Introduction

When first introduced in 1985 as a viable replacement for open cholecystectomy [1], multi-port laparoscopic cholecystectomy (LC) was met with questions of technical feasibility and effectiveness. With improving skill set and comfort of use by surgeons, the technique soon evolved to become the gold standard for the treatment of symptomatic gallbladder disease. Now those same questions and concerns plague the introduction of a newer technology, single-site robotic cholecystectomy (SSRC), the premise of which is to further improve upon the advantages of minimally invasive surgery of the gallbladder [2].

While laparoscopic surgery can be done via a single site, multiple problems including instrument clashing within the port, reduced triangulation and poor surgical ergonomics have negated its prevalent use in teaching hospitals [3, 4]. SSRC offers a way to offset the limitations of single-site laparoscopic cholecystectomy (SSLC) by reducing tremor, restoring good hand–eye coordination and the ability to perform very small movements [5–8], while also maintaining the benefits of single-port operations which include improved cosmesis and quicker healing times for the patient [9]. SSRC has also been shown to decrease pain, blood loss, minimize scarring, provide greater patient satisfaction, while allowing the surgeon to work in a more ergonomically advantageous workstation [10].

Very few studies have looked at the perioperative and postoperative outcomes of this procedure in comparison with the LC. In addition, even fewer studies have looked at long-term postoperative outcomes of patients undergoing these procedures or have small sample sizes. Our study aims to investigate the outcomes of SSRC as compared to LC, all evaluated through a single surgeon’s experience at a high-volume tertiary health care center.

Patients and methods

A retrospective analysis of a prospectively maintained data at a tertiary care center was performed on cases performed between October 2011 and July 2014. All cases undergoing SSRC or LC for symptomatic gallbladder disease were included. Open cholecystectomies, cholecystectomies terminated for anesthesia complications, and patients who underwent surgery simultaneously for other indications (i.e., appendicitis) were excluded from this study. Data collected include patient demographics [age, gender, body mass index (BMI)], necessity of surgery (elective vs. urgent), preoperative diagnosis, American Society of Anesthesiologists (ASA) classification and any history of prior abdominal surgeries. A comprehensive list of all operative data including intraoperative complications (i.e., bile duct injury or spillage), conversion rate to laparoscopic or open surgery, requirement for the utilization of additional ports, total surgery time (incision to skin closure), robotic console time and docking time was also documented. Finally, postoperative information including length of stay, follow-up, visits to the emergency department (ED) within 30 days, any immediate postoperative complications as defined as those recognized within 30 days, and long-term complications were recorded. Postoperative complications included wound infection, incisional hernia, seroma formation, clear leakage, etc. Abdominal pain was defined as any complaint of continued pain that had not improved at the follow-up visit as documented by the primary surgeon in the EMR.

Surgical technique

Single-site robotic cholecystectomy

A 2-cm incision was made at the umbilicus followed by a 2-cm incision within the fascia. A multichannel silicon port was placed at the umbilicus. Pneumoperitoneum was established up to 14 mmHg. A diagnostic laparoscopy was performed. The robot was then docked starting with the camera port in the middle followed by 2 working arms 1 and 2 on the sides of the camera port through the same silicon port. Dissection was then carried out in Calot’s triangle, and after obtaining, the critical view of safety the cystic artery and cystic duct was identified, clipped and divided. The gallbladder was then dissected from the gallbladder fossa by electrocautery. The gallbladder was then removed through the umbilical site using a specimen retrieval bag. The fascia of the umbilical site was then closed with 0-Vicryl suture in an interrupted figure of eight fashion. The skin was closed with a 4-0 Monocryl suture in a sub-cuticular fashion (Fig. 1).

Fig. 1
figure 1

Single-site robotic cholecystectomy. a Multichannel single port with assistant port (green) in the middle. b Showing fascial defect. c Surgeon sitting on the console and performing the operation. d Wound after skin closure

Full size image

Multi-port laparoscopic cholecystectomy (4 ports were used)

A 5-mm incision was made at the umbilicus, and pneumoperitoneum was established using a Veress needle. A 5-mm trocar was then placed using a Visiport, and diagnostic laparoscopy was performed. An 11-mm port was then placed at the sub-xiphoid region followed by two more 5 mm ports: one at the subcostal region and another in the anterior axillary line, in line with the umbilicus for retraction. Steps of dissection of Calot’s triangle were the same as in robotic surgery. Specimen was then removed from the sub-xiphoid port in a specimen retrieval bag. Umbilical and sub-xiphoid ports were closed with size 1-Vicryl suture in a figure of eight fashion (Fig. 2).

