Abstract
Introduction:
The spleen plays a role in both the immune response as well as coagulation, and a splenectomy may be required at the time of ovarian cancer surgery to achieve complete cytoreduction. The aim of the study was to correlate hematologic changes with development of infection and venous thromboembolism in patients undergoing splenectomy during cytoreductive surgery for ovarian cancer.
Methods:
This single-institution retrospective review includes all patients undergoing splenectomy during primary or interval cytoreductive surgery for advanced ovarian cancer March 2001 to December 2016. We compared postoperative hematologic changes (evaluated daily before discharge) in patients developing infection within 30 days post-surgery (Infection group) with those who did not (No-Infection group). We also compared patients developing venous thromboembolism with those who did not.
Results:
A total of 265 patients underwent splenectomy. Median age was 64 years (range 22-88); 146 (55%) patients had stage IIIC and 114 (43%) patients had stage IV. The majority, 201 (76%) patients underwent splenectomy during primary debulking. A total of 132 (50%) patients comprised the Infection group (urinary tract infection was most common (54%)). Median time from surgery to infection was 8 days (range, 0-29). After initial rise in white blood cell count in both groups, the Infection group had a second white blood cell peak on postoperative day 10 (median 16.6K/mcL, interquartile range 12.5-21.2) not seen in the No-Infection group (median 12K/mcL, interquartile range 9.3-16.3). A total of 40 (15%) patients developed venous thromboembolism. Median time from surgery to venous thromboembolism: 6.5 days (range, 1–43). All patients demonstrated continuous rise in platelets from postoperative days 0-15. Thrombocytosis was present in 38/40 (95%) patients with venous thromboembolism versus 183/225 (81%) patients without (p=0.036). Median number of days with thrombocytosis was higher in venous thromboembolism (8 days, range 1-15) versus non-venous thromboembolism (6 days, range 1-16, p=0.049) group.
Discussion:
We identified initial leukocytosis after splenectomy in all patients. The Infection group had a second peak in white blood cell count on postoperative day 10, not present in the No-Infection group. Among patients with venous thromboembolism, thrombocytosis was more frequent and of longer duration.
Keywords: Ovarian cancer, Surgical oncology, Surgical procedures
INTRODUCTION
At the time of debulking surgery for advanced stage ovarian cancer, splenectomy may be required to achieve maximal cytoreduction, due to parenchymal involvement or tumor extension to the peri-splenic region. The need for splenectomy has not been associated with decreased survival [1]. However, splenectomy is a known risk factor for infection, largely because of the spleen’s role in clearing encapsulated bacteria [2, 3]. The spleen also plays a significant role in immediate immune response to blood-borne antigens and is a major site of antibody production [4], removing bacteria and aged, antibody-coated cells from the blood [4]. After splenectomy, patients show immune abnormalities such as decreased phagocytosis, levels of Immunoglobulin M, and antigen response, increasing susceptibility to infection [5, 6]. Infection may be difficult to diagnose, given the frequency of postoperative leukocytosis even in the absence of infection [3, 7, 8]. The transient leukocytosis seen after shock and soft tissue damage is exaggerated post-splenectomy [3]. Trauma literature has focused on using postoperative serum platelet and white blood cell levels to help diagnose infection post-splenectomy, with infections reportedly more likely in patients with platelet-to-white blood cell ratio <20 [5, 8, 9].
Splenectomy is a known risk factor for a hypercoagulable state, with increased rate of thromboembolism—both venous and arterial—and increased risk of death from pulmonary embolism [2, 4]. In patients undergoing splenectomy for hematologic disease, the rate of venous thromboembolism is reported to be between 1.5-55% [4, 10]. The spleen plays an important role in hematologic function and likely a central role in mediating coagulation changes after injury [11]. One of its major functions is regulating platelet activity and clearing platelets from the blood, sometimes sequestering as many as 30-40% of circulating platelets [4, 11]. Thrombocytosis has been associated with increased risk of venous thromboembolism; reactive thrombocytosis (elevation in platelets secondary to the removal of the spleen and its platelet clearing role) is common post-splenectomy, occurring in up to 75% of patients [11]. Several studies have shown improved survival in patients receiving transfusions after cancer surgery, an effect that may be altered by splenectomy [12].
