Preoperative and Postoperative Complete Blood Cell Counts and Prediction of Surgical Site Infection After Cesarean Delivery

A Retrospective Case-Control Study

Cesarean delivery (CD) is the most frequently performed surgery worldwide.¹ In some countries, almost half of all births are by CD.² Surgical site infection (SSI) is a common complication after CD. The incidence of SSI after CD varies between 1% and 23%.³ Postoperative SSI increases morbidity, prolongs hospital stay, and significantly reduces the quality of life of patients. For this reason, it is important to predict and prevent the development of SSI after CD as early as possible.

In the literature, there are hematologic parameters associated with inflammation that are not as widely used as white blood cell count (WBC), C-reactive protein, or erythrocyte sedimentation rate but have attracted attention in recent years. These parameters include neutrophil-to-lymphocyte ratio (NLR), platelet-lymphocyte ratio (PLR), mean platelet volume (MPV), and platelet-large cell ratio (P-LCR). It has been shown that NLR values ​​are highly correlated with the prognosis of inflammatory diseases.⁴ Similarly, PLR is thought to be associated with many clinical conditions accompanied by inflammation, such as rheumatological diseases, sepsis, and cancer.^5,6 The relationship between inflammation-related diseases and other parameters of the hemogram such as MPV was also examined, and significant results were obtained.⁷ Finally, the role of P-LCR in inflammation has been demonstrated among patients with septic shock.⁸

It is known that neutrophils, monocytes, and platelets have important roles in normal wound healing.⁹ In the first stage of wound healing, alpha granules are released by the activated platelets.¹⁰ The inflammatory process begins with a large number of cytokines released from the wound. Monocytes and macrophages also migrate to the wound area to prevent the development of infection.¹¹ With the thought that these blood cells may also play a role in abnormal wound healing processes such as wound infection, the relationship between hematologic parameters and SSI has been investigated in some studies.^12,13 In a recent study, the role of NLR, PLR, and MPV in predicting SSI after CD was examined, and NLR was found as a potential marker.¹⁴ This study is still the only one investigating possible hematologic predictors of SSI after CD. To the best of the authors’ knowledge, there is no study examining the role of P-LCR in the prediction of SSI. Therefore, the current study aimed to evaluate the role of preoperative and postoperative hematologic parameters in predicting the development of SSI after CD.

This retrospective case–control study was conducted with the participation of patients who were treated at the authors’ tertiary hospital between June 2020 and June 2021. Ethics committee approval was obtained (approval number: 2021.09.197). Due to the legal regulations in Turkey and the local ethics committee policy in the hospital, no informed consent is required from the patients in retrospective studies.

The SSI group consisted of patients who were hospitalized because of SSI after CD. The control group was composed of healthy postpartum women who delivered by CD in our hospital and were admitted to the outpatient postpartum care unit for routine evaluation 6 weeks after delivery. Exclusion criteria were long operation time (>60 minutes), staplers used to close the incision, indication for emergency CD, extensive blood loss, a history of 2 or more previous CD, corticosteroid use, presence of diabetes mellitus, infection (eg, chorioamnionitis), or diseases that may affect complete blood count parameters (eg, rheumatologic, renal, hepatic, or immune diseases).

Demographic characteristics of patients such as age, gravidity, parity, total number of previous CD, and weight were recorded. Hemoglobin (Hb), WBC, neutrophil count, lymphocyte count, platelet count (PLT), MPV, P-LCR, NLR, and PLR, which are the parameters of the complete blood count, were recorded before CD and at 24 hours postoperatively. NLR were calculated by dividing the neutrophil count by the lymphocyte count. PLR was determined as the ratio of the platelet count to the lymphocyte count.

The categorical variable of previous CD was presented as frequency and percentage. Distributions of continuous variables were checked by Shapiro-Wilk test. Age, weight, Hb, MPV, and P-LCR levels were normally distributed while other parameters were not normally distributed. Student t test and paired samples t test were used to compare the normally distributed parameters among SSI and control groups and the results were given as mean ± standard deviation. Mann–Whitney U and Wilcoxon signed rank tests were used to compare the non-normally distributed parameters among SSI and control groups, and the results were given as median (25th–75th percentile). Area under the curve (AUC) of significant parameters were analyzed by receiver operating characteristic. All statistical analyses were performed using SPSS v. 17 (IBM). P < .05 was considered statistically significant.

A total of 48 patients who were hospitalized because of SSI after CD comprised the SSI group. The control group (n = 45) consisted of healthy postpartum women who delivered by CD in our hospital and were admitted to the outpatient postpartum care unit for routine evaluation 6 weeks after delivery. Demographic, clinical, and obstetric characteristics of the participants are shown in Table 1. In the current study, inclusion and exclusion criteria were determined by considering several of these risk factors, and there was no difference between the SSI and control groups in terms of age, gravidity, parity, the number of previous CDs, and weight. As a result, the groups were homogeneous in terms of those factors.

There were no differences between SSI and control groups in any of the hematologic parameters in both the preoperative and postoperative periods (Table 2).

