Impact of postoperative infection on changes in leucocyte levels in early postoperative period in patients undergoing splenectomy during cytoreductive surgery for gynaecological malignancy

Abstract

Background

In cytoreductive surgery for gynecological cancers, 13-25% of patients require splenectomy. Therefore, hematological parameters change in the post-splenectomy period, especially leucocytosis and thrombocytosis. In this study, we aimed to evaluate the changing of leucocyte and neutrophil between the groups with and without infection in the early postoperative period in patients who underwent splenectomy during cytoreductive surgery for gynecological cancer.

Methods

This retrospective study included 96 patients who underwent splenectomy during cytoreductive surgery for gynecological malignancies. Leukocyte levels, CRP, procalcitonin, and platelet counts were recorded daily during the first five postoperative days. Postoperative infections were identified based on clinical and laboratory findings.

Results

Postoperative infection was observed in 23 patients (24.0%). On postoperative day 4, the mean leukocyte count was 13.2±4.5 ×10³/µL in infected patients vs. 9.8±3.2 ×10³/µL in non-infected patients (p=0.01). By day 5, leukocyte levels remained significantly elevated in the infection group (p<0.05).

Conclusion

Leukocyte trends, especially between days 4 and 5, may serve as a practical marker for early postoperative infection in patients undergoing splenectomy during cytoreductive surgery.

Introduction

One of the most important prognostic factors in gynecological malignancies, especially in advanced stage ovarian cancer, is the residual tumor after cytoreductive surgery. A significant increase in survival has been demonstrated when cytoreduction is achieved to R1 (residual tumor ≤ 1 cm) or R0 (no macroscopic tumor) cytoreduction at the operation [1–3]. Therefore, resection of additional visceral organs may be required to achieve R1 cytoreduction. In advanced gynecologic cancers, splenectomy may be necessary due to tumor invasion into the parenchymal, perisplenic, or hilar region, or because of iatrogenic splenic injury during surgery. Additionally, it is known that in patients with upper abdominal involvement, splenectomy is performed together with distal pancreatectomy to achieve R0 resection to contribute to the survival of the patients [4]. In cytoreductive surgery for gynecological cancers, 13–25% of patients require splenectomy [5, 6]. Patients who underwent splenectomy for cytoreduction may cause increased morbidity such as acute respiratory distress syndrome, pleural effusion, pancreatic injury, hematologic complications, pulmonary embolism and development of pancreatic pseudocyst [7]. The overall mortality rates after splenectomy have been reported as 1.4% [8].

The spleen is a hematopoietic organ and is involved in the development and regulation of hematologic systems inculding erythroid, myeloid, megakaryocytic, lymphoid, reticuloendothelial, cellular and humoral immunity [9]. As a result, hematological parameters often change in the post-splenectomy period, especially leucocytosis and thrombocytosis. In addition, the spleen is one of the most important peripheral immune organs, playing a crucial role as a secondary lymphoid organ, particularly in providing immunity against encapsulated bacteria [10]. The spleen filters bloodborne pathogens through the reticuloendothelial system, removes aged erythrocytes, and serves as a reservoir for lymphocytes. It facilitates humoral and cellular immunity by promoting antigen presentation and antibody production, particularly against Streptococcus pneumoniae, Haemophilus influenzae, and Neisseria meningitidis, which are all encapsulated organisms. Hence, splenectomy is a known risk factor for infection, with infection rates reported at 3.2% after the procedure [5, 8].

The laboratory tests are routinely used in the diagnosis of postoperative infection, in addition to clinical symptoms. Considering leucocytosis as a secondary physiological response to splenectomy, it may be challenging to diagnose postoperative infection accurately. Additionally, defining the infection becomes more difficult in patients who underwent interval cytoreductive surgery after neoadjuvant chemotherapy. Few studies have investigated early postoperative hematological changes among patients who underwent splenectomy as part of cytoreductive surgery for gynecological malignancy. The study conducted by Bidus et al.. demonstrated that in patients who developed an infection following splenectomy, there was a more significant increase in leukocyte count after the 5th postoperative day compared to those who did not develop an infection. However, the differences in leucocyte and neutrophil changes between infected and non-infected patients in the postoperative period were not statistically significant [9].