Our objective was to evaluate hematologic changes in patients undergoing splenectomy during cytoreductive surgery for advanced stage ovarian cancer, and to evaluate the correlation of these changes with development of postoperative infection and venous thromboembolisms.
METHODS
Following approval by our Institutional Review Board, we conducted a retrospective review of all patients undergoing splenectomy during primary debulking or interval debulking surgery for advanced stage ovarian, tubal, or primary peritoneal cancer from March 2001-December 2016. Data extracted from the electronic medical record included patient demographics, disease and treatment information, hematologic variables up to 30 days post-surgery, presence of infection and/or venous thromboembolism. Patients with a known predisposition to severe infection and/or venous thromboembolism were excluded, aside from increased risk due to cancer diagnosis.
We compared patients developing infection within 30 days post-surgery (Infection group) with those who did not (No-Infection group). Infection was defined on positive culture, or suggestive clinical and/or radiologic findings. Initiation of a work-up for suspected infection and the specific tests obtained, including cultures, were at the discretion of the treating physician. Types of infection included urinary tract, respiratory, gastrointestinal, wound, intra-abdominal, bloodborne. Patients not meeting the specified definition of infection but treated empirically with antibiotics for an infection according to the medical record were included in the Infection group. Women undergoing splenectomy receive appropriate vaccines postoperatively per institutional guidelines, which are in accordance with national recommendations.
We compared patients developing venous thromboembolism during the 30-day postoperative period after splenectomy, with those who did not. Venous thromboembolisms included deep venous thrombosis and pulmonary embolism, and was defined as radiologic evidence of either deep venous thromboembolism or pulmonary embolism on doppler venogram or contrast spiral chest computerized tomography, respectively. Patients diagnosed with venous thromboembolism were initially treated with either continuous intravenous infusion of heparin or subcutaneous enoxaparin and subsequently transitioned to outpatient management, per institutional guidelines. We compared patients developing thrombocytosis in the postoperative period, with those who did not. Thrombocytosis was defined as upper limit of institutional platelet count, 400 K/mcL. Extended venous thromboembolism prophylaxis was instituted at our center in 2013, in accordance with guidelines published by the American Society of Clinical Oncology; all patients received 28 days of prophylactic-dose enoxaparin postoperatively.
Detailed hematologic measurements such as white blood cells, percent neutrophils, percent lymphocytes, and platelet count, were recorded from postoperative day 0 to 15, censoring missing values for patients who were discharged prior to day 15. These values were used to develop line plots demonstrating corresponding hematologic changes. To compare Infection versus No-Infection and venous thromboembolism versus non-venous thromboembolism groups, we simplified postoperative day 0 to 15 measurements to a single number by accepting only the maximum value before infection for infected patients, and the maximum value from postoperative day 0 to 15 for non-infected patients. All results were obtained using these simplified numbers. P-values were obtained using Fisher’s exact test for categorical variables and Wilcoxon rank sum test for continuous variables.
RESULTS
Between March 2001 and December 2016, 265 women underwent splenectomy during cytoreductive surgery for ovarian cancer (Table 1). No splenectomies were performed for intraoperative injury; all patients had intraoperative findings suggestive of splenic involvement by disease. Median age was 64 years (range, 22–88). One hundred and forty-six (55%) presented had stage IIIC disease and 114 (43%) patients had stage IV. Splenectomy was performed at time of primary debulking surgery in 201 (76%) women, at interval debulking surgery in 64 (24%).
Table 1.
