Robotic surgery in the morbidly obese patient: a surgeon’s planning guide

Dr. Silasi is Associate Professor of Obstetrics, Gynecology, and Reproductive Sciences; Director, Discovery to Cure International Fellowship Program in Gynecologic Oncology; Director, Uterine Transplant Program; and Director, Obesity Robotic Pelvic Surgery Program, Yale School of Medicine, New Haven, Connecticut.

He reports receiving consulting fees from Olympus America.

Obesity is an important consideration when making decisions about the optimal surgical approach for a patient. However, regardless of the approach chosen, lack of intraoperative exposure and inability to identify anatomic landmarks in an obese patient heighten the difficulty of surgery and increase operative time.

In our practice, we believe that the robotic surgical system has significantly benefitted morbidly obese patients who require pelvic surgery. Very heavy patients who need surgery are routinely scheduled for a robotic approach.

Here I address some of the challenges to performing robotic surgery on patients with an extreme body habitus.

[Video: 5 steps to robotic hysterectomy]

Preoperative evaluation and planning

In morbidly obese patients, routine preoperative physical examination is frequently suboptimal. Important aspects to consider in order to deliver a specimen transvaginally are the size and mobility of the uterus and the caliber of the vagina. Imaging studies can be done to preoperatively assess uterine size, but we do not feel they are necessary. The delivery method can be determined intraoperatively, based on specimen size and consistency as well as vaginal elasticity and caliber. A small laparotomy by enlargement of an existing port incision or at a different location may be necessary. Power morcellation is currently a subject of significant controversy.

In our experience, large uterine size is not a contraindication for robotic surgery, and no surgeries for uteri larger than 1000 g have required laparotomy.^1,2 However, if cancer of the endometrium is present, a discrepancy between the vaginal caliber and uterine size can pose difficulties. Imaging studies for evaluation of the pelvic anatomy are helpful for planning the operative strategy when large pelvic masses are present. Masses up to 20 cm are not a contraindication for vaginal delivery if they have a cystic component or are malleable enough to fit transvaginally. The mouth of the containing bag is everted at the introitus and the cystic mass can then be deflated and delivered without contamination of the peritoneal cavity.

The limiting factor is the size of the laparoscopic bag, thus avoiding intraperitoneal spillage. The mass can be contained and partially drained in the bag without any spillage and contamination of the peritoneal cavity. Masses larger than 20 cm need to be approached with caution to avoid contamination of the peritoneal cavity with cystic fluid or other contents.

In patients who have a history of intra-abdominal procedures, the previous surgeons most likely encountered the same intraoperative difficulties: limited access and exposure, excessive manipulation of tissues, overuse of thermal energy, and possibly suboptimal hemostasis.

These factors all predispose to adhesion formation. In most of these cases, the omentum and intestines will be involved in adhesions and the risk of injury on first entry is increased. Incisional or spontaneous hernias are not easily recognized on physical examination because of the thickness of the abdominal wall. For this reason, if the umbilicus is not easily explored on palpation, using Palmer’s point as the site for insufflation is a safe option.

Pears versus apples. Commonly used parameters (overall weight and body mass index [BMI]) are not always accurate for preoperative assessment of the difficulty of a procedure. Discerning whether distribution of adipose tissue is in an androgenic (“apple”) or a gynecoid (“pear”) pattern has more value for predicting the difficulty of a procedure. Generally, abdomino-pelvic surgery is more difficult to perform in patients with androgenic fat distribution.

The androgenic pattern, characterized by a round, protruding abdomen and significant truncal fat deposition, is associated with higher intra-abdominal pressures, a large and heavy omentum, thick mesenteries, and hypertrophic epiploicae (Figure 1). These factors severely limit exposure and visualization. In addition, retroperitoneal fat deposition can measure several centimeters in depth and completely obscure the great vessels in the event that para-aortic lymphadenectomy is performed. In these patients, the abdomen is “tighter” and less likely to expand after muscle paralysis.