Fig. 2
figure 2

Multi-port laparoscopic cholecystectomy. a Induction of pneumoperitoneum with Veress needle. b Standard four ports used. c Incision of the fascia for entry into the abdomen. d Wound after skin closure of all four ports

Full size image

Statistical analysis

All results were analyzed using SPSS software, version 23.0. The Chi-square test or Fisher’s exact test was employed to compare both single-site robotic cholecystectomy and multi-port laparoscopic cholecystectomy groups for categorical data, while Student’s t test was used to compare continuous data. A p value of 0.05 was established as clinical significance.

Results

A total of 678 cholecystectomies were performed between October 2011 and July 2014, of which, 415 (61%) were SSRC and 263 (39%) were LC (Table 1). All were performed for symptomatic gallbladder disease. The age range of all patients was 16–100, but there was no difference in the mean age seen between the two groups (54.1 vs. 55.8 years; p = NS). Both groups showed a higher percentage of females than males; however, this distinction was statistically different in the SSRC group (p = 0.004). LC patients had a greater mean BMI (30.5 vs. 29.0 kg/m2; p = 0.008) and were more likely to have undergone prior abdominal surgery (83.3 vs. 41.4%; p < 0.001). There was no statistical difference in the American Society of Anesthesiologists (ASA) classification between the two groups (Table 1).

Table 1 Preoperative data
Full size table

Overall, the percentage of patients with preexisting comorbidities was higher in the LC group (76.1 vs. 67.2%; p = 0.014). Further analysis revealed only coronary artery disease (9.1 vs. 4.9%; p = 0.027) and chronic obstructive pulmonary disease (3.8 vs. 1.5%; p = 0.049) to be statistically significant between the two groups. Pathology of biliary disease was also considered and investigated. Patients with acute cholecystitis were more likely to have undergone LC as compared to SSRC (36.9 vs. 18.3%; p = 0.001); however, this was based on intraoperative evaluation and postoperative pathology results and was not used in the decision-making process regarding which procedure each individual patient should undergo (Table 1).

The total operative times between the two groups did not vary significantly. The mean console time was 57 min, and the mean docking time was 6.8 min (Table 2). Estimated blood loss was reported as negligible in all cases with no cases requiring any blood transfusions. No intraoperative complications were reported. Twenty-five (6.1%) procedures in the SSRC group had to be converted for completion with twelve (2.9%) being converted to LC and thirteen (3.2%) converted to open procedures. The total rate of LC conversion or open conversion between the two groups was not statistically significant (6.1 vs. 4.9%; p = 0.551) (Table 2). Reasons for conversion included but were not limited to, body habitus limiting the use of robotic instrument length, adhesions, inability to define anatomy or achieve critical view, lack of plane or area for traction, inability to identify the gallbladder, inadequate visualization of vascular structures, or a general concern for patient safety as deemed necessary by the surgeon.

Table 2 Perioperative data
Full size table

Table 3 highlights the postoperative data from the two procedures. Mean length of follow-up was 2.5 months in the SSRC group and 3.3 months in the LC group (p = 0.12). Two cases of postoperative bile leak within the laparoscopic group and one in the robotic group were seen at follow-up. All cases were managed by ERCP and stenting. The average length of stay following the procedure was lower for patients undergoing SSRC (1.9 vs. 2.4 days; p = 0.012). Neither the frequency of 30-day postoperative visits to the emergency room or readmission rates in this time period varied significantly within the two groups. Major postoperative complications documented during follow-up at the surgeon’s office or at the tertiary care center in the short term included abdominal pain, post-op ileus, urinary retention, wound infection/non-infectious complications, non-purulent or clear wound discharge, seroma formation, or bile leakages. Abdominal pain (8.4 vs. 4.2%; p = 0.032) and wound infections (3.9 vs. 1.1%; p = 0.037) were seen more commonly in SSRC group, while all others showed no significant differences. At the time of last follow-up, the rate of incisional hernias was higher in SSRC group as compared to LC group (6.5 vs 1.9%; p = 0.006).

Table 3 Postoperative data
Full size table

Discussion

Gallbladder disease continues to be one of the most common medical problems leading to surgical intervention [11]. With over 0.5 million new cholecystectomies being performed within the US each year [11], gallbladder disorders have often been an ideal arena to test and compare new and innovative surgical techniques with current routine practices. All procedures included within this study period were performed by a single experienced surgeon.

Significant differences were noted in patient characteristics prior to undergoing their respective procedures. Both groups comprised of older adults, but the single-site robotic group did have a higher percentage of females. Reasons for this could be attributed to the better aesthetic appeal the robotic platform offers in terms of postoperative cosmesis compared to multi-port laparoscopic cholecystectomy [9]. Patients undergoing LC also typically had a higher BMI, which was not surprising as one of the reasons cited by the surgeon for conversions or the use of additional ports was due to the patient’s body habitus and the limitation of the robotic instruments. Since ASA classification as a stand-alone has been shown as insufficient evidence to describe the physical condition of patients [12] and has a high degree of variability between observers [13–16], we further investigated specific comorbidities in both groups. This sub-analysis showed that a greater percentage of patients within the LC group had preexisting conditions and that both coronary artery disease and chronic obstructive pulmonary disease were significantly more common than within the SSRC group. Factors that have at times included morbid obesity, previous abdominal surgery, acute cholecystitis and a history of severe pulmonary or cardiac diseases were once considered relative contraindications to laparoscopic procedures [17–20].