Patient demographics
| N (%) | |
|---|---|
|
| |
| Median age, years (range) | 64 (22 – 88) |
|
| |
| Stage | |
| ≤IIIB | 5 (2%) |
| IIIC | 146 (55%) |
| IV | 114 (43%) |
|
| |
| Histology | |
| High-grade Serous | 233 (88%) |
| Low-grade Serous | 11 (4%) |
| Clear cell | 7 (3%) |
| Carcinosarcoma | 6 (2%) |
| Endometrioid | 1 (<1%) |
| Other | 7 (3%) |
The Infection group comprised 132 (50%) patients; the No-Infection group, 133 (50%). In the Infection group, 34 (26%) patients were treated empirically with antibiotics; 98 (74%) patients were diagnosed with infection, of whom 64 (65%) patients had one infection and 34 (26%) patients had two or more. The most commonly diagnosed infection was urinary tract (n=53), followed by wound (n=24), intra-abdominal (n=23), respiratory (n=17), gastrointestinal (n=14), bloodborne (n=11). Median time from surgery to infection in the Infection group was 6.5 days (range 0–29). Women with infection were more likely to have undergone primary debulking surgery (Table 2) and had higher median estimated blood loss (1,380 versus 1,000 mL, p<0.001), length of hospital stay (13 versus 8 days, p<0.001), time from surgery to postoperative chemotherapy (39 versus 35 days, p=0.02). Extent of splenic involvement was not associated with increased risk for infection (p=0.73). Twenty-three patients (9%) underwent a segmental or wedge liver resection at the time of splenectomy; 13 (57%) of these patients developed an infection. In 242 patients who did not undergo a wedge liver resection, the infection rate was 49% (n=119, p=0.52).
Table 2.
Comparison of patients with and without infection after splenectomy at the time of debulking surgery for advanced stage ovarian cancer
| Infection (n=132) | No-Infection (n=133) | p-value | |
|---|---|---|---|
| Median WBC count, K/mcL (range) | 19.3 (4.8 – 51.5) | 16.8 (7.7 – 27.5) | <0.001 |
| Median % neutrophils (range) | 91.1 (71.2 – 97.7) | 90.4 (76.2 – 97) | 0.185 |
| Median % lymphocytes (range) | 12 (4.9 – 37.9) | 14 (3.9 – 38.8) | <0.001 |
| Median platelet count, K/mcL (range) | 556.5 (127 – 1,444) | 615 (196-1,397) | 0.315 |
| Median estimated blood loss, mL (range) | 1,380 (90 – 7,000) | 1,000 (100 – 4,000) | <0.001 |
| Median length of stay, days (range) | 13 (9 – 92) | 8 (4 – 27) | <0.001 |
| Time to start of postoperative chemotherapy, days (range) | 39 (9 – 92) | 35 (6 – 70) | 0.02 |
|
Surgery Type Primary Debulking Surgery Interval Debulking Surgery |
108 (54%) 24 (38%) |
93 (46%) 40 (62%) |
0.031 |
|
Splenic Involvement None Capsule only Parenchymal |
20 (15%) 80 (61%) 32 (24%) |
24 (18%) 81 (61%) 28 (21%) |
0.73 |
WBC, white blood cell
Patients in the Infection group had higher median white blood cell count (19.3 K/mcL, range 4.8–51.5 K/mcL) than those in the No-Infection group (p<0.001), although median white blood cell count was still above normal (16.8 K/mcL, range 7.7-27.5 K/mcL). Median neutrophil counts did not differ between groups. Figure 1 shows median white blood cell count between postoperative days 0 and 15 for Infection and No-Infection groups. After initial rise in white blood cell count in both groups (with a peak on postoperative day 2), the Infection group had a second peak on postoperative day 10 (median 16.6 K/mcL, interquartile range 12.5-21.2) not seen in the No-Infection group (median postoperative day 10 white blood cell 12 K/mcL, interquartile range 9.3-16.3, p<0.001).
Figure 1.
(A) Median white blood cell count, (B) median platelet count, (C) median platelet/WBC ratio between postoperative day 0 and 15 for patients with and without an infection after splenectomy
There was no difference in daily median platelet count from postoperative days 0 to 15 between Infection and No-Infection groups; both showed progressive increase (Figure 1). Platelet-to-white blood cell ratio was similar between groups; both showed an increased ratio postoperatively (Figure 1), with ratio >20 for all but three postoperative days.
