Background
The current data available in the literature demonstrate many direct benefits of laparoscopic surgery for gynecologic pathologies when compared to open surgery [
1‐
3]. The known rate of intraoperative and postoperative major complications during gynecologic laparoscopic surgery was 0.7–4% after surgery for benign pathologies [
4‐
6] and 4–21% after surgery for malignant diseases [
7‐
9]. Complications can be identified during laparoscopic surgery, or can develop immediately after, and even considerably after discharge; they lead to extended recovery periods, re-hospitalization, and increased costs. If not diagnosed in time, these complications can cause high morbidity or even mortality; therefore, finding safe methods capable of detecting complications early is important.
Normally, surgery causes trauma to the body, activating the coagulatory and immune responses; complications greatly increases the activity of these systems. These two processes have mostly been studied as functioning independently of each other; however, numerous researches have determined several checkpoints that affect both the hemostatic and the immune response/inflammatory cascades. In fact, coagulation and inflammation are activated by the same types of injuries, usually with a precise temporal correlation. During surgery, when tissues are damaged, macrophages are activated and various monokines are released, mainly tumor necrosis factor α (TNF-α), interleukin 6 (IL-6), and IL-1β [
10], which in turn promote acute-phase protein synthesis in the liver [
11,
12]. C-reactive protein, serum amyloid A, haptoglobin, hepcidin, α1-antitrypsin, α1-acid glycoprotein, and fibrinogen are the main acute-phase reactants [
13]. In case of complications, the organ damage is greater resulting in increased serum levels of such inflammation markers. These increases may be sufficiently sensitive to allow early diagnosis of the same complications.
Among these proteins, fibrinogen, in particular, could be an accurate indicator of an adverse post-operative course, including surgical and non-surgical complications. In fact, as a coagulation factor, it plays important biological roles not only in hemostasis but also in tissue repair and inflammatory responses in several pathologic conditions [
14]. Therefore, the circulating levels of fibrinogen may increase several-fold after the operative trauma, and these levels are greater the more intense the inflammatory response [
15]. To date, the ability of fibrinogen to predict postoperative surgical complication has been shown only retrospectively and in small and very selected populations [
16].
The aim of this research was to evaluate prospectively the predictive properties of fibrinogen for diagnosing complications after laparoscopic gynecologic surgery for benign and malignant conditions. Timely diagnosis and early treatment of these complications will help clinicians to avoid life-threatening situations.
Patients and methods
We enrolled prospectively 1016 patients who underwent elective laparoscopic gynecologic surgery between June 2013 and December 2017 at the Department of Gynecologic Oncology, A. Businco Hospital, Azienda Ospedaliera Brotzu, Cagliari. This study was undertaken with the approval of the Local Institutional Ethics Committee. Enrolled patients gave written informed consent for the surgical procedures and for study participation.
Post-operative complications were defined as those occurring post-operatively up to 8 weeks after surgery. The complications were graded according to the Clavien-Dindo classification, based on the severity and intervention required [
17,
18]. Severity grading increases from one to five: complications of grade III or more were defined as major complications because they required surgical, endoscopic, or radiologic intervention [
18]. In cases wherein more than one complication occurred, the highest-grade complication was used for analysis.
The complications were assessed prospectively. The detection and classification of complications were made by physicians blinded to the laboratory results (i.e., fibrinogen and WBC levels).
We included and graded the following post-operative complications categories: -) infection confirmed by culture or fever (body temperature > 38.0 °C) after 48 h post op requiring antibiotics; −) vascular injury; −) hemorrage: primary hemorrage (i.e., hemorrhage within 24 h from surgery) and secondary hemorrage (haemorrage after 24 h form surgery); −) intestinal complications including intestinal injuries, and peritonitis (postoperative ileus requiring NG tube/total parental nutrition, bowel obstruction, bowel perforation, others such as constipation, diarrhea, fecal incontinence/urgency; −) anastomotic leak of small or large bowel; −) urologic complications such as bladder and ureteral injuries, urinary obstruction, incontinence/urge; −) vaginal vault complications such as infection, bleeding, and vaginal cuff dehiscence; −) fistula (enterocutaneous, enterovaginal, vesicovaginal, ureterovaginal, others); −) wound complications: infection and wound breakdown (deep or superficial); −) pelvic or abdominal abscess/hematoma; −) surgical hernia; −) lymphocyst/lymphoedema; −) thromboembolic disorders (deep venous thrombosis and pulmonary embolia); −) cardiac complications (i.e., atrial fibrillation, myocardial infarction, cardiac failure, and other cardiac problems); −) respiratory complications (i.e., pulmonary oedema, pneumothorax, atelectasia, pleural effusion, and other respiratory problems excluding pneumonia); −) neurological complications: nerve injury with associated neuropathic pain/paraesthesia/nerve palsy; −) psychiatric complications (postoperative delirium, psychosis, depression, and other).
