Hysterectomy is used as a treatment for various benign gynecological conditions including leiomyomas of the uteri (myoma uteri, multiple myomas), adenomyosis, dysfunctional uterine bleeding (DUB), endometrial hyperplasia, uterovaginal prolapse, and endometriosis [1, 2]. In the past, open abdominal hysterectomies were the norm for all hysterectomies. However, given the advances in surgical techniques, minimally invasive alternatives such as laparoscopic and robotic surgeries have become viable options. Nevertheless, there are no well-defined criteria guiding the selection between robotic or laparoscopic surgery. This decision often depends on the surgeon’s preference and experience [3], or occasionally, the cost consideration, with robotic surgery being more expensive compared to the less costly laparoscopic surgery [4, 5].

Robotic surgery has several technical advantages, including a three-dimensional view and articulating arms that feature seven degrees of movement to facilitate precise control; all arms are operated through a console. These features contribute to improve surgeon’s comfort and agility during the procedure [6, 7]. Given these advantages, it is anticipated that robotic surgery would be less invasive and associated with fewer complications than laparoscopic surgery [8]. Short-term complications of gynecological surgery include the need for blood transfusion, an extended hospital stay due to significant bleeding, and a large amount of Jackson-Pratt (JP) drainage [9, 10].

Within this context, we conducted a retrospective analysis of bleeding records of patients who underwent either robotic or laparoscopic hysterectomy at our institution. The results would offer valuable insights into selecting the most appropriate surgical approach when planning a hysterectomy.

We enrolled patients with benign gynecological conditions who underwent robotic or laparoscopic hysterectomy (RH or LH, respectively) from January 2021 to December 2022 at our institution. The diagnoses included leiomyoma of the uterus, adenomyosis, uterovaginal prolapse, endometrial hyperplasia, ovarian cysts, a pelvic mass, dysfunctional uterine bleeding (DUB), menorrhagia, endometrial intraepithelial neoplasia (EIN), polyps of the endometrium, and chronic pyometra. All conditions were confirmed as benign via histopathological examinations conducted during the surgical procedures.

In some cases, additional surgical procedures were combined with the hysterectomies, including adnexal surgeries, bilateral and/or unilateral salpingectomy, ovarian wedge resection, ovarian cystectomy, sacro-colpopexy, anterior and/or posterior colporrhaphy, labioplasty, and cervical polypectomy (in cases of subtotal hysterectomy). Some cases required (minimal) adhesiolysis to ensure surgical success. Cases involving surgeries unrelated to gynecological issues, or performed by a surgical department other than our obstetrics department, were excluded.

We collected the following data: hemoglobin (Hb) levels within 1 month prior to surgery, Hb levels on postoperative days 1 and 3 [11], total JP drainage up to postoperative day 2 or 3, input/output (I/O) difference on the day of surgery, adhesiolysis performed during surgery (classified by grade), the weight of the excised uterus, and the incidence of postoperative blood transfusions. The total JP drainage is an indicator of blood loss [12]. Adhesiolysis have the potential risk of arterial or venous injuries, increasing the risk of bleeding [13]. Data were collected from hospital records of blood tests, clinical observations, intraoperative photographs, and surgical and prescription documentation.

First, we compared all patients who underwent RH and LH. Subsequently, we categorized patients of both group into three subgroups based on the weight of the uterus: (1) 250 g or unknown, (2) 250 g – 500 g, (3) 500 g – 1 kg, and (4) 1 kg or more. Additionally, we divided both RH and LH patients into three subgroups based on the adhesiolysis grade: no adhesiolysis (O), adhesiolysis at a single site (S), and adhesiolysis at multiple sites (M) [14].

We performed detailed analyses considering both uterine weight and adhesiolysis grade. This yielded totals of 12 RH subgroups and 9 LH subgroups. Among LH patient, no uterine weight exceeded 1 kg, precluding a comparison with RH and LH patients with uterine weights exceeding 1 kg.

We compared all subgroups in terms of the percentage decrease in Hb level from within 1 month prior to surgery to postoperative days 1 and 3, total JP drainage to postoperative day 2 or 3, I/O difference on the day of surgery, and the numbers of patients who required transfusions.

$$\text{Hb drop (\%)}=\frac{Hb\left(\mathit{\text{preop.}}\right)-Hb\left(\mathit{\text{postop. day}}1\mathit{\text{or}}3\right)}{Hb\left(\mathit{\text{preop.}}\right)}*100$$

The required sample size was calculated using PASS software (version 12; NCS, LLC Inc., Kaysville, UT, USA). To achieve a power of 0.9 (alpha 0.05, beta 0.1) with a 20% dropout rate, the required sample sizes for both groups were 56. In this study, our sample sizes were 135 in the RH group and 113 in the LH group. All analyses employed IBM SPSS software (version 27; SPSS Inc., Chicago, IL, USA). The independent samples t-test was used for subgroups comparison. A p-value less than 0.05 was considered statistically significant. The study received approval from our Institutional Review Board (approval no. 2023-10-008).