In this study, we aimed to evaluate changes in leucocyte and neutrophil between patients with and those without infection in the early postoperative period among patients who underwent splenectomy during cytoreductive surgery for gynecological cancer.

Material and method

The data of patients who underwent splenectomy during cytoreduction for gynecological cancer between January 2010 and June 2022 were reviewed across two gynecologic oncology centers. Patients with immunosuppressive disease, disease requiring immunosuppressive therapy, granulocyte colony-stimulating factor (G-CSF) and patients with an abscess or other systemic infection in the preoperative period were excluded from the study. Ethical committee approval was obtained (No: 5 Date: 12.07.2021). A total of 96 patients were included in the study. Leucocyte, neutrophil, lymphocyte, platelet, hemoglobin, c-reactive protein (CRP) and procalcitonin values were recorded preoperatively and on the first 7 days postoperatively. Patients were categorized based on whether they received neoadjuvant therapy or not. Ceftriaxone ± metronidazole prophylaxis was administered to all patients except 3 patients. Gentamicin and clindamycin were administered to 2 patients with penicillin allergy and ciprofloxacin prophylaxis was applied to 1 patient. In both clinics, antibiotic prophylaxis was administered 30–60 min before the surgical incision. Metronidazole was also added to prophylaxis in patients with possible gastrointestinal tract resection. To protect against encapsulated bacterial infection after splenectomy, patients were administered pneumococcal, haemophilus influenzae type b and meningococcal vaccines within two weeks postoperatively. The presence of a postoperative infection was defined by an infectious diseases specialist according to fever of 38 °C and above, growth in culture, high procalcitonin and CRP levels and the patient’s clinical condition. Leukocyte count, CRP, and procalcitonin were selected as they are widely used markers for infection and systemic inflammation in the early postoperative period, particularly in abdominal and oncologic surgeries. In cases where infection was detected, antibiotic changes were made by infectious disease doctors.

Postoperative infections were identified based on clinical symptoms (e.g. fever, surgical site findings), laboratory indicators (e.g. leucocyte count, CRP, procalcitonin) and physician-documented diagnoses in the medical record. In cases with available microbiological culture data, infection types were classified accordingly. These included wound infections, intra-abdominal abscesses, bloodstream infections, and fungal infections.

Statistical analysis

Two groups, those who did not develop postoperative infection and those that did, were compared in terms of demographic, clinical, surgical, pathological and haematological parameters. The data were presented utilizing descriptive statistics, expressing continuous variables as mean ± standard deviation (SD) and median (minimum-maximum), while categorical variables were represented as frequencies and percentages. The statistical analysis conducted in this study encompassed an evaluation of various leucocyte series and history of erythrocyte transfusion, in both infected and non-infected patients. Normal distribution assumptions for each variable were confirmed using the Shapiro-Wilk test. Variables meeting the criteria for a normal distribution were subjected to an independent samples Student’s t-test for group comparison, while those failing to meet normality assumptions underwent analysis using the Mann-Whitney U test. To compare clinical and demographic findings between infected and non-infected patients, either the Pearson Chi-square test or Fisher’s exact test was employed as appropriate. Statistical significance was set at a value of p < 0.05. The statistical analysis was conducted using SPSS 11.5 for Windows (SPSS Inc., Chicago, IL, USA).

Results

The median age of the 96 patients was 57 years (range, 34–89 years). Diagnoses included ovarian epithelial carcinoma in 81 (84.3%) patients, endometrial adenocarcinoma in 14 (14.6%) patients and uterine leiomyosarcoma in 1 (1%) patient. Sixteen (16.7%) of the 96 patients received neoadjuvant chemotherapy. Forty-one (42.7%) patients had at least one comorbidity. Distal pancreatectomy was performed concurrently with splenectomy in 5 (5.2%) patients. Erythrocyte transfusion was administered to 3 (3.1%) patients preoperatively, 36 (37.5%) patients intraoperatively and 44 (45.9%) postoperatively. Postoperative infection was diagnosed in 23 (24%) patients. The clinical and demographic features of the patients were detailed in Table 1. In these patients, infection developed on the median 7th postoperative day (range, 2-10th day).

