Introduction

Although laparoscopy has decreased the morbidity related to many urologic procedures, the solutions to complications arising in the minimally invasive environment frequently have no direct open surgical counterpart. For example, laparoscopic access is often obtained through a variety of initially blind punctures where viscera and blood vessels are susceptible to unique injuries. The insufflated abdomen presents specific potential physiologic complications, complex instrumentation can malfunction catastrophically, and inadvertent injury can occur outside the laparoscope’s field of view due to unobserved instrument gestures. Laparoscopy therefore requires a specialized knowledge base, and demands a unique set of troubleshooting skills. Whereas textbooks have been devoted to laparoscopic complications, this chapter will cover the main areas pertinent to the urologist in training, as well as for the practicing urologist wishing to broaden their knowledge.

Since urologic laparoscopy is no longer in its infancy and surgical complexity has increased, warnings and advice about complications from the dawn of laparoscopy may no longer be appropriate. For example, open conversion is less often required for the contemporary surgeon than it was in the past (in early series prior to 2000 conversion was as high as 2.1% [1]). Although the threshold to convert must remain low, and the window to address an evolving complication is short, the options and expertise to intervene without making an incision have expanded. The surgeon must exercise particular care during the learning period; multiple studies in the urologic [2, 3] and general surgical literature [4] have demonstrated an inverse relationship between surgeon experience and complication rates.

Complications in this chapter are organized by major category: (i) access, (ii) physiologic, (iii) patient positioning, (iv) end‐organ (vascular, bowel, solid organ), (v) postoperative, and (vi) miscellaneous. Where appropriate, complications pertaining to specific urologic procedures will be highlighted. Since robotic techniques have essentially duplicated laparoscopic counterparts for all procedures, all complications discussed herein and their management are generalizable to robotic surgery. The classification schema proposed by Clavien–Dindo will not be specifically reviewed, but the reader is urged to review this separately [5].

Access complications

Regardless of the method used to gain initial deliberate entry into the peritoneal (or retroperitoneal) cavity, access remains a significant source of morbidity and potential mortality. The incidence of any type of access injury ranges from 5 in 10 000 to 3 in 1000 [6]. In a recent literature review of 41 articles, including almost 700 000 procedures for which Veress needle access was used, 1575 (0.23%) injuries occurred [7]. The majority (92%) of these injuries were minor. Although the risk of access injury is low, it is not negligible, and a surgeon is statistically likely to encounter several over the course of a career.

The following access techniques are available:

• Veress needle (closed entry), whereby a hollow needle with a retractable blunt tip is passed across the abdominal wall. Upon transperitoneal entry, the blunt cannula springs back into place, theoretically blunting the needle tip. Insufflation is then carried out through the needle.
  
• Hasson (open entry), whereby access is gained via a mini‐laparotomy approach and all layers are entered sharply [8].
  
• Optical trocar, a camera‐guided approach where the camera sits within the trocar’s transparent conical‐tipped obturator, and all layers of the abdominal wall can be seen as the trocar is advanced [9]. This technique can be employed alone in a desufflated abdomen or following initial insufflation via a closed (Veress) needle.
  
• Other techniques, less commonly used in urology, include radially expanding trocars and direct trocar insertion.

A 2008 Cochrane literature review comparing open and closed entry techniques found that there was insufficient evidence to recommend one technique over another [10]. No technique is foolproof. Experts acknowledge, however, that a prospective study powered to detect a meaningful difference would require thousands of patients and is unlikely to be organized. Generalized underreporting of complications further hampers attempts to define the true incidence of entry complications. However, a careful review of smaller studies reveals some take‐home points. The incidence of vascular injury is probably higher with closed techniques [11]. Bowel injury is equally likely with open and closed entry techniques [12], but there may be a higher chance of immediate diagnosis with an open technique, since the injured bowel is directly visualized. Even so, delayed diagnosis of bowel injury has occurred following open access as well [13].

The nuances of safe access are highly subject to individual surgeon preference. As such, evidence‐based literature only goes so far, and there is no substitute for hands‐on bedside training. In our opinion, for instance, Veress needle safety checks, such as aspiration and saline drop, can provide valuable clues when performed correctly and in conjunction with verifying a low opening pressure (ideally < 9 mmHg). A published report on this matter, however, concluded that the ancillary needle checks provided little useful information apart from opening pressure [14]. The literature must be interpreted with caution in this arena, since the most important factor in gaining safe access is surgeon familiarity and mastery of a given approach.