To detect the occurrence of complications, a daily clinical assessment was performed postoperatively and additional examinations (i.e., contrast-enhanced computed tomography or magnetic resonance, exploratory laparoscopy) were performed as indicated clinically. After discharge, the patients were checked by phone call every 2 days and visited after 1 week and then every 4 weeks, or sooner in the presence of symptoms, for 2 months after surgery.
The descriptive analysis of the enrolled population also included some demographic and anthropometric data (age, body mass index), parity, indication to surgery, previous abdominal surgery, as well as operative data (i.e., operative time, blood loss, length of hospital stay, time taken to achieve well-being).
Assessment of circulating levels of fibrinogen and inflammatory parameters
Fibrinogen level was determined along with the other coagulation parameters (prothrombin, partial thromboplastin time, d-dimer) preoperatively, on the first post-operative day and at the appearance of symptoms indicative of an irregular post-operative course or at the time of re-hospitalization for persistent symptoms such as pain and/or bleeding. Additionally, white blood cell (WBC) count was assessed at the same time intervals. The changes (% variation) of fibrinogen and WBC between first POD value and the value determined at the appearance of symptoms indicative of an irregular post-operative course or at the time of re-hospitalization for persistent symptoms was considered an independent variable to predict the occurrence of complication (dependent variable).
As additional parameters evaluated at the time of suspicion of postoperative complications (on the basis of fibrinogen values), we also tested the C-reactive protein (CRP) and the IL-6 levels.
For the analysis, blood samples were obtained and placed in tubes containing sodium citrate for the fibrinogen assay. Plasma was separated with centrifugation at 4 °C within 15 min. Plasma fibrinogen levels were assessed with the routine method described by Clauss (normal range 200–400 mg/dL) [
19]. The WBC count was analyzed via an automatic hematological blood analyzer (Coulter Gen-S; Beckman Coulter, Fullerton, CA). Serum concentrations of CRP (mg/L) were measured by nephelometry using an autoanalyzer with a lower detectable limit of 1 mg/dL. Serum IL-6 levels were assessed with an enzyme-linked immunosorbent assay (ELISA) using a commercially available kit (DRG Instruments GmbH, Marburg, Germany). All the assays were performed at the clinical laboratory at the Businco Hospital, Cagliari, Italy, according to the manufacturer’s protocol. The laboratory personnel were blinded to the clinical information.
Perioperative management
Laparoscopic gynecologic surgeries followed a standardized procedure and were performed by the same qualified/experienced principal surgeon with a minimal caseload of 50 operations/years and by a small selected team throughout the study duration. All major surgical procedures performed at the Gynecologic Oncology Department at the Businco Hospital were included in the study. The procedure included surgery for benign conditions such as ovarian cysts with a pre-operative suspicion of cancer or fibromatous uteri and malignancies (ovarian, endometrial, cervical cancers, and uterine sarcoma); cases with a complex surgical history that were referred to the gynecologic oncology team (i.e., deep endometriosis); and risk-reducing surgery. Minor diagnostic procedures were excluded. In cases of malignant gynecological tumors, the extent of surgery was determined in each case by the surgeons, who aimed to achieve a radical resection according to specific cancer treatment guidelines. All patients received preoperative antibiotic prophylaxis (2000 mg ceftriaxone intravenously) 30 to 60 min before surgery. Anticoagulation was achieved with low-molecular-weight heparin administered at a prophylactic dose (adjusted for body weight and thromboembolic risk according to clinical history) starting from the day of surgery in all patients.
Statistical analysis
Categorical data were reported as a number (and percentage) of patients. To define the distribution, continuous variables were explored for skewness and kurtosis. Normally distributed continuous variables were reported as the mean ± standard deviation and non-normally distributed continuous variables were reported as the median with a range of minimum and maximum values. Comparisons of means were performed with Student’s t test for quantitative variables following normal distribution and the Mann-Whitney method for non-normal distributions. Comparison of qualitative variables was performed with the chi-square test. Correlation analysis was performed by Pearson’s correlation methods for variables following the normal distribution.