Postoperative venous thromboembolism was diagnosed in 40 (15%) patients; 13 of these were diagnosed with a deep venous thromboembolism, 30 with a pulmonary embolism, and 3 developed both, with the pulmonary embolism diagnosed later in time in all three cases. Median time from surgery to diagnosis of venous thromboembolism was 6.5 days (range, 1–43). The development of a venous thromboembolism did not directly contribute to death in any of the patients, and were either grade 1 or 2. Thrombocytosis was more common after primary than interval debulking surgery (p<0.001) and lasted longer (7 versus 4 days, p<0.001); there was no difference in surgery type (primary versus interval debulking) for women with or without venous thromboembolism (p=0.07, Table 3). Risk of venous thromboembolism was not associated with operative time or intraoperative blood transfusion. In total, 150 patients received intraoperative blood transfusion: 146, packed red blood cells (median 2 units, range 1-15); 60, fresh frozen plasma and/or cryoprecipitate (median 2 units, range 1-17); 13, transfused platelets (median 1 unit, range 1-3). Transfusion of fresh frozen plasma and/or cryoprecipitate or platelets was not associated with increased venous thromboembolism risk (p=0.08 and 0.25, respectively).
Table 3.
Comparison of patients with and without a venous thromboembolism after splenectomy at the time of debulking surgery for advanced stage ovarian cancer
| VTE (n=40) | No VTE (n=225) | p-value | |
|---|---|---|---|
| Median WBC count, K/mcL (range) | 17.9 (4.8 – 51.5) | 18.3 (11.5 – 33.9) | 0.332 |
| Median % neutrophils (range) | 90.3 (71.2 – 97.9) | 91.9 (83 – 96.4) | 0.018 |
| Median % lymphocytes (range) | 13 (3.9 – 38.8) | 12.6 (5.5 – 24.3) | 0.248 |
| Median platelet count, K/mcL (range) | 591 (127 – 1444) | 622 (211 – 1364) | 0.36 |
| Median estimated blood loss, mL (range) | 1000 (90 – 5500) | 1350 (225 – 7000) | 0.033 |
| Median operative time, min (range) | 424 (131 – 898) | 420.5 (205 – 749) | 0.966 |
| Median length of stay, days (range) | 9 (3 – 48) | 15 (6 – 48) | <0.001 |
|
Intraoperative blood transfusion (%) No Yes |
13 (32.5%) 27 (67.5%) |
102 (45.3%) 123 (54.7%) |
0.166 |
|
Surgery Type (%) Primary Debulking Surgery Interval Debulking Surgery |
5 (12.5%) 35 (87.5%) |
59 (26.2%) 166 (73.8%) |
0.072 |
VTE, venous thromboembolism; WBC, white blood cell
Figure 2 shows daily median platelet count from postoperative days 0 to 15, for patients with and without venous thromboembolism. Both groups demonstrated a similar and continuous rise in platelet count. Thrombocytosis was present in 38/40 (95%) patients with venous thromboembolism versus 183/225 (81%) patients without (p=0.03). Among the 221 patients with thrombocytosis, median number of days with platelets >400K/mcL was higher in the venous thromboembolism (8 days, range 1-15) than the non-venous thromboembolism (6 days, range 1–16, p=0.049) groups.
Figure 2.
Median platelet count between postoperative day (POD) 0 and POD15 for patients with and without a VTE
A greater number of patients (n=31, 21.5%) developed venous thromboembolism prior to institution of extended prophylaxis in 2013, than afterwards (n=9, 7.4%) (p=0.002).
DISCUSSION
In our cohort, we demonstrated an initial leukocytosis in all patients undergoing splenectomy during ovarian cancer debulking surgery, with a peak on postoperative day 2. We also demonstrated a second peak in white blood cell count in women who subsequently developed a post-operative infection, peaking on postoperative day 10. Additionally, we found that women who developed a postoperative thromboembolism were more like to develop thrombocytosis, which also lasted longer than in women without a postoperative thromboembolism.
