Abstract
Abstract
Introduction:
Since the use of robotic surgery (RS) revolutionized some adult surgery procedures such as radical prostatectomy, it has been progressively and increasingly introduced in pediatric surgery. The aim of this study is to evaluate how the Da Vinci® Si HD technology impacts a pediatric public hospital and to define the use of a robotic system in pediatric surgery.
Materials and Methods:
We prospectively included patients older than 6 months of age undergoing RS or conventional minimal access surgery (MAS): Study period ranges between February 2015 and April 2016. Surgical indications were defined after a detailed disease-specific diagnostic work-up. We analyzed surgical outcomes and the most relevant economic aspects. The 30-day postoperative complications were evaluated and retrospectively collected in an electronic database.
Results:
From February 2015 to April 2016, we performed 77 procedures with RS and 84 with conventional MAS in patients with a median age of 77 and 98 months at surgery and a median weight of 20 and 23 kg, respectively. Median operative times were 130 and 109 minutes, respectively. We observed 9.1% of complications in the RS group and 6% in the MAS group and the difference was not statistically significant. Of note, 8 out of 77 RS procedures would have been performed with open classic surgery in case of conversion or failure of RS.
Conclusions:
This initial experience confirms that RS is as safe and effective as conventional MAS. A number of selected procedures performed with RS would only benefit from this approach, as it is not suitable for conventional MAS. Although economically demanding, in particular for a pediatric hospital, we firmly believe that centralization of care would allow pediatric surgeons adopting RS to perform complex reconstructive surgical procedures with great advantages for the patients and a minimal increase in overall costs for the health system.
Introduction
S
Thanks to three-dimensional visualization, articulated instrumentation, intuitive movements, tremor reduction, motion scaling, and surgeon ergonomics, RS improves both dissection accuracy and feasibility of intra-corporeal knotting required in case of complex procedures without increasing the surgeon's distress.10,11 Furthermore, RS maintains other advantages of classical and well-established minimally invasive surgery (minimal access surgery [MAS]), including improved cosmetic results, reduced post-operative pain, and shorter hospital stay. 12
Currently, RS has a number of drawbacks that must be overcome before it can be routinely adopted in pediatric surgery, including high costs and limited re-usability of instrumentation, need for a specific training, and lack of tactile feedback.13–16
The vast majority of European Pediatric Surgical Units involved in a robotic program are located in a general hospital, and pediatric utilization involves a minority of the overall performed procedures as reported by Anderberg et al., with pediatric surgical procedures accounting for 4.2% of the total. 17
There are several ways to get a robotic system in a hospital. This technology can be purchased, rented, or acquired thanks to research funding. However, once it has been introduced in the hospital, the overall expenses for the instrumentation employed for each usage remain a significant issue to be addressed. This study is aimed at evaluating how the Da Vinci Si HD® technology impacts a pediatric public third-level hospital and at validating the hypothesis of a positive role of RS in pediatrics. Our study will improve the knowledge regarding feasibility, organization, and convenience of RS in a pediatric hospital. It will also demonstrate the versatility of RS in treating well-known urogenital issues as well as gastrointestinal, oncological, and thoracic malformations and acquired diseases in all fields of pediatric surgery.
Materials and Methods
Setting in the hospital
The introduction of the RS in our pediatric surgical unit with longstanding experience in MAS required a preliminary training period of 30 days, with simulation of the layout of the operating room and on-site surgeon's training on the simulator. Site visits to other adult centers were also required to observe RS sessions directly on the patient. A dedicated staff of surgeons and nurses was formed, and the first RS procedures were performed with the supervision of tutors who were dedicated to the disease addressed in each case (general surgeons for gastrointestinal diseases and urologists for genitourinary disorders).