Patients who are “pears” (gynecoid fat distribution), may have very high weight but adiposity distributed less around the core, making intraoperative exposure easier to achieve. In these patients, the abdomen extends laterally after induction of anesthesia (Figure 2).

Figure 1. Apples. Massive core adiposity with only moderately high BMI. Surgical procedures can be very difficult in patients with this androgenic pattern of fat distribution.

Figure 2. Pears. In patients with gynecoid pattern of fat distribution, most of the excess weight is distributed away from the midsection.

Bowel preparation

Preoperative mechanical bowel preparation is a common clinical practice for laparoscopic pelvic surgeons. Current clinical data, mostly obtained from colorectal surgery practice, offer no evidence to support the claim that preoperative colon cleansing reduces the risk of anastomotic leaks or infectious complications. On the contrary, it may increase the rate of anastomotic complications. Randomized controlled trials have shown no improvement in exposure with use of laparoscopic gynecologic procedures.^3,4 Our patients are given instructions for a clear liquid diet the day before the procedure.

Intestinal length is the same in obese and non-obese patients but because the mesentery and epiploicae in obese patients are heavily laden with fat, their intestines occupy a much larger volume in the peritoneal cavity. This impairs manipulation of bowel and its storage in the upper abdomen during pelvic surgery. In addition, the hypertrophic omentum also takes up more space in the upper abdomen, thus displacing the bowel toward the pelvis. In extreme obesity cases, preoperative bowel preparation may be of value, in that it decreases the amount of intraluminal chime, thus facilitating bowel handling and, it is hoped, allowing better exposure and visualization. In cases in which an operator struggles to obtain visualization, considerations other than patients’ comfort, such as intraoperative safety and reduced operative time, should be prioritized.

Anesthesia concerns

The anesthesiology team plays an integral role in the success of robotic surgery in obese patients. For very large individuals, we request an experienced anesthesiology team.

Starting in the preoperative holding area, routine aspects of the preoperative activities, such as establishing adequate intravenous (IV) access, can pose challenges. Once docked, the size of the robotic system will interfere with the ability to quickly re-establish IV access and limit access to the heart, should direct defibrillation be necessary.

Obese patients have large tongues and crowded necks, and intubation can be difficult. After intubation, high tidal volumes are frequently miscalculated based on the patient’s weight, which can cause alveolar barotrauma and impaired pelvic field exposure. If the diaphragmatic excursions interfere with exposure in the operative field, the tidal volume needs to be decreased, if only temporarily.

Airway pressures in an obese patient will be high and achieving optimal ventilation will be more complex in steep Trendelenburg position. Especially when core adiposity is massive, a patient’s incline may need to go up to the maximum allowed by the operating table, usually 40 degrees.

Trendelenburg position and pneumoperitoneum affect respiratory mechanics in many ways. The intra-abdominal pressure and the weight of the abdominal contents impair the excursions of the diaphragm, and insufflation of the peritoneal cavity decreases lung compliance. Also, when the abdominal wall and its fascia are weak, CO₂ will leak more into the subcutaneous tissues around the trocar insertions. The amount of CO₂ absorbed from the peritoneal insufflation can be compounded by the CO₂ from subcutaneous emphysema. Very large patients have some degree of restrictive pulmonary disease and they are less efficient in eliminating excessive CO₂ even with increased minute-volume of ventilation. Permitting mild hypercapnia may be necessary to perform a robotic procedure, but it is not easily accepted by anesthesiology teams, because normocarbia helps preserve cerebrovascular homeostasis. As with any surgical procedure, the operative time should be kept to a minimum.

Intraoperative infusion of fluids is limited with this positioning. Urine production will be decreased because of the preoperative fluid deficit, patient positioning during surgery, and intra-abdominal pressure from the pneumoperitoneum.

