complications that justify splenectomy, although palliation and prolonged survival have occasionally been reported in some patients with hemolytic anemia or symptomatic splenomegaly [II]. There is even less enthusiasm for splenectomy in patients with chronic myelogenous leukemia, the common belief being that the procedure should be limited to patients with acute symptoms who are at risk for splenic rupture or symptomatic massive splenomegaly. Previous controversy over elective splenectomy in chronic myelogenous leukemia to delay the blastic phase and prolong survival has now been resolved by two large randomized studies that have suggested no benefit from elective splenic removal in patients with chronic myelogenous leukemia [12,13]. This review confirms the conclusion of previous reports regarding splenectomy in patients with chronic leukemia and documents the additional postsplenectomy risk of infection that must be considered. Based on the data presented, we suggest
Volume 152, December 1988
that splenectomy in patients with chronic leukemia influences the disease process and, therefore, symptoms of the disease must be carefully balanced against the significant risk of postspenectomy infectious morbidity. Summary The added risk of infectious complications due to splenectomy in patients already immunocompromised because of chronic leukemia was studied over a 22 year period. When compared to patients with chronic leukemia who did not undergo splenectomy, survival was not influenced. Splenectomy did significantly increase the total number of serious infections (65 percent versus 35 percent, p
585
Mower et al
trial of early splenectomy in chronic myeloid leukemia. Br J Haematol 1983;54:415-30.
Discussion Carey P. Page (San Antonio, TX): Patients with leukemia manifest altered humoral cell-mediated immunity and phagocytosis. Removal of the spleen makes all of these conditions worse. By examining the 50 isolates that were reported, it becomes clear that the majority of the increased infections were accounted for by those organisms cleared primarily by humoral mechanisms. As you pointed out, Dr. Hawkins, Pseudomonas was primarily responsible. Organisms cleared primarily by cell-mediated or phagocytic mechanisms actually occurred with increased frequency in the group of patients who did not undergo splenectomy. Dr. Hawkins, you have taken an unusual approach for a surgical series to the evaluation of postsplenectomy infections in patients with leukemia. You have ignored the perioperative morbidity and mortality associated with splenectomy in these patients and have documented an increased incidence of infection, frequency of episodes, and increased risk of death from overwhelming sepsis in patients who survived splenectomy. These patients were matched for duration of disease, age, and gender; still, there is something different about the two groups of patients, that being that one group required splenectomy. My first question is, specificahy, why did these patients require splenectomy? Is the difference you observed in septic episodes intrinsic or caused by splenectomy? Secondly, did you quantitate in the splenectomy and nonsplenectomy groups, the disease activity at the time of infection by examining their marrow or peripheral blood? You reviewed spontaneously occurring infections and those associated with hospitalization and drug therapy for leukemia. Was there any difference between the two groups? Did one group have, for instance, a higher rate of
566
infection associated with treatment and the other a higher rate of spontaneous infection? In retrospect, Dr. Hawkins, can you account for the “unidentified” organisms that caused the septic course in your patients? I believe that another important aspect of this study is its failure to demonstrate a shortened survival period in patients with leukemia requiring splenectomy despite a higher mortality from infection. This encouraging observation should prompt aggressive management of febrile episodes, comprehensive immunization programs, and documentation of the efficacy of immunization. Kent C. Westbrook (Little Rock, AR): Dr. Hawkins, you said you gave consideration to antibiotics. Specifically, did you administer antibiotics, and if so, which ones? John A. Hawkins (closing): Dr. Page, we did not participate in any of the prospective, randomized trials, so all of these patients had splenectomy for very definite indications. The indications included hemolytic anemia unresponsive to steroid therapy, thrombocytopenia, and a few cases of massive symptomatic splenomegaly. Another of your questions dealt with unidentified organisms in the study. Those were believed to be viral infections. This phenomenon has been reported in patients with Hodgkin’s disease in whom an increased incidence of viral infections after splenectomy was also noticed. Again, this occurred in an uncontrolled retrospective study. Dr. Westbrook, I think the word “consideration” is a nice way to hedge on a treatment when you do not really have hard data to back it up. I think that patients with chronic leukemia who have had a splenectomy and are going to undergo a surgical procedure, no matter how minor, should receive prophylactic antibiotics to cover both the gram-positive and the gram-negative organisms. Some investigators believe that young splenectomized patients should be given long-term prophylactic antibiot: its.