We compared the 135 patients who underwent RH and the 113 patients who underwent LH. In terms of indications for hysterectomy, 131 patients (97%) had leiomyomas or adenomyosis, 4 patients (3%) had endometrial disease in the RH group; 99 patients (87.6%) had leiomyomas or adenomyosis, 9 patients (8%) had endometrial disease, and 5 patients (4.4%) had uterovaginal prolapse. There were no statistically significant differences in the basic patient characteristics (age, body mass index (BMI), baseline hemoglobin and hematocrit) between the two groups. Statistically significant differences were found in the operation time and the total JP drainage volume on postoperative day 2 or day 3 (Table 1).

For further analysis, patients in the RH or LH groups were categorized into four subgroups based on uterine weight. However, no LH patient had a uterine weight exceeding 1 kg; limiting comparisons to the three other subgroups from either group. In all comparisons, RH patients consistently exhibited lower total JP drainage up to postoperative day 2 or day 3 than LH patients (Table 2).

In addition, we compared the three adhesiolysis subgroups of both RH and LH groups. In all comparisons, total JP drainage to postoperative day 2 or 3 was lower in RH than LH patients (Table 3).

Subsequently, all patients who underwent RH or LH were categorized into 12 or 9 subgroups respectively, based on both uterine weight and the extent of adhesiolysis. In the majority of comparisons, the percentage decreases in Hb levels from within 1 month prior to surgery to postoperative days 1 and 3 did not exhibit significant differences. However, noteworthy decreases (in comparison to other subgroups) from within 1 month prior to surgery to postoperative day 1 approached significance in subgroups 1 and M (uterine weight less than 250 g or unknown and adhesiolysis multiple sites) and 2 and M (uterine weight 250 g or more but less than 500 g and adhesiolysis multiple sites). Additionally, the percentage decrease in Hb compared to other subgroups from within 1 month prior to surgery to postoperative day 3 approached significance in subgroup 3 and 0 (uterine weight 500 g or more but less than 1 kg and adhesiolysis 0 site) (Table 4).

When comparing the total JP drainage to postoperative day 2 or 3, significant differences were observed between all subgroups except 1 and S, 2 and S, 3 and 0, 3 and S, and 3 and M (Table 4).

I/O difference on the day of surgery did not demonstrate significant differences among the subgroups (Table 4). Only two cases, both undergoing LH, required blood transfusions (Table 1).

Our patient selection intentionally focused on patients with benign conditions. In patients with uterine malignancies, such as ovarian, endometrial, or cervical cancer, malignancy-related vascular proliferation [15] can lead to an increase in blood volume within the pelvic cavity, inducing inflammation and adhesions around the pathological regions [16]. Had we included patients with both benign and malignant conditions, we would have failed to capture only bleeding factors attributable to surgical approach; we wished to exclude factors that were affected by various malignancies.

The strength of our study thus lies in the careful selection of the sample; limiting enrollment to patients with benign conditions were enrolled. This approach allowed us to determine whether differences of bleeding-related factors were solely influenced by variation in surgical method of RH and LH. Furthermore, we excluded potential bleeding associated with additional surgical processes used for lymphadenectomies or extensive adhesiolysis, as well as the removal of malignant masses.

We exclusively enrolled patients who underwent surgery at our institution. This ensured the involvement of a single surgical team with consistent surgical protocols to minimize non-surgical variations. By maintaining this, we aimed to eliminate factors unrelated to surgical procedures.

We compared bleeding parameters between RH and LH including the percentage decrease in Hb levels, total JP drainage, I/O difference on the day of surgery, and the number of patients who required red blood cell transfusion. The results suggest that RH is not inferior to LH but rather superior in certain aspects. This superiority is attributed to the excellent three-dimensional view, articulating arms that afford seven degrees of movement for precise control, and reduced surgeon fatigue operating at a console [6, 17].