Study period, inclusion and exclusion criteria, and addressed procedures
Study period ranged between February 2015 and April 2016 (15-month period). Patients younger than 6 months of age were excluded, mostly due to the limited working space provided by the available robotic system (8–10 cm represents the minimal distance between ports for our device). Surgical indications were defined according to detailed disease-specific diagnostic work-ups. The choice to adopt “classical” MAS or RS was not randomized but based on the surgeon's attitude and instrumentation availability. One single surgeon (G.M.) performed all RS. Procedures were divided into gastrointestinal, urological, thoracic, oncologic, and hepato-biliary.
Informed consent and Institutional Review Board
Benefits, risks, and expectations related to RS were explained in detail to the parents by the same pediatric surgeon. The parents in all cases signed a detailed and tailored informed consent. Given the observational nature of this study, Institutional Review Board Ethical Committee approval was not requested. Moreover, indications and surgical techniques were those widely approved and adopted by international disease-specific treatment guidelines.
Case–control study and assessed variables
Procedures were grouped according to the adopted approach. The RS procedures (robotic minimally access surgery [RMAS] group) were compared with similar procedures performed with the standard MAS technique (no-RMAS group) in the same study period. We prospectively recorded details regarding patients' demographics (age and weight at surgery), surgical details, post-operative pain using FLACC score 18 , length of procedures, docking time, and surgeon console time (the latter two items only for RMAS group). All intraoperative and 30-day postoperative complications were retrospectively collected with note revision and recorded according to the classification of Clavien-Dindo. 19 The economic analysis included expenses for operative room occupation, instruments consumption, draping, sterilization, and trocars (if disposable). The expenses for purchase and maintenance of both the system and all instruments were obtained, thanks to the data provided by the Institutional Pharmacy Department. Diagnosis-related group (DRG) as determined by the Health System and relative procedure-specific weight 20 were calculated for each procedure.
All data were recorded and stored in a digital database according to the Data Protection Act.
Statistical analysis
Descriptive statistics were reported as percentages. Median and range or mean and standard deviation were used according to the variability in our series. Differences in the frequencies of each categorical variable were evaluated by the chi-square, fisher exact tests or with a t test in case of a comparison of means. A comparison of categorical data was performed by using the fisher two-tailed unpaired test. A P value lower than .05 was considered statistically significant.
Results
Overall and demographics
A total of 413 pediatric patients underwent 432 MAS procedures during the study period, 77 of which were performed with an RS approach (RMAS) (40 men, 37 women). According to our inclusion criteria and addressing similar or overlapping procedures, 84 out of 355 MAS procedures were included in the no-RMAS group. (See Table 1 for details.)
In the RMAS group, we performed a total of 77 procedures in 76 patients (32 urologic procedures, 29 gastroenterologic procedures, 7 oncologic procedures, 6 thoracic procedures, and 2 hepato-biliary procedures). In the no-RMAS group, we performed a total of 84 procedures in 80 patients (31 gastroenterologic procedures, 23 urologic procedures, 21 hepato-biliary procedures, 6 thoracic procedures, and 3 oncologic procedures). Pieloplasty (pyeloureteric surgery) and cholecystectomy were the most common procedures in the RMAS group (15/77; 19.5%) and no-RMAS group (21/84; 25%), respectively. Eight out of the seventy-seven procedures performed with RS would have been performed alternatively with open conventional surgery given the features of working space and demanding dissection required for safety and effectiveness.
MAS, minimal access surgery; RMAS, robotic minimally access surgery; RS, robotic surgery.
Median age at surgery was 77 months (range 8–288 months) and 98 months (range 6–403 months) for RMAS group and no-RMAS group, respectively. Median weight was 20 kg (range 7–82 kg) and 23 kg (range 5–89 kg) for RMAS and no-RMAS groups, respectively.
Procedures
We observed a prevalence of urological procedures (41.5%) in the RMAS group and gastrointestinal procedures in the no-RMAS group (37.3%). However, the prevalence of specific subsets of procedures, divided as stated earlier (see Table 1 for details), did not significantly differ between the two groups, with the exception of hepato-biliary procedures that were included in the RMAS group in only 2 out of 77 cases, with the difference being extremely statistically significant compared with the no-RMAS group (P = .0001).