Most of what is known about post-splenectomy infections comes from trauma literature. In a cohort of 15,411 trauma patients undergoing laparotomy and splenectomy, there was a 40% risk of pneumonia [2]. In a sample of patients treated in the intensive care unit post-surgery—after adjustment for trauma and bowel resection—patients undergoing splenectomy were at significantly higher risk for infectious complication (48.5% versus 28.6%), specifically for intra-abdominal abscess (9.1%) [6]; the average day for diagnosis of infection was postoperative day 7 [5]. The rate of infectious morbidity after splenectomy during debulking surgery for ovarian cancer is reportedly 15-33%, occurring between postoperative days 3 and 10 [7, 13]. We identified initial leukocytosis in all patients after splenectomy, with a peak on postoperative day 2. However, patients with infection exhibited a second peak in white blood cell count on postoperative day 10, after a decrease of initial leukocytosis; this second peak was not present in patients without infection. This pattern has also been demonstrated in trauma patients [3]. We did not identify a difference in platelet count between patients with or without infection. Trauma literature has focused on using the platelet-to-white blood cell ratio to facilitate diagnosis of infection post-splenectomy, with infections more likely in patients with a ratio of <20 [5, 8, 9]. In our population, as there was no difference in platelets between the two groups, platelet-to-white blood cell ratio was driven solely by the white blood cell value and did not facilitate diagnosis of infection. It is possible that the hematologic changes in our patients were also driven by the malignancy, as well as non-surgical treatments such as chemotherapy, and supportive medications commonly including steroids. Several other laboratory tests can be used to aid in the diagnosis of infection, such as C-reactive protein and procalcitonin. Most of the available data come from patients undergoing gastrointestinal procedures, showing the utility of C-reactive protein in diagnosing of postoperative infection [14, 15]. However, the evidence for the utilization of these markers for post-ovarian cancer surgery is not robust, and neither of these markers are utilized universally at our institution.
In a prospective observational multicenter trial of trauma patients, those undergoing splenectomy had twice as many venous thromboembolisms (17.5% versus 7.8%, p=0.015), and significantly higher platelet counts starting on postoperative day 3 [11]. In a study of patients treated in the intensive care unit after surgery, those undergoing splenectomy had a significantly higher incidence of venous thromboembolism (29.7% versus 12.1%, p=0.005) [16]. On average, venous thromboembolisms occurred 9.7 days after splenectomy [16]. Patients undergoing splenectomy had a significantly higher postoperative maximum platelet count [16]. In patients with thrombocytosis, venous thromboembolism rates were higher after splenectomy; even in the absence of thrombocytosis, venous thromboembolism rates were higher after splenectomy [16]. A Danish population study of all patients undergoing splenectomy between 1996 and 2005 for a variety of reasons (trauma, hematologic disease, hematologic and solid cancers), showed that within 90 days of surgery, venous thromboembolism rates were elevated for all patients irrespective of indication [17]. The increased venous thromboembolism risk continued past 90 days, with a relative risk of 3.4 after one year compared with controls [17]. In a study of ovarian cancer patients undergoing splenectomy during either primary or secondary debulking surgery, the rate of venous thromboembolism was 8% within 30 days post-surgery [13].
We demonstrated a similar and progressive postoperative increase in platelet count in all patients after splenectomy, irrespective of development of venous thromboembolism. Patients developing venous thromboembolism had higher rates of thrombocytosis than those who did not. This data suggests that venous thromboembolism risk may be related to presence of thrombocytosis in these patients, rather than to absolute platelet count.
We found no difference in venous thromboembolism rates between patients undergoing primary versus interval debulking surgery; however, rates of thrombocytosis were higher after primary versus interval debulking. This difference is probably partially explained by administration of neoadjuvant chemotherapy before interval debulking surgery, as the commonly-used regimen of carboplatin and paclitaxel often leads to myelosuppression. Extent of surgery, which is generally less during interval debulking, may be associated with thrombocytosis as a function of acute phase protein. A study involving a larger cohort of interval debulking surgery patients would help determine whether lower rates of thrombocytosis result in lower rates of venous thromboembolism. There is also evidence that preoperative thrombocytosis in patients with ovarian cancer is an independent prognostic factor for worse survival [18], but more data are needed to evaluate if postoperative thrombocytosis has a similar effect.
