Paediatric robotic surgery

Abstract

A review of the current status of robotic surgery use in paediatrics

Paediatric robotic surgery offers unique challenges within this rapidly advancing field. The financial costs forproviding this advanced technology are significant for what are essentially low-volume complex cases. There has been a slow rate of uptake within most paediatric surgical centres around the world.due to both to finance as well difficulties associated with equipment primarily designed for adults. Consequently there has been a slow rate of uptake within most paediatric surgical centres around the world. The ergonomics required for the da Vinci® master–slave-type platform currently challenge the small working space in very small children. Only two centres in the UK currently offer robotic surgery for children.^* This article aims to review the current state of robotics within the field of paediatric surgery and surgical subspecialties.

Paediatric urology

Urology has arguably seen the greatest uptake of robotic surgery within paediatrics. One of the first uses of robotics in children was a (dismembered Hynes-Anderson) pyeloplasty for pelviureteric junction obstruction (PUJO), because the ureterico-pelvic anastomosis was a significant technical challenge using conventional laparoscopic surgery.¹

Robotic-assisted pyeloplasty (RAP)

A meta-analysis of RAP in 2014 examined 12 retrospective studies comparing open pyeloplasty (OP) with RAP and laparoscopic pyeloplasty (LP) with RAP.² Darzi’s group found a tendency towards better RAP success rates in comparison with LP; although when compared with OP, RAP was equivalent. There was no difference in rates of re-operation and complications between groups. However, OP was statistically significantly faster and cheaper than RAP – although it resulted in a significantly longer hospital stay (+1 day). The studies included were limited by retrospective technique, variable methods of group selection and inconsistent comparison of outcomes, which limit their own conclusions. Subsequent to this, two large multi-centre comparison of OP, LP and RAP^3,4 support Cundy’s findings that RAP demonstrated a high success rate and resulted in a significantly shorter hospital stay.^3,4 There was also some evidence that there is a lower rate of complication amongst RAP. However, Chan et al’s larger study found no difference in rates of complications.

Ureteral re-implantation⁵

The standard operative approach for ureteral re-implantation is via an open approach,⁵ although there is some evidence of comparable success in RAUR. Kasturi et al achieved VUR resolution with a robotic approach in 99.3% of cases⁶ and this has been supported in a comparison study finding equivalence of RAUR (97%) with open technique (100%),⁷ as well as in a case matched series that also examined an intra-vesical technique.⁸ Marchini et al conclude that there is no significant difference in postoperative complications; however, it is worth noting all significant complications – urinary retention, bladder leak and ureteral leak are within the robotic arm of their comparison. A higher rate of bladder spasm and haematuria in the open group account for an overall finding of no difference in complication rates.⁸ One RAUR was readmitted with a ureteric leak.⁶ In line with other minimally invasive surgery, there does appear to be a reduced length of stay and postoperative pain.^7,9

Ureteroureterostomy

Ureteroureterostomy is performed in a number of indications as the primary procedure (duplex systems with an upper pole ectopic ureter, obstructed ureterocele, etc) and there are several case series reporting successful robotic-assisted ureteroureterostomy.^10–12 Most recently, a small comparison of robotic-assisted ureteroureterostomy with open ureteroureterostomy concluded: ‘Operative times and complication rates were comparable with slightly shorter length of hospitalisation in robotic cases.’¹³

Mitrofanoff and reconstructive bladder surgery

The appendicovesicostomy (Mitrofanoff) has been established as feasible to perform with robotic assistance by a number of case series and in 2015 Grimsby et al compared the open and robotic approaches. They found no difference in rates of complication (26% vs 29%); however, the severity of complications in the robotic group was clinically greater 7.6% (Clavien>3). Furthermore, rates of continence were just 90% when compared with 97% in the open group.¹⁴ Similar results are seen in larger multi-centre study of reconstructive surgeries involving Mitrofanoff.¹⁵ Grimsby et al attribute these issues to technical adaptation of the operation to robotic platform. Other complex reconstructive surgery has also been examined in a useful, small comparison of open and robotic bladder augmentation (and associated procedures) by Murphy et al, who found similar functional outcomes and rates of complication but with significantly reduced length of stay.¹⁶