The Amerken Journal04 Surgery
Postsplenectomy Infection in Patients With Chronic Leukemia
Willam R. Mower, MD, Job A. Hawkins, MD, and Edward W. Nelson, MD, Salt Lake Ci,
The spleen is known to play an important role in various immunologic functions as demonstrated by the increased risk of infection due to encapsulated bacteria after splenectomy [I]. Although well documented in the pediatric age group [2], the question of an increased susceptibility to infection in otherwise normal asplenic adults has only recently been addressed [I]. The added potential risks of splenectomy in specific populations of immunocompromised patients, such as organ transplant recipients [3,4] and patients with Hodgkin’s disease, has also been reported [5,6]. This review was designed to examine the risk of postsplenectomy infection in a specific subset of immunocompromised hosts, namely patients with chronic lymphocytic leukemia and chronic myelogenous leukemia. PatiWltSandbthOdS The clinical records of all patients admitted to the University of Utah Medical Center from January 1962 until December 1984 with the diagnosis of chronic leukemia were examined, with the exception of patients who had hairy cell leukemia and erythroleukemia. All patients with the diagnosis of chronic myelogenous leukemia or chronic lymphocytic leukemia who had undergone splenectomy were then more closely reviewed for postsplenectomy infectious complications. For comparison, a group of age- and sex-matched control subjects seen during the same time period with the diagnosis of chronic myelogenous leukemia and chronic lymphocytic leukemia but not undergoing splenectomy were similarly reviewed. A serious infection was defined as one that occurred during hospitalization for the treatment of leukemia or that required rehospitalization. In the patients who underwent splenectomy, this included only those infections that occurred after both the diagnosis of leukemia and operative removal of the spleen. For the patients who did not underFrom the Dep&me~ of Surgery, University of Utah School of Medicine, Salt Lake City, IJteb. Requests for reprints shod be addressed to Edwerd W. Nelson, MD, Department of Surgery, Universtty of Uteh htedical Center, 50 North Medical Drive. salt L&e City, Uteh 84132. Preserited at the 38th Annual Meeting of the Southwestern Surgical Congress, San Francisco, California. April 21-24. 1986.
votume 152, thamber
1986
Utah
go splenectomy, serious infection included all infections occurring after the diagnosis of chronic leukemia had been made. For each patient in the study, data were collected regarding the type of chronic leukemia, age at diagnosis, sex of the patient, whether or not splenectomy had been performed, length of survival, and history of serious infection. All serious infections were categorized by date of occurrence, type of organisms, location, and relation to patient survival. Statistical analysis was performed using the chi-square test for contingency tables and the Student’s t test for comparison of means. Probabilities of less than 5 percent (p <0.05) were reported as statistically significant.
Real&S From January 1962 until December 1964, a total of 23 patients with chronic leukemia underwent splenectomy for the following indications: splenic infarction and pain, severe splenomegaly, and hypersplenism. This group included 14 patients with chronic lymphocytic leukemia and 9 patients with chronic myelogenous leukemia. A matched group of 48 patients with chronic lymphocytic leukemia and 39 patients with chronic myelogenous leukemia treated without splenectomy was also evaluated. There was no difference in age or in the sex distribution between the two groups. The mean age of patients with chronic myelogenous leukemia with and without splenectomy was 44 years; whereas the mean age of patients with chronic lymphocytic leukemia who had undergone splenectomy was 62 years, and for those with chronic lymphocytic leukemia who had not undergone splenectomy, 64 years. Mean overall survival from the time of diagnosis did not differ significantly between the patients with chronic lymphocytic leukemia who underwent splenectomy 18% months) and the patients with chronic lymphocytic leukemia without splenectomy (48 months). A similar but still nonsignificant difference in mean survival was seen between patients with chronic myelogenous leukemia who underwent splenectomy (61 months) and those patients with
583
Mower eta1
TABLE I
Types of OrganismsIsolated In Chronic Leukemia Patients Wlth and Without Splenectomy No.of isolates
CUMULATIVE PEACENT 6o
j
/--ti
0
I
2
3
4
5
6
7
8 9 IO
YFBRS FIgwe 1. Cunwlatlve mnnber ot Intectkma aftef sptenectomy In pattents wlttt chrmlc Ieukemla comparad with the cumulative ttumber after the dlagnnsts of chronic leukemia In patlents wlthout aplenectomy. Mast Infectknis occurred during the tkst 3 years atter splenectomy. 7Ra dltterence between gruups was ltt#ly s&rMlcant (p <0.007).