Flexibility to use a different technique when called for is crucial. When the surgeon encounters difficulty gaining access by any approach, the advice to change something is prudent. A quote widely attributed to Einstein captures this concept: “The height of insanity is repeating the same thing over and over and expecting a different result.” Stubborn repetition of the same maneuver invites complications. In a difficult access situation, careful reassessment of landmarks, checking for faulty equipment, and evaluating approach angles, can make all the difference. The trocar design must not create false reassurance, as even shielded trocars have not been shown to decrease access injuries. Also, be aware that in obese patients, reliable surface landmarks, such as the association between the umbilicus and aortic bifurcation, are lost.

The preventative guidelines in Box 87.1 are not data driven, but rather based on our experience. They are applicable for all types of laparoscopic access.

Physiologic complications

Pneumoperitoneum is associated with potential complications and physiologic alterations both from its mechanical compressive effects on blood vessels and viscera, and by its potential entry into the bloodstream and subcutaneous tissues. Carbon dioxide is utilized because it is widely available, noncombustible, colorless, and readily absorbed. An unfortunate by‐product of its favorable diffusion properties is the potential to decrease blood pH with prolonged absorption. Patients with severe pulmonary disease in whom the ability to counteract acidosis with hyperventilation is impaired may have a relative contraindication to laparoscopy [15], but this is rarely encountered clinically.

Although other pulmonary changes are best considered expected alterations rather than complications, they warrant mention nevertheless. Metabolic acidosis from hypercapnia can typically be measured in the form of expired CO₂. With mechanical ventilation, the clinical consequences are typically insignificant [16]. The mechanical compressive effects of pneumoperitoneum on the diaphragm causes increased intrathoracic pressure, increased peak airway pressure, and decreased vital capacity.

Hemodynamic effects of pneumoperitoneum are variable and difficult to generalize [16], but a prospective study by Meininger et al. thoroughly assessed this question by studying 10 patients undergoing robotic prostatectomy using a central venous catheter in place [17]. They were able to accurately measure cardiac output using thermodilution, cardiac index (CI), heart rate, mean arterial pressure (MAP), systemic vascular resistance (SVR), and central venous pressure (CVP), and correlate these with Trendelenburg position as well as pneumoperitoneum. The only hemodynamic parameter associated with Trendelenburg was an increase in CVP. Pneumoperitoneum only resulted in an increased MAP. Importantly, cardiac output was not affected either by Trendelenburg position or by pneumoperitoneum. The reader should note that other studies have demonstrated conflicting results in this arena.

Two important clinical scenarios are frequently encountered. First, the patient is found to have profound bradycardia as an initial response to peritoneal insufflation. The abdomen must be swiftly desufflated and anticholinergics administered. If the physiologic response occurs during a closed (Veress) entry technique, rapid desufflation is not immediately possible, as this can only be carried out after primary trocar placement. In this difficult scenario, the surgeon must proceed rapidly without placing the patient at risk from overly aggressive trocar insertion. Attention to safe trocar placement is well worth a few extra seconds of bradycardia.

Second, if an air embolus is suspected, the often cited Durant maneuver [18] is suggested to prevent the “air lock” from sitting in the right ventricular outflow tract. In the original study, animals that were positioned left side down had improved hemodynamics. In this position with the right side up, and furthermore in Trendelenburg, theoretically the air is prevented from entering the pulmonary outflow tract. A central venous catheter may be inserted to aspirate the air. Some authors have since pointed out that these effects have not been reliably reproduced in humans, and the position may be cumbersome and impractical in an emergency situation [19]. Nevertheless, the presentation of air embolus is dramatic, especially if Veress needle high‐pressure insufflation directly into a major vessel is the cause. The patient develops sudden hypoxia, hypercarbia, and cardiovascular collapse.

The overall incidence of subcutaneous emphysema is 0.4–2.3% [20]. Subcutaneous emphysema is typically clinically insignificant and resolves without specific intervention. It may occur with initial preperitoneal Veress needle insufflation. Alternatively, a combination of a snug skin incision and loose fasciotomy around a trocar allows gradual development of ongoing subcutaneous emphysema during surgery. During lower tract laparoscopy, subcutaneous emphysema in the scrotum and penis is not a subtle finding, and should be manually decompressed at the conclusion of the case.

The presence of subcutaneous emphysema underscores the ability of pressurized gas to diffuse between body compartments [20]. Gas may similarly enter the pleural cavity, mediastinum, and pericardial space. Small collections of air in these spaces, without clinical instability, can typically be managed conservatively since CO₂ is readily absorbed.

Complications of patient positioning

The surgeon often relies on gravity and patient positioning for adequate exposure during minimally invasive genitourinary procedures. Patient positioning, and at times prolonged operative times, predisposes the patient to four potentially painful and debilitating problems if precautions are not taken: neuropraxia/neuropathy, rhabdomyolysis, compartment syndrome, and pressure ulcers [15]. Complications in this arena are position‐ and therefore procedure‐specific. In a large series of urologic laparoscopic procedures compiled from 15 centers and over 1600 cases, the overall incidence of neuropraxia was 2.7%, including rhabdomyolysis in 0.4% [21]. Risk factors included obesity, longer operative times (>5 hours), and advanced patient age. In a postoperative patient with oliguria and myalgias, clinical suspicion of rhabdomyolysis is paramount, creatinine kinase should be checked, and the patient should be alkalinized expeditiously.