In order to assess the predictive role of routine laboratory parameters (% change of fibrinogen and WBC) to detect the onset of postoperative complications (primary aim of the study), we used the receiver operating characteristic (ROC) curve analysis and the respective areas under the curve (AUC) to check the sensitivity and specificity of the continuous variable (% change of Fbg and WBC). The AUC of ROC analyses using Fbg and WBC changes showed higher than 0.60, and so ROC analyses were used to determine the appropriate threshold value of these two variables. Then, the identified thresholds were used to perform the univariate regression analysis in order to identify the association between laboratory variables (independent variables) and the presence of post-operative complications (outcome dependent variable). Factors significant in the univariate analysis entered the multivariate logistic regression analyses, running a stepwise elimination model (variable entered if p < 0.05, variable removed if p > 0.1). Odds ratios and confidence intervals were determined. A 2-sided p value < 0.05 was considered significant. All statistical analyses were performed using SPSS version 17.0 (SPSS Inc., Chicago, IL).
Discussion
Although minimally invasive surgery in gynecology with laparoscopic procedures is now a very common and well-established practice, the incidence of complications is variable and often unpredictable. In the present study, we observed that the rate of major complications after gynecological laparoscopic surgery in the total population was 2.85%, with a rate of 3.2% seen after surgery for malignant diseases. This is within the range reported in previously published studies [
7,
20,
21]. The early identification of patients with complications would be undoubtedly of great clinical value because it would facilitate optimal timing of therapeutic interventions to minimize the sequelae of such complications, particularly major ones that require surgical intervention for correction with a significant threat to short- and long-term outcomes [
22]. In the present study, we investigated the ability of fibrinogen concentration to enable clinicians to rapidly identify postoperative complications following gynecological laparoscopic procedures. We found that the increase in fibrinogen concentration on postoperative days had a high diagnostic accuracy to identify postoperative complications early, with an AUC of 0.90 (
p = 0.012), sensitivity of 0.89, and specificity of 0.84. Moreover, we identified that a postoperative change in the fibrinogen concentration
> 20% was useful for predicting postoperative major complications following gynecologic laparoscopic surgeries.
The role of acute-phase proteins in predicting the development of postoperative complications have been already investigated, but, to date, the majority of studies have focused on CRP [
23‐
32].
Moreover, although the data in different surgery settings for both malignant and benign pathology are numerous, to date, the data in the setting of laparoscopic gynecologic surgery are very scarce and of low quality [
33,
34]. One prospective study carried out in patients who underwent laparoscopic bowel resection for deep infiltrating endometriosis showed that postoperative CRP values were significantly higher in those patients who developed postoperative anastomotic leakage or ureteral injury, while on the contrary, the decrease in CRP levels was associated with an uncomplicated postoperative course [
35].
Another laboratory parameter, historically used as a marker of complications, is the WBC count. In our study population, the changes in postoperative WBC counts showed a low predictive ability of postoperative complications. This result is in accordance with those of other studies [
16,
32,
36]. In particular, Swets et al. [
36] found that significant changes in the WBC count occurred 7 days after surgery when most postoperative complications have already had a negative clinical evolution. In the present study, we also measured IL-6 levels. This has recently been reported as a predictor, in particular, of postoperative infectious complications after major abdominal surgery [
37]. In the present work, IL-6 was considered as an additional parameter. In fact, several of our previous studies have demonstrated that the increase in IL-6 levels correlates perfectly with the increase in fibrinogen levels [
38,
39]. However, the use of IL-6 as routine marker is more expensive (less cost-effective) than fibrinogen.
The results of the present study also report a significant association between the magnitude of the postoperative changes in the fibrinogen levels and complication severity, using the Clavien-Dindo grade, following laparoscopic surgery for gynecological diseases. These findings are consistent with those of some previous studies that have shown a significant association between the magnitude of the postoperative systemic inflammatory response, measured by CRP levels, and the severity of complications following surgery for colorectal, gastric, and esophageal cancers [
30,
40,
41].
Notably, our study revealed that in patients with early complications the high fibrinogen level may lead to prompt identification of the complication; however, in those with late complications, this increase is always associated with the core symptoms of the complication that results in readmission of the patient. Therefore, during hospitalization, when the patient’s clinical condition does not progressively improve as expected, the assessment of fibrinogen levels could help determine whether the patient has a previously unnoticed or misdiagnosed complication during surgery. This evidence would allow the diagnosis of the complication very early and to allow us to adopt any measure to promptly resolve the complication in the shortest possible time before it compromises the clinical picture and increases morbidity and hospitalization costs. Thus, the early diagnosis of complications, particularly after laparoscopic gynecological surgery, is essential for better outcome of the patient. It is especially very important when the patient is planned to undergo an early discharge, as in minimally invasive laparoscopic surgery.
Publisher’s Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.