In almost every case, steep Trendelenburg position increases intraocular pressure (IOP). Fortunately, blindness as a consequence of high IOP is exceedingly rare. The threshold for postoperative visual loss has been reported as 45–55 mm Hg. Even brief (30–45 min) increases in IOP at this level can cause apoptosis of retinal ganglion cells. Our anesthesiology teams routinely monitor IOP with a handheld tonometer. If it exceeds 40 mm Hg, it can be decreased with application of dorzolamide/timolol ophthalmic solution. Also, undocking the robotic system and placing the patient in reverse Trendelenburg position for only 10–15 minutes will decrease the IOP.

During a robotic procedure, obese patients may develop facial and airway edema. That will influence the decision to extubate because the anesthesiology team will have to consider the possibility of a difficult reintubation in the setting of obesity hypoventilation syndrome.

Patient positioning

Patient positioning is a crucial part of robotic surgery. Without adequate positioning, a simple case can become tedious, while difficult cases can pose insurmountable problems and a patient’s safety can be compromised.⁵ For exposure of the pelvic surgical field, a Trendelenburg position is necessary. Generally, the incline of the operating table steepens with increasing body weight.

Areas including but not limited to the upper back, shoulders, and posterior aspect of the calves will support the body weight and are subjected to the risk of decubitus injury. Careful and extensive padding is required to prevent soft-tissue pressure injuries and nerve injuries, especially at the level of the brachial plexus.

In addition, a patient must be immobilized on the table to prevent her from sliding down when the robotic system is docked. The trocars of the robotic system are in fixed position and sliding on the table will place undue tension on the trocars piercing the fascia. If sliding occurs, the patient cannot be repositioned with the robotic system docked.

We use 2 methods of stabilizing large patients on the operating table, depending on physician preference.

1. Patients are positioned on a gel pad over a bean bag (Vac-Pac Olympic Medical) and then placed in a low lithotomy position using Yellofin stirrups (Allen Medical Systems). A sling improvised from a bed sheet is wrapped around the bean bag and secured to the table (Figure 3). The disadvantage to using the bean bag is that a Bookwalter retractor cannot be secured to the operating table side rails in case of conversion to laparotomy.

2. An egg-crate foam pad and padded shoulder supports can achieve excellent patient stabilization. However, without a bean bag to cradle the patient, we use lateral extensions to the operating table if a patient is wider than the operating table. The disadvantage to lateral extensions is that the ergonomics of the assistant operator will be negatively affected (Figure 4).

When a patient’s arms cannot be placed alongside the body (Figure 5), the armboards are kept at an angle and the arms are bandaged to the arm boards with 3M Coban Self-Adherent Wrap (3M Healthcare) (Figure 6).

Figure 3. Gel pad over a bean bag.

Figure 4. Padded shoulder supports.

Figure 5. The arms cannot be tucked along the body.
 

Figure 6. Securing the arms to the armboards.

When the stirrups are not large enough to safely accommodate a patient’s legs, the legs can be secured to the stirrups with the same wrap (Figure 7).

Figure 7. Securing the legs to the stirrups.

A pendulous pannus will shift significantly after insufflation when a patient’s position is changed from horizontal to Trendelenburg. Depending on the amount of redundant tissue, it may be beneficial to stabilize the pannus before inserting the trocars. 3M Ioban Antimicrobial Incise Drape provides excellent skin adherence and the pannus can be taped to the thighs (Figure 8).

Figure 8. A pendulous pannus can be taped to the thighs to stabilize it.

We have also used Medfix Montgomery Straps (Medline Industries) with less efficacy (Figure 9).

Figure 9. The pannus is stabilized with Montgomery straps.

The jointed operating table cannot be in flexion of the patient’s thorax over the abdomen. That would significantly affect the exposure and visualization for most heavy patients. Wedge-shaped foam or folded blankets used by the anesthesiology team to hyperextend the neck and facilitate intubation should be removed from under the patient’s shoulders (Figure 10).

Figure 10. Intraperitoneal exposure can be compromised by flexion of the thorax, while extension has the opposite effect.

Pneumoperitoneum and port placement

The risks and benefits of different laparoscopic entry methods have been evaluated in numerous studies.^6-8 Our preferred way to achieve pneumoperitoneum is by direct insertion of the Veress needle, with insufflation running at the highest flow rate.

