Intraoperative and postoperative complications of gynecological laparoscopic interventions: incidence and risk factors

Purpose The aims of this study were to determine the incidence of intraoperative and postoperative complications of laparoscopic gynecological interventions and to identify risk factors for such complications. Methods All patients who underwent laparoscopic interventions from September 2013 to September 2017 at the Department of Gynecology, Obstetrics and Reproductive Medicine, Saarland University Hospital were identified retrospectively using a prospectively compiled clinical database. Binary logistic regression analysis was used to identify independent risk factors for intra- and postoperative complications. Results Data from 3351 patients were included in the final analysis. Overall, 188 (5.6%) intraoperative and 219 (6.5%) postoperative complications were detected. On multivariate analysis, age [odds ratio (OR), 1.03; 95% confidence interval (CI) 1.01–1.04], surgery duration (OR, 1.02; 95% CI 1.02–1.03), carbon dioxide use (OR, 0.99; 95% CI 0.99–1.00), and surgical indication (all p ≤ 0.01) were independent risk factors for intraoperative and duration of surgery (OR, 1.01; 95% CI 1.01–1.02; p ≤ 0.01), carbon dioxide use (OR, 0.99; 95% CI 0.99–1.00; p ≤ 0.01), hemoglobin drop (OR, 1.41; 95% CI 1.21–1.65; p ≤ 0.01), and ASA status (p = 0.04) for postoperative complications. Conclusion In this large retrospective analysis with a generally low incidence of complications (5.6% intraoperative and 6.5% postoperative complications), a representative risk collective was identified: Patients aged > 38 years, surgery duration > 99 min, benign or malignant adnex findings were at higher risk for intraoperative and patients with surgery duration > 94 min, hemoglobin drop > 2 g/dl and ASA status III at higher risk for postoperative complications.


Introduction
Minimally invasive techniques have replaced a plethora of open surgical interventions [1]. Numerous studies have shown that laparoscopy offers many advantages over open surgery, such as reduced intraoperative blood loss, postoperative pain levels, hospitalization and faster recovery times, including for obese patients, general surgeries, and the treatment of benign and malignant gynecological diseases [2][3][4][5][6]. Although minimally invasive techniques are the gold standards in many fields of surgery, they are intra-abdominal procedures, and complications of laparoscopic interventions are not less severe [7]. Reported overall complication rates vary between different authors, with ranges of overall complication rates between 4-41.21%, possibly due to differences in study design and cohort size [8,[17][18][19][20]. In addition, a major problem in research on laparoscopic complications is the lack of standardized definition and recording of complications. The aim of this observational study was to determine the rates of intraoperative and postoperative complications of laparoscopic gynecological interventions, and to identify risk factors for the occurrence of these complications.

Study design and participants
All patients who underwent laparoscopy at the Department for Gynecology, Obstetrics and Reproductive Medicine, Saarland University Hospital between September 2013 and September 2017 were identified retrospectively using a prospectively compiled institutional clinical database. All patients who underwent laparoscopic gynecological surgeries were included in this study. Those who underwent laparoscopically assisted vaginal hysterectomies and those for whom intervention or patient data were incomplete were excluded.
A gynecological fellow collected the study data by systemic chart review. Data on patient characteristics [age (years), height (cm), weight (kg), body mass index (BMI; kg/m 2 )], surgical parameters such as surgery duration (min), carbon dioxide (CO 2 ) use (l) (defined as amount of carbon dioxide in liter insufflated from the insertion of the Veress needle to the removal of the last trocar), hemoglobin (Hb) drop (g/dl) (defined as difference between the last preoperative and the first postoperative hemoglobin control in g/dl), surgical indications [symptomatic uterine myoma, benign or malignant adnexal finding, cervical cancer, endometrial carcinoma, endometriosis, urogynecological indication, cervical intraepithelial neoplasia (CIN)], histological findings (benign/malignant), laparoscopic intervention difficulty according to Barakat [low-medium (levels I and II), medium high-high (levels III and IV)], American Society of Anesthesiologists (ASA) physical status, adhesiolysis and duration of postoperative hospital stay (days) were extracted [9,10].
Surgical complications were recorded as intra-and postoperative. Bleeding, blood transfusion, organ injury, conversion to laparotomy, resuscitation and skin emphysema were defined as relevant intraoperative complications. Postoperative complications (all deviations from the normal postoperative course) occurring during a 6-week period after surgery were classified using the five-grade Clavien-Dindo system and as minor (levels I and II) and major (levels III to V) [8,11].