WBC, neutrophil count, and NLR levels were increased while Hb, PLT, and lymphocyte count levels were decreased after CD in both post SSI and control groups (11.4 vs 12.7, P = .002; 8.1 vs 10.0, P < .001; 3.6 vs 6.6, P < .001; 11.8 vs 9.9, P < .001; 221 vs 190, P < .001; and 2.1 vs 1.6, P < .001, respectively in the SSI group; 10.9 vs 12.4, P < .001; 7.9 vs 9.7, P < .001; 3.8 vs 5.6, P < .001; 11.5 vs 10.1, P < .001; 209 vs 186, P < .001; and 2.0 vs 1.7, P = .003, respectively in the control group). There were no differences in P-LCR and MPV levels after CD in both groups (36 vs 35, P = .42 and 11.4 vs 11.4, P = .73, respectively in the SSI group; 35 vs 35, P = .97 and 11.2 vs 11.2, P = .99, respectively in the control group).

As shown in the Figure, the PLR was increased after CD in the SSI group but remained stable in the control group (P = .007 and P = .51, respectively). Unfortunately, the difference in the PLR was unable to significantly predict SSI after CD (AUC = 0.60; P = .09).

The prediction of SSI after CD is important for many reasons. SSI often develops within the first week after CD, and there are risk factors such as maternal age, obesity, number of previous CD, presence of related maternal disease (diabetes mellitus or immune diseases), corticosteroid use, poor antenatal care, emergency delivery, duration of operation, stapler use for incision closure, and heavy bleeding during surgery.^14,15

A complete blood count is performed for almost all patients perioperatively. Therefore, the idea that the parameters evaluated in the complete blood count can be used in prediction, diagnosis, or follow-up of a clinical condition could be a cost-effective approach. Especially in the past 20 years, the relationship between these hematologic markers and many diseases has been examined. In the literature, the relationship between inflammation-related complete blood count parameters such as WBC, neutrophil count, lymphocyte count, MPV, P-LCR, NLR, and PLR and many diseases has been examined.^4-8

It has been shown that blood cells such as neutrophils, monocytes, and platelets have a role in wound healing.⁹ Because SSI is a clinical condition in which healing is impaired as a result of an infection, blood cells will have an important effect on the development of SSI, both in terms of inflammation and wound healing. However, the number of studies examining the relationship between preoperative and postoperative inflammatory hematologic markers and SSI is limited. In a study directly examining the value of preoperative hematologic parameters to predict wound healing failure in 2017, lower levels of preoperative neutrophil count ratio, NLR, and PLR were found to be useful in head and neck reconstruction cases.¹⁶ In another study conducted with patients who had also undergone head and neck cancer surgery, preoperative NLR was found to be significant in predicting SSI.¹³ However, in this study, higher level of NLR was found to be associated with SSI. Therefore, the results of these 2 studies were inconsistent.

Considering the studies examining the relationship between postoperative hematologic markers and SSI, lymphocyte count ratio and NLR were found to be significant predictors of SSI after spinal decompression surgery.¹² In a study by Zhuo et al,¹⁷ high preoperative NLR and PLR were found as predictors of SSI after surgery for groin hernia.

In light of these studies, it is thought that there may be a relationship between the postoperative NLR level and SSI, but the evidence is not clear. In a study examining the role of hematologic markers in predicting SSI after CD, Rotem et al¹⁴ found that NLR and PLR ​​measured within the first 24 hours postoperatively were statistically significantly higher in the SSI group. In this study, AUC for NLR was calculated as 0.670 (P = .01), while PLR was found to be statistically insignificant in the receiver operating characteristic analysis.

Because the results of studies on the value of hematologic parameters in terms of SSI prediction in the literature are conflicting and there is only 1 study on SSI after CD, we aimed to close this gap with the current study. We evaluated all parameters both preoperatively and postoperatively and found no differences between the SSI and control groups in any of the hematologic parameters. The only statistically significant finding was increased PLR in the SSI group, but this difference was not useful to predict SSI.

The fact that this is, to the best of the authors’ knowledge, only the second study in the literature on SSI after CD limits comparability of the results. The small number of patients and the retrospective design are also limitations.

This retrospective case–control study showed that neither preoperative nor postoperative hematologic parameters could predict SSI after CD. The PLR was increased significantly after CD in the SSI group, but the difference could not predict SSI. Prospective studies with larger populations are needed.

Fatih Aktoz, MD¹; Can Tercan, MD¹; Busra Cigdem, MD¹; and Eren Vurgun, MD²

¹Department of Obstetrics and Gynecology, Basaksehir Cam and Sakura City Hospital, Istanbul, Turkey
²Department of Medical Biochemistry, Prof. Dr. Cemil Tascioglu City Hospital, Istanbul, Turkey

Address all correspondence to: Fatih Aktoz, MD, Department of Obstetrics and Gynecology, Basaksehir Cam and Sakura City Hospital, Istanbul, Turkey; tel: +905321773290; email: fatihaktoz@gmail.com