Those surgeries that must access the pelvis and therefore have a narrow field may well suit a robotic approach

Miscellaneous

Paediatric urology procedures without a reconstructive element have adapted well to traditional laparoscopic approach; nevertheless, there is extensive literature documenting the use of robotics in hemi-nephrectomy¹⁷ and nephroureterectomy.¹⁸ Those surgeries that must access the pelvis and therefore have a narrow field may well suit a robotic approach. Indeed, there are reports of robotic-assisted excision of bladder diverticulum,¹⁹ urachal cyst excision,²⁰ excision of posterior urethral diverticulae,²¹ prostatic utricles,²² seminal vesicle cyst,²³ and varicocele.²⁴

Paediatric general surgery

As in adult surgery, inguinal hernia repairs are commonplace in paediatrics, although they are performed usually through a smaller open inguinal incision. The paediatric laparoscopic hernia repair is also far less involved than its adult counterpart and does not use a mesh, making robotic assistance an unnecessary technical addition in its current format. Other more complex procedures have been carried out robotically.

Fundoplication

A meta-analysis in 2014 reporting outcomes of 297 children²⁵ found that despite a tendency towards conversion to open surgery in the laparoscopic fundoplication (LF) group (6.1% vs 3%), there was no significant difference in postoperative complications (RF 8.9% vs 8% LF) found. In one study the most common complication in the RF and LF was a tight wrap, requiring dilatation (8% and 6%), whereas in the open series wound infections were more common (4%). The meta-analysis was limited as most studies were retrospective with non-synchronous selection of controls, but perhaps most significantly limited by their absence of long-term follow-up of success.

Figure 1.

Paediatric robotic procedure

Figure 2.

Robotic pyeloplasty (Courtesy of Nisha Rahman, Chelsea and Westminster NHS Foundation Trust)

Hepatobiliary surgery

HPB surgery in children inevitably involves intricate and demanding MIS procedures. Choledochal cyst excision and reconstructive Roux-en-Y hepaticoenterostomy are technically complex and, with the exception of centres in South East Asia,²⁶ open procedures are still relatively prevalent. The laparoscopic technique often involves extending the umbilical incision to allow extra-corporeal anastomosis. Meehan et al describe a robotic approach outlining how additional degrees of freedom offered by the robot conferred a real advantage;²⁷ a view shared by others with experience in the area.²⁸ This approach has also been repeated in small infants (<10kg) (although they use an extracorporeal anastomosis)²⁹ and by the same group in a larger series;³⁰ within this series they converted 19% of their cases, although only 1 patient had any complications. A similar rate of conversion is also seen in another case series,³¹ which also used extracorporeal anastomosis for the Roux-en-Y loop. Recent evidence,³² however, suggests that laparoscopic Kasai portoenterostomies may have significantly worse outcomes than an open approach. This may reduce enthusiasm for further robotic work.

Miscellaneous

Robotic-assisted cholecystectomies and splenectomies are relatively prevalent in the literature.^33–38 However, all authors emphasise that – although these are useful training opportunities in the robot platform – neither robotic-assisted splenectomy nor cholecystectomy seem to offer additional benefit over the laparoscopic approach. Indeed, there is no comparative research in the field.

There are also case reports and series that document a diverse array of successful robotic general and gynaecological surgery such as robotic-assisted diaphragmatic hernia repair,^39,40 Heller’s cardiomyotomy for achalasia,^41,42 duodenojejunostomy for SMA syndrome,⁴³ repair of duodenal atresia,⁴⁴ anorectal pull-through for anorectal malformations,³⁸ ovarian cystectomies and salpingo/oophorectomies.⁴⁵ Further study is needed to assess whether these procedures are indeed effective and whether they confer any benefit above traditional minimally invasive surgery (MIS).

Paediatric cardiothoracic surgery

Cardiac

The surgical interruption of a patent ductus arteriosus (PDA) in robot-assisted thoracoscopic surgery (RATS) vs video-assisted thoracoscopic surgery (VATS) has been described.⁴⁶ The authors, however, concluded that the VATS procedure is relatively straightforward and that no advantage is offered by RATS, especially allowing for increased complexity of setup and cost.⁴⁶ One study since examined robotic cardiac surgery specifically in children,⁴⁷ involving the division of congenital vascular rings, and found RATS comparable but did not offer further advantage over VATS.