In the RMAS group, median operative time was 130 minutes (range 30–525), median console time was 65 minutes (range 5–250), and both median docking and undocking time accounted for 10 minutes (range 5–30 minutes and 4–10 minutes, respectively). Median hospital stay was 5 days (range 1–33). In the no-RMAS group, median operative time was 109 minutes (range 25–425) with a median hospital stay of 4 days (range 1–20). Comparing operative time in the two groups, we could observe that no-RMAS procedures lasted significantly less than did RMAS procedures (P = .0030). Comparing hospital stay in the two groups, we could not observe any statistically significant difference (P > .05).
Each subset of procedures in the RMAS group was specifically analyzed, and results with anesthesiological (preoperative and postoperative time) and surgical details (surgical time, docking time, console time, and length of hospital stay) are shown in Table 2.
All variables are expressed as median and range.
These data show the differences among the different subgroups of surgery. Thoracic and oncology subgroups have a higher median age and weight compared with the other subgroups. FLACC and NRS scales confirm a good pain control with RS.
FLACC, face, legs, activity; Cry, consolability; Nr, numeric rate; NRS, numeric rate scale; ASA, American Society of Anesthesiologists.
RMAS group (Table 1)
Urological procedures—conversions and complications
A total of 32 urologic procedures were performed with RS in the study period. We experienced one conversion to laparoscopy in a year-old male infant scheduled for redo pyeloplasty, due to difficulties with instruments insertion during the docking phase. One conversion to open surgery was required to complete the isolation and identification of a residual ureter in a girl previously treated in another institution for a duplex collecting system.
No intraoperative complications occurred. We experienced one postoperative complication: The same patient as described earlier experienced omental herniation during drain removal on postoperative day 7. This issue was resolved with bedside sedation (Grade IIIa of Clavien-Dindo Classification).
Gastrointestinal procedures—conversions and complications
A total of 30 gastrointestinal procedures were performed with RS in the period of study. No conversions either to laparoscopy or to open surgery were required.
No intraoperative complications occurred. We experienced four postoperative complications. One 3-year-old male patient with anorectal malformation experienced anastomotic leak on postoperative day 5 that required diverting ileostomy (Grade IIIb—Clavien-Dindo Classification). A girl treated for left para-retrorectal cyst experienced a residual persistence of a small part of the cyst but she only required repeated imaging and radiological follow-up (Grade I of Clavien-Dindo Classification). A 12-month-old male infant who underwent Nissen fundoplication and gastrostomy experienced wound infection and leak on postoperative day 6 requiring surgical revision under general anesthesia (Grade IIIb of Clavien-Dindo Classification). A 7-year-old male patient who underwent esophagogastric disconnection after two failed laparoscopic Nissen fundoplications experienced a colonic perforation requiring reoperation, colonic suturing, and diverting ileostomy (Grade IIIb of Clavien-Dindo Classification).
Thoracic procedures—details and complications
A total of six procedures were performed with RS during the study period.
We experienced two conversions to open conventional thoracotomy. One was due to difficulties in defining the consistency of a bronchial carcinoid during an inferior right lobectomy; the other was due to the presence of extensive adhesions of the lung preventing the possibility to inflate the thorax and start RS for left superior lobectomy.
We experienced one intraoperative complication. Massive bleeding was experienced by a 15-year-old boy who required emergent splenectomy as a consequence of diaphragmatic and splenic trocar-related injury (Grade IIIb of Clavien-Dindo Classification) during resection of a thoracic neuroblastoma. No postoperative complications were experienced.
Hepato-biliary procedures—details and complications
A total of two hepato-biliary procedures were performed with RS during the study period. No conversions were needed. We did not experience any intra- or postoperative complications.