TABLE II
Ty~ar of lnfectlcns In Chrcnk Leukemia Patients Wlth and Wlttwut Spianectomy No.of cases
infection Pneumonia Sepsis Abscess Other
Splenectomy 23 14 5 8
No Splenectomy 33 12 1 4
chronic myelogenous leukemia who did not undergo eplenectomy (36 months). Although survival was not significantly changed for eplenectomy and nonsplenectomy patients, there was a significant influence on the incidence of serious infections. Fifteen of 23 patients undergoing splenectomy (65 percent) were found to have had at least one serious infection (10 in the chronic lymphocytic leukemia group and 5 in the chronic myelogenous leukemia group) compared with 29 of 87 patients (35 percent) who did not undergo splenectomy (15 in the chronic lymphocytic leukemia group and 14 in the chronic myelogenous leukemia group) (p
584
Organisms
1
Splenectomy
Pseudomonas aeruginosa Staphylococcus aureus Streptococcus pneumoniae Salmonella species Hemophilus influenza Escherichia coli Pneumocystis carlnii Fungal Unidentified l
11 7 4 4 3 2 0 10 13
No Splenectomy 3’ 11 3 0 5 4 3 16 14
p < 0.005.
mia patients without splenectomy (28 months) (p
The American Journal of Surgery
Postsplenectomy Infection in Chronic Leukemia
matic splenectomy in healthy adults, although previously controversial, appears to have increased slightly but significantly over that in the adult population who have undergone spelenctomy without any postoperative trauma [I]. Past reviews of specific patient groups other than children and healthy adults have concentrated on patients who are otherwise immunocompromised due to chemotherapy, radiotherapy, immunosuppressive drugs, or hematologic malignancies [4-71. For example, Corker et al [5] and Shimm et al [6] have reported that patients undergoing staging laparotomy for Hodgkin’s disease rarely have an increased incidence of serious postsplenectomy infection unless they are otherwise predisposed by extremes of age, advanced stage of disease, leukopenia, or ingestion of steroids. Others who have reviewed infectious complications after splenectomy in a wide variety of hematologic disorders have found the highest rates (17 percent to 20 percent) in patients with chronic leukemia and myeloid metaplasia [8]. Although the potentials for immunologic consequences after splenectomy are consistent with the contribution of the spleen to the body’s total lymphatic tissue and reticuloendothelial mass [9], the exact mechanism of postsplenectomy infection has not been defined. The loss of a variety of well known splenic functions, including filtration of bacteria, the production of immunoglobulin M, and loss of various opsonins (tuftsin and properdin), are all presumably of importance [2,9]. Possibly, the addition of impaired cellular and humoral immunity found during chronic leukemia and secondary chemotherapy have an additive effect on the loss of splenic function. Overall, other than hairy cell leukemia, splenectomy has a limited role in chronic leukemia [IO]. Only a minority of patients with chronic lymphocytic leukemia have development of complications that justify splenectomy, although palliation and prolonged survival have occasionally been reported in some patients with hemolytic anemia or symptomatic splenomegaly [II]. There is even less enthusiasm for splenectomy in patients with chronic myelogenous leukemia, the common belief being that the procedure should be limited to patients with acute symptoms who are at risk for splenic rupture or symptomatic massive splenomegaly. Previous controversy over elective splenectomy in chronic myelogenous leukemia to delay the blastic phase and prolong survival has now been resolved by two large randomized studies that have suggested no benefit from elective splenic removal in patients with chronic myelogenous leukemia [12,13]. This review confirms the conclusion of previous reports regarding splenectomy in patients with chronic leukemia and documents the additional postsplenectomy risk of infection that must be considered. Based on the data presented, we suggest
Volume 152, December 1988
that splenectomy in patients with chronic leukemia influences the disease process and, therefore, symptoms of the disease must be carefully balanced against the significant risk of postspenectomy infectious morbidity. Summary The added risk of infectious complications due to splenectomy in patients already immunocompromised because of chronic leukemia was studied over a 22 year period. When compared to patients with chronic leukemia who did not undergo splenectomy, survival was not influenced. Splenectomy did significantly increase the total number of serious infections (65 percent versus 35 percent, p
585
Mower et al
trial of early splenectomy in chronic myeloid leukemia. Br J Haematol 1983;54:415-30.