Compared to LH, RH was not significantly superior in terms of the percentage pre- and postoperative Hb level reductions or the postoperative blood input: output ratio. However, RH was clearly better in terms of the total JP drainage to 2 or 3 days after surgery, except when the uterine weight was between 500 g and 1 kg. Also, RH (compared to LH) reduced the need for blood transfusions. Consequently, robotic surgery was superior to laparoscopic surgery in terms of bleeding control.

In cases where surgery was challenging or could be complicated due to factors such as severe obesity, a history of multiple obstetrical and gynecological surgeries, and a high number of previous cesarean sections, or when there was a possibility of cancer considering results of ultrasonography, computed tomography (CT), or magnetic resonance imaging (MRI), robot-assisted hysterectomy had been preferred over laparoscopic hysterectomy. This study included cases that had potential for cancer but were histopathologically proved to be benign only. Therefore, despite the likelihood of increased intraoperative bleeding in robot-assisted surgery which led to more Hb drop rate between preop and postop than laparoscopic hysterectomy (p-value $>$ 0.05), the advantages of robotic surgery allow for effective hemostasis during the surgery, resulting in a less amount of postoperative JP drainage than laparoscopic hysterectomy (p-value 0.05).

While this study did not compare hospitalization durations, complications, or long-term outcomes, the advantages afforded by RH in terms of JP drainage and the reduced need for blood transfusions suggest advantages of the lower bleeding associated with RH which may lead to shorter hospital stays and a decrease in short-term complications [18, 19]. Ultimately, this could afford superior long-term outcomes. Future studies to prove this hypothesis are needed.

However, we did not differentiate patients undergoing total and subtotal hysterectomies, leading to an imbalance in the rates of total and subtotal hysterectomies within the RH and LH groups. Although the difference in bleeding between surgical methods of total and subtotal hysterectomy may be a factor affecting bleeding risks, this was not considered due to study design limitations. A future study that differentiates between patients undergoing total and subtotal hysterectomy would provide valuable insights. This could help determine the optimal choice among robotic total, robotic subtotal, laparoscopic total, and laparoscopic subtotal hysterectomy when bleeding must be minimized.

When considering a blood transfusion, it is important to review the percentage decrease in Hb levels from within 1 month prior to surgery to postoperative day 1 or 3. Additionally, cases with significantly low Hb levels on postoperative days 1 or 3 were included. This suggests that postoperative blood transfusion may not solely reflect bleeding during the surgical procedure. In this study, of the two patients who received blood transfusions, one exhibited a substantial drop in Hb level, which is an absolute value of 4.8 g/dL (a 36.36% reduction) from 1 month prior to surgery to postoperative day 1. However, the other patient did not exhibit a significant decrease with an absolute value was 2.3 g/dL (a 20.20% reduction). The decision for transfusion in this case was based on the abnormally low absolute Hb level on postoperative day 1.

Our study demonstrates that RH is not inferior, but rather superior to LH in terms of certain bleeding outcomes. In particular, RH exhibited significant advantages in terms of the total postoperative JP drainage level in almost all subgroups. In subgroups with more severe adhesions, RH was associated with a lower decrease in the Hb levels after surgery compared to LH.

Our findings provide valuable guidance for collaborative decision-making involving both a medical team and a patient. It is essential to minimize surgical blood loss when patients refuse blood transfusions for religious or personal reasons, or dealing with individuals of Rh-negative blood types. In the latter cases, the availability of compatible blood may challenge for prompt transfusions. Moreover, for patients facing financial constraints and unable to afford robotic surgery due to the high cost [17], our findings offer informed insights to make well-considered surgical choices.

The datasets generated and analyzed during the current study are not publicly available due to the Institutional Review Board has not approved the disclosure of patient data, but are available from the corresponding author on reasonable request.

Conceptualization, TYK and SHC; Data curation, TYK and SHP; Formal analysis, JHS and SHC; Methodology, JIC and SHC; Validation, JHS and JIC; Writing—original draft, TYK, JIC and SHC; Writing—review and editing, TYK, SHP, JHS, JIC and SHC. All authors contributed to editorial changes in the manuscript. All authors read and approved the final manuscript. All authors have participated sufficiently in the work and agreed to be accountable for all aspects of the work.

The study was conducted in accordance with the Declaration of Helsinki, and the protocol was approved by the Ethics Committee of Soonchunhayng University Bucheon Hospital (IRB No. 2023-10-008). This study is a retrospective study conducted through medical chart review. Therefore, informed consent was not required and the IRB had already approved it.

We would like to express our gratitude to all those who helped us during the writing of this manuscript. Thanks to all the peer reviewers for their opinions and suggestions.

This study was supported by a research fund from Soonchunhyang University.

The authors declare no conflict of interest.