Table 1.

Clinical-demographical features of entire cohort

Parameter		n (%)
Neoadjuvant chemotherapy	Not received	79 (82.3)
	Received	16 (16.7)
	Not reported	1 (1.0)
Tumor diagnosis	Ovarian epithelial carcinoma	81 (84.3)
	Endometrial adenocarcinoma	14 (14.6)
	Uterine leiomyosarcoma	1 (1.0)
Comorbidity	Absent	55 (57.3)
	Present	41 (42.7)
Preoperative erythrocyte transfusion	Not received	76 (79.2)
	Received	3 (3.1)
	Not reported	17 (17.7)
Intraoperative erythrocyte transfusion	Not received	13 (13.5)
	Received	36 (37.5)
	Not reported	47 (49)
Postoperative erythrocyte transfusion	Not received	14 (14.5)
	Received	44 (45.9)
	Not reported	38 (39.6)
Postoperative infection	Negative	73 (76.0)
	Positive	23 (24.0)

There was no statistically significant difference in neoadjuvant chemotherapy, tumor diagnosis and presence of comorbidity among patients with or without postoperative infections (Table 2). Furthermore, the rates and volumes of preoperative, intraoperative, and postoperative erythrocyte transfusions were statistically similar between patients with and without postoperative infections, with no significant differences observed (p > 0.05) (Table 2 and Table 3).

Table 2.

Clinical and demographical findings in patients with or without postoperative infection

Parameter		Postoperative infection		p Value
		Positive	Negative
		n (%)	n (%)
Neoadjuvant chemotherapy	Not received	19 (82.6)	60 (83.3)	1.000
	Received	4 (17.4)	12 (16.7)
Tumor diagnosis	Ovarian epithelial carcinoma	19 (82.6)	62 (84.9)	0.850
	Endometrial adenocarcinoma	4 (17.4)	10 (13.7)
	Uterine leiomyosarcoma	0 (0.0)	1 (1.4)
Comorbidity	Absent	11 (47.8)	44 (60.3)	0.293
	Present	12 (52.2)	29 (39.7)
Preoperative erythrocyte transfusion	Not received	17 (94.4)	59 (96.7)	0.545
	Received	1 (5.6)	2 (3.3)
Intraoperative erythrocyte transfusion	Not received	3 (33.3)	10 (25.0)	0.683
	Received	6 (66.7)	30 (75.0)
Postoperative erythrocyte transfusion	Not received	2 (11.8)	12 (29.3)	0.195
	Received	15 (88.2)	29 (70.7)

Table 3.

The comparison of history of erythrocyte transfusion in patients with or without postoperative infection

Time of erythrocyte transfusion	Infection		No infection		Total		p Value
	Mean ± SD	median (min-max)	Mean ± SD	median (min-max)	Mean ± SD	median (min-max)
Preoperative	2 ± NC	2.00 (2.00–2.00)	1.50 ± 0.71	1.50 (1.00–2.00)	1.67 ± 0.58	2.00 (1.00–2.00)	0.667
Intraoperative	2.83 ± 0.98	2.50 (2.00–4.00)	2.43 ± 1.07	2.00 (1.00–6.00)	2.50 ± 1.06	2.00 (1.00–6.00)	0.349
Postoperative	2.53 ± 0.99	2.00 (1.00–4.00)	2.21 ± 0.73	2.00 (1.00–4.00)	2.32 ± 0.83	2.00 (1.00–4.00)	0.274

SD Standard Deviation, NC Not calculated because of patient number (n=1)

Among the 23 patients who developed postoperative infections, the infection types were classified based on microbiological or clinical documentation. The most frequent infections included wound or incision site infections (n = 6), intraabdominal abscesses (n = 3), gram-positive bacterial infections such as Staphylococcus or Streptococcus (n = 4), and gram-negative infections including E. coli, Klebsiella, and Pseudomonas (n = 4). Fungal infections (Candida species) were identified in 2 patients, while one patient had COVID-19. Additionally, 3 patients were diagnosed with clinical sepsis without microbiological confirmation but received infection-targeted treatment based on clinical and laboratory findings. Of the 23 patients diagnosed with postoperative infection, 9 also presented with major surgical complications including anastomotic leakage, intra-abdominal abscess formation, and reoperation.