Overwhelming post-splenectomy infection as a late complication was first described in 1951 [2, 4]. Administration of vaccines—ideally at least two weeks before splenectomy—is aimed at preventing this complication [4]. However, the majority of overwhelming post-splenectomy infections occur 2-3 years after splenectomy [6]. Given that patients with advanced stage ovarian cancer recur within the first 2-3 years and are commonly treated with chemotherapy for recurrence, the potential for overwhelming post-splenectomy infection should be considered. The liver serves to replace some of the functions lost following splenectomy, and it is possible that patients undergoing both splenectomy and partial liver resection during debulking surgery are more susceptible to postoperative infection. In our cohort, the rate of infection was the same regardless of liver resection (p=0.522). However, further research is needed to delineate this relationship.
Our study has several strengths. It is the largest report to date on patients undergoing splenectomy during either primary or interval debulking surgery for ovarian cancer, and includes detailed records of intraoperative procedures, pathology, postoperative blood tests, complications, clinical outcomes. Limitations include its retrospective nature; additionally, it is difficult to assess long-term complications of infection and venous thromboembolism in a population characterized by multiple rounds of myelosuppressive chemotherapy and the hypercoagulable state of malignancies.
We have demonstrated that splenectomy leads to initial leukocytosis in all patients during the first 2-3 postoperative days. A second white blood cell count rise occurred in patients diagnosed with infection, not seen in those who did not have infection. Thus, a decreased white blood cell count and then a second postoperative leukocytosis may be a sign of infection requiring intervention. We found that platelet-to-white blood cell ratio was not a reliable indicator of infection. All patients demonstrated a progressive postoperative increase in platelets after splenectomy, irrespective of venous thromboembolism development. Thrombocytosis was present more frequently and lasted longer in those who developed venous thromboembolism.
HIGHLIGHTS.
Immediate post-operative leukocytosis was present in all patients after a splenectomy
Patients who developed an infection developed a second peak in white blood cell count
Thrombocytosis was more common and of longer duration in patients who developed venous thromboembolism
Funding:
This study was funded in part through the NIH/NCI Support Grant P30 CA008748.
Footnotes
CONFLICT OF INTEREST STATEMENT: None of the authors declare conflicts of interest.
DISCLOSURES:
Dr. Chi reports personal fees from Bovie Medical Co., personal fees from Verthermia Inc. (now Apyx Medical Corp.), personal fees from C Surgeries, other from Intuitive Surgical, Inc., other from TransEnterix, Inc., personal fees from Biom ’Up, outside the submitted work.
Dr. Iasonos reports personal fees from Mylan, outside the submitted work.
Dr. Long Roche reports other* from Intuitive Surgical Inc., outside the submitted work. (*Airfare to a survivorship conference, where she spoke).
REFERENCES
- 1.Eisenkop SM, Spirtos NM, Lin W-CM. Splenectomy in the context of primary cytoreductive operations for advanced epithelial ovarian cancer. Gynecol Oncol 2006;100:344–8. doi: 10.1016/j.ygyno.2005.08.036. [DOI] [PubMed] [Google Scholar]
- 2.Fair KA, Connelly CR, Hart KD, Schreiber MA, Watters JM. Splenectomy is associated with higher infection and pneumonia rates among trauma laparotomy patients. Am J Surg 2017;213:856–61. doi: 10.1016/j.amjsurg.2017.04.001. [DOI] [PubMed] [Google Scholar]