The total operative time did not vary significantly between the two groups. Though we did not specifically compare pure surgical dissection times, the average console time was 57 min, in line with the high degree of precision achieved in tissue dissection with the robotic platform. This is quite comparable to a study recently published where console time was 52.8 min [21]. The conversion rates in our series are similar to reported conversion rates for LC [22, 23].

While a prior comparative study did not show any statistically significant difference in the length of postoperative stay [21], our study showed an increased length of stay in the LC group, explained by the presence of more emergent cases and patients with more comorbidities in this group. In our series, abdominal pain was more frequently reported in SSRC group which is in sharp contrast to some studies comparing single-port procedures to multi-port procedures [24–27]. Another study has actually shown increased postoperative pain following single-incision laparoscopic cholecystectomy when compared to multi-port laparoscopic cholecystectomy [28]. It is reasonable to assume that the larger incision, continuous movements of instruments within a confined space and higher pressure around the umbilical port may contribute to the increased intensity of postoperative pain in the SSRC group. Since postoperative pain after LC is already minimal and no objective pain measuring scale was used in our study, a potential reporting bias in the SSRC group can also not be excluded. While pain can often be a subjective measure [29], all patients in this study were discharged on a standard medication regimen for pain similar to that used in other settings from other studies.

Another finding of significance was the incidence of wound infections. Since operative characteristics were controlled with the same surgeon having performed the procedures in both groups, other risk factors for wound infection were examined including diabetes, cigarette smoking, obesity, or patient colonization [12]. None of these factors were found to be significant. Furthermore, among patients diagnosed with wound infections, no differences were seen in the mean BMI between the robotic or laparoscopic groups.

Finally, the most significant finding of this study was a high rate of incisional hernia in the SSRC group. Twenty-seven patients (6.5%) were diagnosed with umbilical incisional hernias after having undergone SSRC as opposed to five patients (1.9%) in the LC group. The surgical technique and suture used for port site closure were similar in all cases. On subgroup analysis, the rate of incisional hernia did not vary significantly between the first hundred, second hundred, third hundred and subsequent remaining cases (7, 7, 6, 7; p = 0.984). This would suggest that the learning curve and surgical technique were likely not the cause of incisional hernia. No statistically significant differences were seen in the mean BMI of patients diagnosed with hernia after undergoing robotic or laparoscopic procedures. Of the five hernia cases found within the laparoscopic group, one patient had undergone 4 abdominal surgical procedures prior to their laparoscopic cholecystectomy; another patient had a previous history of multiple ventral hernia repairs prior to their laparoscopic cholecystectomy; another patient was a case that had been converted to open procedure; and finally another one of the patients had undergone colon resection for myxoma of the appendix, which had been completed robotically via single-incision prior to the laparoscopic cholecystectomy. This then leaves only one patient (0.4%) who had been diagnosed with incisional hernia postoperation, where the procedure was the likely cause of the defect. While all hernias diagnosed within the robotic group were hernias occurring within the umbilicus, excluding cases with any form of previous abdominal surgery (even those not including the umbilicus in the surgical field, i.e., Pfannenstiel incision), the data still reflects a greater incidence of hernia in the SSRC group than in LC arm (2.9 vs. 0.4%; p = 0.020).

Previous studies looking at SSLC versus LC have reported higher rates of surgical site infection (4.0 vs. 1.6%) and incisional hernia (2.2–5.8 vs. 0.3–1.8%) in the SSLC group [30–34]. The higher rate of incisional hernia in SSLC has been ascribed to local ischemia induced by placement of a single large port or multiple ports at a single site, which could potentially weaken the fascia [30]. Similarly, potential ischemia combined with the fact that the fascial incision for SSRC represents a 400% increase in the length of incision as compared to LC (20 vs. 5 mm) could contribute to the higher incidence of wound infection and subsequent incisional hernia following SSRC.

The major limitation of this study is the fact it was nonrandomized and retrospective in nature. The preoperative data were also not exactly comparable between the two groups with BMI, comorbidities, prior abdominal surgery, and more urgent cases being skewed in a manner producing more unfavorable cases within the LC group.

Decreased incisional hernia and wound infection incidence make it seem that LC should continue to remain the gold standard for treating gallbladder disease. If balanced against the higher rate of postoperative complications, SSRC is an effective alternative in uncomplicated benign gallbladder disease.