The change in leucocyte series due to infection

A comparison of splenectomy patients with and without infection showed a similar trend in postoperative leucocyte and neutrophil counts. Until the 4th postoperative day, leucocyte levels in both groups were above normal values. However, from the 4th to the 5th day, leucocyte and neutrophil levels increased in the group with infection, while they decreased in the group without infection (Fig. 1 A and 1B). This change in leucocyte values between the 4th and 5th days was statistically significant (p = 0.030). In infected patients, the leukocyte count change was 822.86 (95% CI: − 535.44 to 2181.16), whereas it was − 1266.80 (95% CI: − 2104.39 to − 429.21) in non-infected patients. However, neutrophil count changes between the 4th and 5th days were borderline statistically significant (p = 0.079) (Table 4). For neutrophils, the change was 362.14 (95% CI: − 1041.50 to 1765.78) in infected patients and − 1727.60 (95% CI: − 2563.41 to − 891.79) in non-infected patients. These changes in the 4th and 5th days were highlight a critical period for monitoring potential infections.

Fig. 1.

Postoperative changes in leucocyte (A) and neutrophil (B) levels in patients with and without infection

Table 4.

Change from 4th day to 5th (5th days minus 4th days) in patient with or without postoperative infection

Parameter	Postoperative infection positive		Postoperative infection negative		p Value
	Mean ± SD	Median (min-max)	Mean ± SD	Median (min-max)
Leucocyte count changes between days of from 4th to 5th	822.86 ± 3141.07	1160 (5140–7410)	−1266.80 ± 3589.92	−850 (−6800-6860)	0.030
Neutrophil count changes between days of from 4th to 5th	362.14 ± 3245.91	1015 (−7800-4750)	−1727.60 ± 3582.29	−1500 (−7070-6220)	0.079

SDStandard Deviation

In the comparative evaluation of inflammatory biomarkers over time, procalcitonin levels in the infection group peaked on postoperative day 2 (mean 59.29 ng/mL), then gradually declined, while remaining consistently below 1.0 ng/mL in the non-infection group throughout the early postoperative period. Similarly, CRP levels in the infection group were elevated in the early postoperatively days, with a peak observed around the 3rd postoperative day, whereas the non-infection group maintained lower and more stable CRP levels. These temporal differences were statistically significant for both PCT (p = 0.002) and CRP (p = 0.001). No significant difference was detected in platelet trends between the two groups (p = 0.184).

Discussion

The spleen is a hematopoietic organ that has immunological and hematological functions [11]. A normally functioning spleen removes senescent red blood cells, bacteria and other particles from the circulation, as well as aiding in the retention and circulation of platelets [11]. Therefore, it is well known that splenectomy increases infection rates [12, 13]. The incidence of infection and overall mortality were reported as 3.2% and 1.3% during 7 years median follow-up period for splenectomized patients, respectively [8]. However, the reason for splenectomy for included patients was hematological diseases in most of those studies. The number of patients who underwent splenectomy due to malignancy was notably low.