- 3.Horowitz J, Leonard D, Smith J, Brotman S. Postsplenectomy leukocytosis: physiologic or an indicator of infection? Am Surgeon 1992;58:387–90. [PubMed] [Google Scholar]
- 4.Weledji EP. Benefits and risks of splenectomy. Int J Surg 2014;12:113–9. doi: 10.1016/j.ijsu.2013.11.017. [DOI] [PubMed] [Google Scholar]
- 5.Banerjee A, Kelly KA, Zhou HY, et al. Diagnosis of infection after splenectomy for trauma should be based on lack of platelets rather than while blood cell count. Surg Infect (Larchmt) 2014;15:221–6. doi: 10.1089/sur.2012.176. [DOI] [PubMed] [Google Scholar]
- 6.Barmparas G, Lamb AW, Lee D, et al. Postoperative infection risk after splenectomy: a prospective cohort study. Int J Surg 2015;17:10–4. doi: 10.1016/j.ijsu.2015.03.007. [DOI] [PubMed] [Google Scholar]
- 7.Bidus MA, Krivak TC, Howard R, et al. Hematologic changes after splenectomy for cytoreduction: implications for predicting infection and effects on chemotherapy. Int J Gynecol Cancer 2006;16:1957–62. doi: 10.1111/j.1525-1438.2006.00725.x. [DOI] [PubMed] [Google Scholar]
- 8.Weng J, Brown CV, Rhee P, et al. White blood cell and platelet counts can be used to differentiate between infection and in the normal response after splenectomy for trauma: prospective validation. J Trauma 2005;59:1076–80. doi: 10.1097/01.ta.0000189001.00670.d2. [DOI] [PubMed] [Google Scholar]
- 9.Toutouzas KG, Velmahos GC, Kaminski A, Chan L, Demetriades D. Leukocytosis after posttraumatic splenectomy: a physiologic even or sign of sepsis? Arch Surg 2002;137:924–8. doi: 10.1001/archsurg.137.8.924. [DOI] [PubMed] [Google Scholar]
- 10.Watters JM, Sambasivan CN, Zink K, et al. Splenectomy leads to a persistent hypercoagulable state after trauma. Am J Surg 2010;199:646–51. doi: 10.1016/j.amjsurg.2010.01.015. [DOI] [PubMed] [Google Scholar]
- 11.Pommerening MJ, Rahbar E, Minei K, et al. Splenectomy is associated with hypercoagulable thrombelastography values and increased risk of thromboembolism. Surgery 2015;158:618–26. doi: 10.1016/j.surg.2015.06.014. [DOI] [PubMed] [Google Scholar]
- 12.Weitz J, D’Angelica M, Gonen M, et al. Interaction of splenectomy and perioperative blood transfusions on prognosis of patients with proximal gastric and gastroesophageal junction cancer. J Clin Oncol 2003;21:4597–603. doi: 10.1200/JCO.2003.12.136. [DOI] [PubMed] [Google Scholar]
- 13.Magtibay PM, Adams PB, Silverman MB, Cha SS, Podratz KC. Splenectomy as part of cytoreductive surgery in ovarian cancer. Gynecol Oncol 2006;102:369–74. doi: 10.1016/j.ygyno.2006.03.028. [DOI] [PubMed] [Google Scholar]
- 14.Facy O, Paquette B, Orry D, et al. Diagnostic accuracy of inflammatory markers as early predictors of infection after elective colorectal surgery. Ann Surg 2016;263:961–6. [DOI] [PubMed] [Google Scholar]
- 15.Adamina M, Steffen T, Tarantino I, Beutner U, Schmied BM, Warschkow R. Meta-analysis of the predictive value of C-reactive protein for infectious complications in abdominal surgery. Br J Surg 2015;102:590–8. [DOI] [PubMed] [Google Scholar]
- 16.Lee DH, Barmparas G, Fierro N, et al. Splenectomy is associated with a higher risk for venous thromboembolism: a prospective cohort study. Int J Surg 2015;24(Pt A):27–32. doi: 10.1016/j.ijsu.2015.10.011. [DOI] [PubMed] [Google Scholar]
- 17.Thomsen RW, Schoonen WM, Farkas DK, Riis A, Fryzek JP, Sørensen HT. Risk of venous thromboembolism in splenectomized patients compared with the general population and appendectomized patients: a 10-year nationwide cohort study. J Thromb Haemost 2010;8:1413–6. doi: 10.1111/j.1538-7836.2010.03849.x. [DOI] [PubMed] [Google Scholar]
- 18.Ye Q, Cheng J, Ye M, Lui D, Zhang Y. Association of pretreatment thrombocytosis with progression in ovarian cancer: a systemic review and meta-analysis. J Gynecol Oncol 2019;30:e5. [DOI] [PMC free article] [PubMed] [Google Scholar]