As the Veress needle penetrates the abdominal wall into the peritoneal cavity, gas pressure readings provide the surgeon with useful information and decrease the “blindness” of the blind insertion. The camera port is inserted once pneumoperitoneum is achieved. Insertion injuries of the solid organs (liver, kidney, pancreas) are exceedingly rare and a clean needle stick does not need to be addressed. The only life-threatening injury that demands immediate attention is to the greater vessels. For obese patients, the point of entry into the abdomen in relationship to the retroperitoneal greater vessels is higher than for thin patients, and for patients with very high BMI, the needle is not long enough to reach the vessels. However, a surgeon should keep in mind the possibility of vessel injury and retroperitoneal hemorrhage. A retroperitoneal hematoma should be observed for expansion, with intra-abdominal pressure kept to a minimum. If it appears stable, the surgeon can continue the procedure. Reinsufflation of the peritoneal cavity for the duration of the procedure will add to the hemostatic effect. Retroperitoneal exploration is usually counterproductive for small injuries of the great vessels and should be reserved for large iatrogenic defects.

Bowel penetration at initial entry can be hidden by the hypertrophic mesentery or epiploicae and stay unrecognized. However, an unrecognized clean puncture with the Veress needle will have no or minimal clinical consequences.

For most gynecologic procedures on obese patients, umbilical insertion is adequate. When the planned procedure includes para-aortic lymph node dissection, total omentectomy, or hysterectomy of uteri >18 weeks, a supraumbilical/epigastric location is more appropriate. In a patient who has had several prior surgeries or in whom an umbilical or para-umbilical hernia is present or suspected, it is prudent to place the insufflation needle at Palmer’s point.

In non-obese patients, the intra-abdominal pressure is <5 mm Hg. In very obese patients, the intra-abdominal pressure can be much higher and sometimes the entry pressure set at 15 mm Hg is inadequate. In that case, temporarily increasing the entry pressure to 20–25 mm Hg can overcome the weight of the abdominal wall. After pneumoperitoneum is achieved, the pressure can be decreased to 12–15 mm Hg or even lower if ventilation is difficult. However, dropping the pressure too low will make it more difficult to keep the intestines in the upper abdomen and the bowel will keep rolling into the operative field. While lower insufflation pressures aid with ventilation, higher pressures help with hemostasis.

Meticulous hemostasis maintains a dry operative field, which is indispensable for an exact operative technique. Anatomic landmarks will be indistinguishable because of obliteration by fat, and surgical planes will be difficult to identify in wet tissues. For the same reason, we do not use irrigation unless absolutely necessary.

After insufflation, the abdominal shape can change significantly, especially when the fascia is weak. The reach of the camera and instruments may become inadequate (Figure 11).

Figure 11. The planned port locations by need to be adjusted after insufflation.

We often place the ports symmetrically, in an arch centered by the camera port (“sunrise” or “half-moon” distribution). To the left of the camera port, at the same level or slightly below it, is the port for arm 2. Opposite and symmetrical to this is the port for the assistant. At the ends of the imaginary arch we insert the port for arm 1 on the right and arm 3 on the left. Additional ports may be needed during difficult cases, mostly to assist with bowel retraction and exposure of the operative field. However, these are only guidelines. Depending on a patient’s torso length and width, as well as the goals of the surgical procedure, the ports should be placed so as to maximize maneuverability and visualization. For patients with morbid obesity, we recommend performing the operation using all 4 arms of the robotic surgical system. After insertion of the laparoscopic ports, the patient is placed in a steep Trendelenburg position and the da Vinci System is docked. The robot is most often positioned between a patient’s legs. Side docking can be difficult because of a patient’s bulk; it also severely limits the reach of arm 3 into the upper abdomen. The newest robotic surgical system, the Xi, allows for great flexibility in docking site.