Banerjee et al.. evaluated post-splenectomy leucocyte changes in patients following trauma, finding an infection rate of 40% but no statistical difference in daily maximal leucocyte values between infected and non-infected groups over two weeks post-splenectomy [14]. Toutouzas et al.. reported that leucocyte > 15 × 10³µL on postoperative day 5th were independent risk factors for infection following splenectomy due to trauma [15]. Weng et al.. reported that the leucocyte counts were not different until postoperative 5th days. Leucocyte levels were significantly higher than 15 × 10³µL and persisted until postoperative day 10th in patients with infection after postoperative day 5th compared to patients without infection [16]. In the study conducted by Bidus et al.. on gynecological cancers, it was shown that the changes in leucocyte counts between the infected and non-infected groups were similar within the first 4 days. However, after the 4th postoperative day, an increase in leucocyte count was observed in the infected group compared to the non-infected group. Nevertheless, when the entire early postoperative period was examined, this change in leucocyte count between the two groups was found to be not statistically significant [9]. In our study, in patients with splenectomy because of the gynecologic malignancy, distinction of leucocyte between the 4th and the 5th days was significantly different for patients with and without postoperative infection. In a similar to other studies [9, 16], there was no significant difference in leucocyte values between patients with and without infection in the first 4 postoperative days. According to our results, receiving neoadjuvant chemotherapy had no effect on leucocyte changes between infected and non-infected patients.

The restrospective nature is a limitation of our study. This study, which includes 96 patients, is one of the very few studies examining hematological changes following splenectomy in gynecological cancers. Additionally, it represents the largest cohort in the literature regarding the number of patients, demonstrating that changes in leucocyte and neutrophil counts are indicative of postoperative infection during this process.

In the comparative evaluation of inflammatory biomarkers, a day-to-day trend analysis revealed that CRP and procalcitonin levels were consistently higher in the infection group throughout the early postoperative period. Procalcitonin levels peaked on postoperative day 2 (mean 59.29 ng/mL) in the infection group, whereas they remained below 1.0 ng/mL in the non-infection group at all time points. Similarly, CRP values showed a peak around postoperative day 3 in patients with infection, followed by a gradual decrease, while CRP levels remained relatively low and stable in the non-infection group. These differences were statistically significant for both CRP (p = 0.001) and procalcitonin (p = 0.002). Platelet counts did not significantly differ between groups (p = 0.184). This temporal divergence in procalcitonin and CRP levels underscores the potential of these markers in early differentiation of postoperative infection in splenectomized gynecologic oncology patients.

Although infection was documented in all cases based on clinical and laboratory data, comprehensive microbiological culture results were not uniformly available due to the retrospective nature of the study. This hindered detailed organism-level classification in some patients.

Conclusion

Our study showed that within the first 4 days following splenectomy, there were no significant differences between the infected and non-infected groups. However, the daily increase in leucocyte values from the 4th to the 5th days is considered significant in favor of infection. These findings suggest that serial monitoring of leucocyte trends, particularly on postoperative days 4 and 5, may aid in the early detection of infection in high-risk gynecologic oncology patients undergoing splenectomy.

Monitoring hematological parameters could thus be vital in the early identification and management of postoperative infections in patients undergoing cytoreductive surgery for gynecological malignancies. Larger prospective studies are needed to establish clear cut-off values for post-splenectomy hematological parameters in patients with gynecological malignancies, which is critical for reducing postoperative infection rates.

Authors’ contributions

Yesım Ozkaya Ucar: Conceptualization, Methodology, Formal analysis and investigation, Writing - original draft preparation, Writing - review and editing, Resources, and Supervision.Tufan Arslanca: Formal analysis and investigation, Writing - original draft preparationOkan Aytekın: Formal analysis and investigation, Writing - original draft preparationGulsah Tıryakı Guner: Formal analysis and investigation, Writing - original draft preparationSerpil Ünlü: Formal analysis and investigation, ResourcesGürsu Kimyon Comert: Formal analysis and investigation, Writing - review and editingFatıh Kılıc: Formal analysis and investigation, Writing - review and editingAziz Ahmet Türel: Formal analysis and investigation, Resources, and SupervisionTaner Turan: Conceptualization, Writing - original draft preparation, Writing - review and editing, Resources, and Supervision.

Funding

None.

Data availability

No datasets were generated or analysed during the current study.

Declarations

Ethics approval and consent to participate

The Ankara Bilkent City Hospital ethical committee evaluated and approved the research procedure in accordance with the Declaration of Helsinki (Approval date: 12/07/2021, Approval No: 5). All participants provided informed consent to participate.

Consent to for publication

Not applicable.

Competing interests

The authors declare no competing interests.