Procedures
All laparoscopies were performed by experienced consultants. All patients received perioperative antibiotic prophylaxis with cefuroxime (single shot, 1.5 g i.v.; Fresenius Kabi, Bad Homburg, Germany) and, if necessary, metronidazole (single shot, 500 mg i.v; Fresenius Kabi, Bad Homburg, Germany) and a perioperative gastric tube and indwelling urinary catheter until mobilization on the first postoperative morning. For perioperative thromboembolism prophylaxis, low-molecular-weight heparin (enoxaparin sodium, 40 mg; Sanofi, Paris, France) was injected daily during patients' entire hospital stays [12]. For hysterectomies of benign and malignant indication, such prophylaxis was administered for 1 and 4 weeks, respectively. Laparoscopy was performed in lithotomy position using the four-port technique; specifics of the techniques used for different interventions have been described previously [12][13][14][15][16]. After surgery, all patients were taken to postoperative ward for 4 to 6 hours of monitoring vital signs, then moved to general ward; blood count control was performed on the first postoperative day. Before discharge from hospital, final examination with speculum insertion and transvaginal and kidney ultrasound was performed in all cases. Patients were advised to present to the emergency ward if any deviation from the normal postoperative course occurred.

Statistical analysis
All data were collected with Microsoft Excel (Excel 2014, Microsoft Corporation, Redmond, WA, USA). The statistical analysis was performed using IBM SPSS (Version 25, IBM SPSS Inc., Chicago, IL, USA). For quantitative data, the Kolmogorov-Smirnov test was used to assess the normality of distribution. As the data were not distributed normally, medians and ranges were calculated and the Mann-Whitney U test was used for comparison. For categorical variables, absolute and relative frequencies were determined, and the Pearson chisquared test was used for comparison. Patient characteristics and surgical parameters were compared according to the presence or absence of intraoperative and postoperative complications. Multivariate binary logistic regression with stepwise forward and backward selection was used to identify factors predicting the occurrence of these complications. A receiver operating characteristic (ROC) curve was used to vary the discrimination thresholds for quantitative risk factors.

Patient characteristics and surgical parameters
Of 3409 patients identified from the database, 58 were excluded due to lack of information; thus, data from 3351 patients were examined in the present analysis. The patient characteristics and surgical parameters are presented in  Postoperative complications were recorded in 219 (6.5%) cases; 92 (2.7%) were minor and 128 (3.8%) were major. According to clinical manifestation, revision of postoperative hemorrhage and hematomas was applied most commonly [n = 41 (1.2%)], followed by the antibiotic treatment of infections, hematomas, and abscesses [n = 28 (0.8%)] and transfusion for anemia [n = 26 (0.8%)]. Twenty-three (0.7%) patients presented at the hospital due to the post-discharge exacerbation of pain, 13 (0.4%) patients had cardiopulmonary complications, 7 (0.2%) patients had ileuses treated conservatively with prokinetic stimulation, and 5 (0.1%) patients had ileuses requiring surgery. Minimal vaginal cuff dehiscence that could be treated conservatively was documented in 3 (0.1%) cases; 12 (4%) cases of dehiscence requiring re-intervention occurred. Eighteen (0.5%) patients with intestinal problems (e.g., peritonitis, anastomosis insufficiency, fistula) and four (0.1%) patients with bladder injury, ureteral lesion, or vesicovaginal fistula required surgical treatment. Nerve lesions and paresthesia were registered in five (0.1%) cases, trocar site hernia or infection were recorded in two (0.1%) cases, lymphocele was recorded in one (0.02%) case, and compartment syndrome was recorded in two (0.1%) cases. Wound healing disorders were treated conservatively in two (0.1%) cases and operatively in four (0.1%) cases. Eight (0.2%) abscess revisions were recorded. Three (0.1%) patients had multiorgan dysfunction and 12 (0.4%) patients experienced other types of adverse event ( Table 2).