Mediastinal

RATS has limited further examination in current literature. The largest series reports on 11 cases including mediastinal cyst excision, diaphragmatic hernia repair, Heller’s myotomy, oesophagoplasty and oesophageal atresia repair via RATS.⁴⁸ There were several conversions to open surgery in neonatal patients.⁴⁸ The small neonatal thorax represents the greatest obstacle in adapting the large 5 or 8mm instruments of most robotic platforms into paediatric surgery, and the authors conclude that RATS seems only appropriate in patients with a weight >20kg.

Paediatric oncological surgery

Despite widespread use of MIS in adult oncological surgery and in non-oncological paediatric surgery, open surgery is the usual standard of care for resection of paediatric solid tumours. Paediatric oncological MIS and robotic assistance is a relatively recent development that is lacking high-level evidence,^49,50 although there is a wide range of case literature.

Thoracic

Anecdotally, the robot seems well adapted to intricate mediastinal dissection and has been used in the excision of left ventricular myxoma⁵¹ and removal of complex massive oesophageal leiomyoma.⁵² There is support for the robot’s applicability to the mediastinum in a larger case series.⁵³ There is also a relatively large case series that demonstrated in neurogenic chest tumours: ‘Resection [R0] can be as complete as an open procedure without having to complicate the operative technique in the same operating time.’ And the authors felt that ‘the surgeon has a better visualisation of the tumour and its anatomic connections’.⁵⁴

Abdominal

There are mostly individual case reports for robotic-assisted abdominal oncological surgery in children. Excision of juvenile cystic adenomyoma,⁵⁵ a radical cystoprostatectomy,⁵⁵ partial nephrectomy for RCC,⁵⁷ retroperitoneal lymph node dissection⁵⁸ and partial adrenalectomy for pheochromocytoma.⁵⁹ A common theme discussed by many of the authors is of the suitability of the robotic approach to extended lymph node dissection; indeed, one asserts that ‘vision was excellent throughout the procedure, which is very important during dissection and not always the case during open surgery.’⁵⁶ The solitary case series supports these claims and the authors experienced no complications and achieved R⁰ resection in all.⁶⁰

There is some dispute as to whether the fundamental oncological principles of no tumour spillage and total resection of tumour margins can be adhered to by robotic-assisted surgery; a specific concern being the lack of haptics having an impact on the surgeon’s ability to differentiate cancerous from healthy tissue. However, it has been noted that ‘the loss of tactile feedback is, in our opinion, very well compensated for by the excellent optical system’.⁵⁶ Cancer patients are necessarily followed up for recurrence and only prospective long-term studies of robot resections can give assurances of robotic adherence to oncological principles.

Paediatric neurosurgery

Computer technology has long been used in neurosurgery, specifically in ‘image-guided’ surgery and surgical planning. This has been applied to robotics in the form of the robotised stereotactic assistant (‘ROSA’).⁶¹ Robotic-assisted paediatric neurosurgical procedures are sparsely reported, although there is a recent large case series of an array of procedures, making up 128 cases. The authors demonstrated a high rate of success (97.7%) and low rate of complications (3.9%) and no incidences of permanent neurological deficit.⁶¹ Such results in a diverse case series is encouraging for the safety and utility of robotic assisted paediatric neurosurgery. Indeed, as is noted in previous case reports, ‘precision of movement millimetre resolution make robotic tools highly attractive in the treatment of intracranial lesions’.⁶²

The future of robotics in paediatric surgery

With advancing technology and the demand for more compact robotic platforms, the future for robotic surgery will no doubt result in a reduction of instrument size and an improvement in haptic feedback. This puts the paediatric patient – in particular, the newborn – at the forefront. Reconstructive surgery such as oesophageal and intestinal anastomosis, all of which require a delicate and more magnified approach will benefit enormously from these advances. The financial restraints that exist in public health systems currently restrict the advancement and training of many in robotics. With a more expanded competitive market, this should improve the landscape but in the interim the paediatric and neonatal patient must be at the forefront of research into the future of robotic surgery.