Oncology procedures—details and complications
A total of 11 oncologic procedures were performed during the study period, four of which were already included among thoracic procedures.
We experienced two conversions to open surgery. One 12-year-old girl required conversion due to uncontrolled bleeding during nephrectomy for angiomyolipoma of the right kidney. A boy scheduled for excision of a peri-hepatic formation required conversion to obtain a better definition and isolation of the mass. We also experienced one conversion to laparoscopy due to difficult vascular management during resection of a left adrenal neuroblastoma with image-defined risk factors (IDRFs).
No intraoperative complications occurred. We experienced one postoperative complication in a 15-year-old girl who underwent excision of a pseudopapillary tumor of the pancreas and experienced postoperative omental bleeding and pseudocyst formation in the long term. Both issues required reiterative laparoscopy and endoscopic gastro-cyst drainage, respectively (Grade IIIb of Clavien-Dindo Classification).
Overall morbidity and outcome
A total of seven conversions (five to open surgery and two to conventional MAS) and seven complications were reported in the RMAS group (9.1%). During the same study period, six conversions were reported in the no-RMAS group (7%), in which we observed four major complications (5%), two of which were related to equipment malfunction. Comparing the incidence of conversion and that of complications in the two groups, the differences turned out to not be statistically significant (P = .7748 and .3546, respectively).
Operative time in the RMAS group lasted 147 ± 96 minutes whereas that of the no-RMAS group lasted 109 ± 62 minutes, with the difference being very statistically significant (P = .0036). Similarly, operative room occupation in the RMAS group lasted a mean of 237 ± 114 minutes whereas that of the no-RMAS group lasted a mean of 168 ± 79 minutes. The difference proved to be very statistically significant (P = .0001). Comparing data available from the no-RMAS group with those reported in the RMAS group (Table 3), RS proved to be significantly more expensive than conventional MAS (P < .0001), with a median difference of 3587 € (range: 1103–5461) for procedures performed during the study period. The median value of DRG of the procedures included in the RMAS group was 6185.5 € (range 1442–16,715). The median of those included in the no-RMAS group was 5950 € (range 1312–16,552 €) (P < .002).
DRG is the refund for hospital costs and is calculated for each patient based on the Italian Ministry of Health system. Relative weight is specified for each procedure and represents a measure of complexity of each procedure. For example, refunds in case of cholecystectomy with the robotic approach are relatively low and prove to be inadequate, thus not supporting the use of this approach.
DRG, diagnosis-related group.
Discussion
Since the first report by Meininger et al. in 2001, 21 a number of experiences regarding RS in pediatric surgery were published.10,22–28 At present, although it is difficult to define both applicability and reproducibility of RS in children, it is clearly evident how easy it is to shift from laparoscopy to RS. In particular, most surgeons proved to appreciate this change from classic open laparotomic/thoracotomic surgery to RS, thanks to the overcoming of some major limits of “classic” laparoscopy.
Symmetry of information is anyway mandatory. In fact, although RS is more expensive than conventional laparoscopy, Hassan et al. 29 recently demonstrated that RS is easier to learn and more intuitive when compared with “conventional” laparoscopic surgery for naive users, thus allowing a wider applicability of MAS. 30
Certain anatomic structures such as the deep pelvis or the esophagogastric junction at the diaphragmatic crura can be particularly difficult to address and technically demanding when approached with conventional laparoscopy due to very narrow working spaces, particularly in newborns, toddlers, and youngsters. This is the case of restorative proctectomy with J-pouch ileo-rectal anastomosis or esophagogastric disconnection that are usually among the most difficult procedures in pediatric surgery. In fact, stomach manipulation required for the creation of the retrogastric space proves to be straightforward with RS. The same applies to the “hand-sewn” esophago-jejunual anastomosis that becomes feasible, reliable, and safe, particularly if compared with what is experienced in case of conventional laparoscopy. Similarly, rectal dissection can be easily carried down to the levator ani with minimal traumatism of perirectal structures with RS. This implies a subsequently reduced likelihood of continence impairment.