Discussion
In this large retrospective study based on prospective clinical data, the intraoperative and postoperative complication rates of laparoscopic gynecological surgeries were 5.6% and 6.5%, respectively. Patient age, duration of surgery, carbon dioxide use, and surgical indication were independent risk factors for intraoperative complications. These complications were more frequent in patients with benign and malignant adnexal findings than in those with endometriosis (reference group). Risk factors for postoperative complications were duration of surgery, carbon dioxide use, hemoglobin drop, and ASA status, with ASA grade III associated with a greater risk than ASA grade I. These low complication rates are similar to those obtained in other studies and confirm the safety of laparoscopic interventions. The standardization of surgical procedures and identification of risk collectives might help to further reduce complication rates.
A few previous major studies also focused on the incidence of complications in laparoscopic gynecological surgeries. Chapron et al. evaluated surgical complication rates of diagnostic and operative procedures (29,966 cases) and found a total rate of 0.46% [17]. Similarly, Härkki-Siren et al. evaluated the nationwide incidence of laparoscopic complications in a cohort of 70,607 patients and assessed a rate of 0.36% [18]. A higher incidence of complications was recorded in the analysis of Mac Cordick et al. with a rate of 2.9% out of 743 procedures and Leonard et al. with 3% in a cohort of 1033 patients [19,20].
Two main explanations can be offered for the difference between these rates and those obtained in the present study.
First, in contrast to the previous studies, the Clavien-Dindo classification, which includes minor complications (e.g., need for analgesic or antibiotic treatment) and thus results in higher overall complication rates, was used in this study [11].
For instance, the postoperative complications considered by Chapron et al. would be classified as of Clavien-Dindo grades ≥ III; the exclusion of minor complications can be argued to have led to the underestimation of complication occurrence [17]. In a study from Radosa et al., in which the Clavien-Dindo classification was used, the total complication rate among 7438 laparoscopies was 13% [8].
Another possible explanation for the "increase" in complication rates is the development of laparoscopic techniques over time and their expanded application to more advanced procedures, with concurrent evolution of systems    [17]. The "advanced" category of this classification encompasses levels III and IV of the Barakat classification used in this study. Advanced procedures comprised 9.4-11.5% of surgeries in the previous studies, and comparable (Barakat levels III and IV) procedures comprised 24.1% of surgeries in the present study.
As in this study, Mirhashemi et al. found that the surgical indication was an independent risk factor for intraoperative complications; they reported lower rates associated with endometriosis and ovarian cystectomy [21]. Similarly, Saidi et al. reported that ovarian cystectomy was associated with the lowest rate of common complications [7]. We found that endometriosis was associated with a lower rate of intraoperative complications than were benign and especially malignant adnexal findings. Significant correlations between malignancy and complications have been reported previously [9,22]. The differences between these findings and ours may be explained by differences in study design; in the present study, we examined a large sample of patients undergoing a wide range of gynecological interventions.
Increasing age was an independent risk factor for the occurrence of intraoperative complications in this study. Three recent studies with designs similar to that of the present study yielded similar results. Age was an independent risk factor for complications of gynecological laparoscopic interventions in a sample of 1451 cases (along with previous radiation therapy and malignant diagnosis), and for bowel injury during hysterectomy in a cohort of 15,557 patients (along with endometriosis, and abdominal surgical approach) [9,23]. Aside from surgery type, increasing age was the most important predictor of complications of laparoscopic gynecological surgery in a sample of 843 patients; the risk of operative injury or bleeding was five times greater among women aged ≥ 35 years than among CIN cervical intraepithelial neoplasia, ASA Classification American Society of Anaesthesiologists Classification (n = 3351) those aged < 35 years [21]. In the present study, we found that intraoperative complications were more likely to occur in women aged > 38 years. Older age may be a significant risk factor in this context due to the overall higher incidence of comorbidities with increasing age. Of note, malignancy was not identified as a risk factor in our analysis, as it was in a previous study [9]. Chi et al.'s sample included more surgeries with malignant (n = 724) than with benign (n = 691) indications, whereas the majority of interventions examined in the present study were performed due to benign indications [9]. Thus, this discrepancy may be attributable to the larger percentage of malignant cases in Chi et al.'s smaller cohort [9].