Uterine manipulator

In this patient population, abundant pre- and retroperitoneal adipose deposits completely obscure anatomic landmarks. We use a uterine manipulator to delineate the vaginal fornices. An assistant pushes on the manipulator when the preperitoneal fat of the bladder flap or other developed planes roll into the operative field. Also, in very difficult cases, pushing the manipulator cephalad adds up to 1 cm to the distance between the point of sealing the uterine vessels and the trajectory of the ureters, thus adding a maneuver to protect the ureters from thermal injury. In obese patients, the distance between the cervix and ureters may be slightly less than in thin patients, but the difference is clinically insignificant.⁹

Because of the heavy retroperitoneal fat, the ureters are usually not visualized transperitoneally, so it is important for a surgeon to be aware of the fact that the distance between at least one ureter and the cervix is <0.5 cm in 3.6%–12% of cases. Also, adding to the difficulty caused by lack of visualization and exposure, the distances between the ureters and cervix are not the same on the right and left in many cases.^9-11

Conclusion

Performing robotic procedures in extremely morbidly obese patients can be challenging even for experienced surgeons. The familiar landmarks of the surgical anatomy are obliterated by fat deposition and operative field exposure is suboptimal. The robotic surgeon, as team leader, must ensure that all preparatory steps are taken to minimize the morbidity of these high-risk procedures.

References

1. Silasi DA, Gallo T, Silasi M, Menderes G, Azodi M. Robotic versus abdominal hysterectomy for very large uteri.JSLS. 2013;17(3):400–406.

2. Gallo T, Kashani S, Patel DA, Elsahwi K, Silasi DA, Azodi M. Robotic-assisted laparoscopic hysterectomy: outcomes in obese and morbidly obese patients.JSLS. 2012;16(3):421–427.

3. Siedhoff MT, Clark LH, Hobbs KA, Findley AD, Moulder JK, Garrett JM. Mechanical bowel preparation before laparoscopic hysterectomy. A randomized controlled trial. Obstet Gynecol. 2014;123:562–567.

4. Won H, Maley P, Salim S, Rao A, Campbell NT, Abbott JA. Surgical and patient outcomes using mechanical bowel preparation before laparoscopic gynecologic surgery: a randomized controlled trial. Obstet Gynecol. 2013;121:538–546.

5. Ramanathan R, Carey RI, Lopez-Pujals A, Leveillee RJ. Patient positioning and trocar placement for robotic urologic procedures. In: Patel VR, ed. Robotic Urologic Surgery. 2^nd ed. London: Springer-Verlag; 2012:107–120.

6. Zakherah MS. Direct trocar versus Veress needle entry for laparoscopy: a randomized clinical trial. Gynecol Obstet Invest. 2010;69(4):260–263.

7. Vilos GA, Ternamian A, Dempster J, Laberge PY, The Society of Obstetricians and Gynaecologists of Canada. Laparoscopic entry: a review of techniques, technologies, and complications. J Obstet Gynaecol Can. 2007;29(5):433–465.

8. Deffieux X, Ballester M, Collinet P, Fauconnier A, Pierre F; French National College of Gynaecologists and Obstetricians. Risks associated with laparoscopic entry: guidelines for clinical practice from the French College of Gynaecologists and Obstetricians. Eur J Obstet Gynecol Reprod Biol. 2011;158(2):159–166.

9. Hurd WW, Chee SS, Gallagher KL, Ohl DA, Hurteau JA. Location of the ureters in relation to the uterine cervix by computed tomography. Am J Obstet Gynecol. 2001;184(3):336–339.

10. Gemer O, Simonovsky A, Huerta M, Kapustian V, Anteby E, Linov L. A radiological study on the anatomical proximity of the ureters and the cervix. Int Urogynecol J Pelvic Floor Dysfunct. 2007;18(9):991–995.

11. Gellhaus PT, Bhandari A, Monn MF, et al. Robotic management of genitourinary injuries from obstetrical and gynecological operations: a multi-institutional report of outcomes. BJUInt. 2015;115(3):430–436.