For upper tract laparoscopy, with the patient flexed in flank position, the following maneuvers are important: use of axillary roll, avoiding excessive flexion, and padding all pressure points. For lower tract surgery, with the patient supine with arms tucked and legs variably positioned per surgeon preference, hands should be in an anatomic position (thumbs up) and sheets to “tuck” arms must not be overly snug around the arm.

End‐organ complications

Vascular injury

Bleeding complications account for a substantial portion of laparoscopic complications. For example, intraoperative and postoperative hemorrhage accounted for 40% of all perioperative complications in a single institution analysis of urologic laparoscopic procedures from 1997 to 2006 [22]. The overall incidence of vascular injury during urologic laparoscopy is in the range 0.4–1.7% [1, 2].

The most feared vascular injuries carrying the highest chance of mortality when injured are the great vessels and their immediate branches: aorta, vena cava, and iliac vessels. Mesenteric vascular injury can also be a source of major bleeding. The most common scenario is injury by initial blind Veress needle puncture. This may initially manifest as blood aspirated from the needle. Alternatively, vascular puncture from primary trocar insertion can occur. In either case, possible causes include incorrect angle of insertion and/or uncontrolled axial insertion force. In case of injury, if secondary trocars can be swiftly inserted, holding pressure at the bleeding site with a gauze sponge and laparoscopic grasper (unrolled gauze will fit through a 12 mm trocar) usually gives the surgeon time to assess the situation and, if required, proceed with laparotomy in a controlled fashion [23].

Depending on the bleeding severity and origin, the surgeon must decide whether to use thermal energy, clips, or suture ligatures. With massive hemorrhage, immediate laparotomy is often required. One has precious seconds to determine whether initial laparoscopic attempts at control are feasible or futile. A hand‐assist port may be placed as an intermediary to open conversion. If open conversion is required, the laparoscope should be lifted up to the abdominal wall and rapid incision made directly over it. In such an emergency, the incision should be generous for maximal exposure. Compression of the bleeding site allows time for anesthesia to catch up with resuscitation, and allows surgeons with appropriate expertise, such as vascular surgery, to join the team.

The epigastric vessels are the most commonly injured minor blood vessels during trocar or Veress needle insertion [24], when these are placed laterally off midline (Figure 87.1). Using a Carter–Thomason wound closure device [25], sutures placed above and below the bleeding site are usually sufficient to control the bleeding.

A vascular complication specific to retroperitoneal laparoscopic right‐sided nephrectomy is inadvertent ligation of the inferior vena cava (IVC) mistaken for the renal vein [26]. The cause is a lack of orientation to anatomic landmarks. During retroperitoneoscopic surgery, the psoas muscle must be maintained as the “surgical horizon” to maintain orientation. The renal vein must be verified to be heading towards the kidney (90° up, if the psoas horizon is maintained), with expected caliber. Most importantly, the renal vein–IVC junction above and below must be confirmed, as a prerequisite to renal vein ligation. If IVC transection is recognized (Figure 87.2), open vascular repair is suggested.

Bleeding is the main complication following partial nephrectomy (1.3–4.3%) [27]. One of the more common causes of intraoperative bleeding is an unclamped, unrecognized accessory renal artery. Great care must be taken both in evaluating preoperative axial imaging and during intraoperative hilar dissection to detect such variations. Delayed bleeding in the setting of hemodynamic stability can be addressed with selective embolization. In cases of hemodynamic instability, re‐exploration is warranted. An attempt to ligate bleeding vessels is reasonable, but nephrectomy may be indicated if a single bleeding site is elusive or patient instability does not allow for renorrhaphy. In a large series of laparoscopic partial nephrectomy, reoperation was required in 2% and elective laparoscopic radical nephrectomy was performed in 0.5% [28]. When delayed bleeding manifests as hematuria, vascular fistula should be considered, and can be managed effectively with endovascular embolization.

As with any type of injury, prevention of vascular injury is best. A vascular injury “kit” should be prepared with each case (Box 87.2). Dissection should be performed in layers, widely completing more superficial layers before progressing deeper. Vascular injury of varying severity will occur, but poor exposure and “working in a hole” at the outset will severely limit the ability to control an injury. If staplers are used for vascular control, the stapler should be applied a few millimeters distally to provide a stump to grasp in case of catastrophic stapler malfunction [29].