The duration of surgery (i.e., longer operative time) was an independent risk factor for intraoperative and postoperative complications in this study. Two previous studies yielded similar results. In one study, the operative time (along with arterial hypertension, chronic obstructive pulmonary disease, and ASA status) was an independent risk factor for postoperative complications of 9145 laparoscopic procedures performed for the treatment of endometrial cancer [24]. In the other study, the operative time was associated independently with 30-day complications after laparoscopic and robotic hysterectomies [28]. Longer operative times can be associated with greater intraoperative blood loss and thus greater intraoperative and postoperative transfusion requirements and an increased risk of surgical site infection [25,26]. They can also result in increased postoperative pain and increased risks of deep vein thrombosis and pulmonary artery embolism [27].
The ASA status describes patients' preoperative health and is a predictor of postoperative mortality and outcomes of laparoscopic and open surgeries [29]. In this study, ASA grade III was associated with a significantly greater risk of postoperative complications than was ASA grade I. Two previous studies yielded similar results. The ASA status was an independent risk factor for postoperative complications of laparoscopic procedures in a sample of 9145 patients with endometrial cancer, and higher ASA status was a significant predictor of postoperative complications and length of hospitalization [24,30]. Dean stated that this factor was useful for preoperative risk assessment [30].
The postoperative complication risk was also associated with the hemoglobin drop in this study. This variable has been reported as an independent risk factor for postoperative complications of other types of procedure (e.g., [31]). The hemoglobin drop presumably correlates directly with intraoperative blood loss and thus a greater postoperative transfusion requirement (a Clavien-Dindo grade II complication).
Carbon dioxide use was a risk factor for intraoperative and postoperative complications in this study. ORs for this variable were < 1, indicating that risk increased with reduced carbon dioxide use. For intraoperative complications, this association may reflect the conversion of laparoscopy to laparotomy (with no carbon dioxide use). We could not find a report on a comparable assessment of the association of carbon dioxide use with complications in the literature.
Based on the findings of this study, we can identify a representative risk collective, who may benefit from close postoperative monitoring and surveillance. The implementation of a standardized approach to surgical management and procedures that reduces the surgery duration and involves preoperative hemoglobin optimization might contribute further to the prevention of complications. Patients with ASA statuses > III might benefit from a multi-disciplinary approach to the preoperative optimization of health status.

Limitations
This study was conducted at a single center, which limits the generalizability of the findings. However, the inclusion of a large number of patients in the analysis helps to minimize this limitation. In addition, possible bias in the recording of surgical complications was avoided by having an advanced medical student (rather than a layperson) perform data acquisition using a standardized classification system [32]. In addition, although many studies have revealed a significant correlation between surgeon expertise and the occurrence of complications [18,19], we did not consider this variable as a possible risk factor in our analysis. However, highly qualified and experienced surgeons perform all laparoscopic procedures at our institution.

Conclusion
In this large, single-center study, the overall complication rate for laparoscopic gynecological surgeries was low and independent risk factors for such complications were identified. Patients aged > 38 years, surgery duration > 99 min, benign or malignant adnex findings were at higher risk for intraoperative and patients with surgery duration > 94 min, hemoglobin drop > 2 g/dl, ASA status III at higher risk for postoperative complications. The findings can be used to develop strategies aiming to further reduce intraoperative and postoperative complications of these procedures.
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http:// creat iveco mmons. org/ licen ses/ by/4. 0/.