Discussion Carey P. Page (San Antonio, TX): Patients with leukemia manifest altered humoral cell-mediated immunity and phagocytosis. Removal of the spleen makes all of these conditions worse. By examining the 50 isolates that were reported, it becomes clear that the majority of the increased infections were accounted for by those organisms cleared primarily by humoral mechanisms. As you pointed out, Dr. Hawkins, Pseudomonas was primarily responsible. Organisms cleared primarily by cell-mediated or phagocytic mechanisms actually occurred with increased frequency in the group of patients who did not undergo splenectomy. Dr. Hawkins, you have taken an unusual approach for a surgical series to the evaluation of postsplenectomy infections in patients with leukemia. You have ignored the perioperative morbidity and mortality associated with splenectomy in these patients and have documented an increased incidence of infection, frequency of episodes, and increased risk of death from overwhelming sepsis in patients who survived splenectomy. These patients were matched for duration of disease, age, and gender; still, there is something different about the two groups of patients, that being that one group required splenectomy. My first question is, specificahy, why did these patients require splenectomy? Is the difference you observed in septic episodes intrinsic or caused by splenectomy? Secondly, did you quantitate in the splenectomy and nonsplenectomy groups, the disease activity at the time of infection by examining their marrow or peripheral blood? You reviewed spontaneously occurring infections and those associated with hospitalization and drug therapy for leukemia. Was there any difference between the two groups? Did one group have, for instance, a higher rate of
566
infection associated with treatment and the other a higher rate of spontaneous infection? In retrospect, Dr. Hawkins, can you account for the “unidentified” organisms that caused the septic course in your patients? I believe that another important aspect of this study is its failure to demonstrate a shortened survival period in patients with leukemia requiring splenectomy despite a higher mortality from infection. This encouraging observation should prompt aggressive management of febrile episodes, comprehensive immunization programs, and documentation of the efficacy of immunization. Kent C. Westbrook (Little Rock, AR): Dr. Hawkins, you said you gave consideration to antibiotics. Specifically, did you administer antibiotics, and if so, which ones? John A. Hawkins (closing): Dr. Page, we did not participate in any of the prospective, randomized trials, so all of these patients had splenectomy for very definite indications. The indications included hemolytic anemia unresponsive to steroid therapy, thrombocytopenia, and a few cases of massive symptomatic splenomegaly. Another of your questions dealt with unidentified organisms in the study. Those were believed to be viral infections. This phenomenon has been reported in patients with Hodgkin’s disease in whom an increased incidence of viral infections after splenectomy was also noticed. Again, this occurred in an uncontrolled retrospective study. Dr. Westbrook, I think the word “consideration” is a nice way to hedge on a treatment when you do not really have hard data to back it up. I think that patients with chronic leukemia who have had a splenectomy and are going to undergo a surgical procedure, no matter how minor, should receive prophylactic antibiotics to cover both the gram-positive and the gram-negative organisms. Some investigators believe that young splenectomized patients should be given long-term prophylactic antibiot: its.
The Amerken Journal04 Surgery