Furthermore, RS proved to be particularly suitable for oncologic surgery, given the reduced need for mobilization and manipulation of tissues together with a sharper and more precise dissection permitted by the increased degrees of freedom of robotic arms, observed during tumor excision.
Many criticisms were raised more than 20 years ago when MAS was being introduced in pediatric surgery, including difficult tissues manipulation in narrow spaces, discrepancy between the size of the child and that of available instruments, and higher costs compared with traditional open surgery.31,32 Nowadays, RS is experiencing the same hurdles and skepticisms.
Although many studies regarding economic aspects related to the use of RS in the adult population have been already reported, this issue has not yet been adequately examined in pediatric patients. Mahida et al. recently reported data from a series of patients belonging to a tertiary-care children's hospital who were operated on with RS and showed increased costs without significant differences in terms of postoperative length of stay or pain control compared with “conventional” laparoscopic surgery. 33 Similarly and not unexpectedly, in our series of patients, RS procedures proved to be significantly more expensive with differences as much as 5000 € and higher. Conversely, we recorded higher DRG refunds, given the increased complexity of our series of patients who were approached with RS. In fact, we experienced an improved confidence and attitude to resort to RS for procedures that are not routinely approached with conventional MAS. An example is represented by neurogenic tumor excision with IDRFs that have been approached and excised with RS. 34
When coming to the safety issue, we must address complication and conversion rates. The rate of early and intraoperative complications of our series of patients resembles that of previous publications that reported an incidence of complications ranging between 2.5% and 8%.15,35 The same applies to conversions that have been reported to occur with an incidence ranging between 5% for abdominal RS 36 and as much as 30% for thoracic RS. 37 Our series of patients confirm that gastrointestinal reconstructive and urological procedures can be completed with RS in most cases (overall 8% conversion rate) whereas thoracic RS and complex oncologic procedures such as those with IDRFs are those with the highest likelihood of conversions (nearly 24%) and complications (roughly 12%), with an incidence that does not significantly differ from that of previously reported open surgical series.38,39
Surgery duration interferes with both costs and safety with regard to theater occupation time and anesthesia duration. Our series of RS lasted significantly longer, but this issue can be mostly related to the increased complexity of case load included in the RS group. As a consequence, given the heterogeneity of the two groups, a realistic comparison is not feasible. More likely, the setting of the robotic system could interfere with the overall operative time by prolonging the overall duration of surgery by 7–40 minutes for each procedure, which is acceptable in our personal opinion.
The Society of American Gastrointestinal and Endoscopic Surgeons's (SAGES) Technology and Value Assessment Committee (TAVAC) 40 has compared RS with conventional MAS in the adult population in different surgical areas (foregut, colorectal, hepatobiliary, and pancreatic surgery), analyzing safety, efficacy, and costs. Although the authors found RS to be as safe and effective as conventional MAS, they underlined that the increased costs must be taken into maximal consideration when resorting to such approaches. 40 On the other hand, Curet, 41 scientific consultants of Intuitive Surgical, Inc., brought some criticism to the SAGES assessment in terms of modality of comparison between robotic and laparoscopic surgery, adoption of exclusion criteria for analyzed studies, and modality of cost comparison. Of note, one of the key points raised by Curèt was that RS should be compared with open surgery and not with conventional MAS, given the procedures that are amenable of the RS approach. 41 Although our study compared RS with conventional MAS, as done by the SAGES TAVAC, we must agree with Curet that some of the procedures that we performed with RS are the most likely approached with open laparotomy/thoracotomy. This is confirmed by our series of patients in whom endorectal pull-through for adult Hirschsprung's disease, IRDF's positive neurogenic tumor resections, redo esophagogastric disconnections, and other procedures would have been approached with conventional surgery. Besides, we could not find procedures that would be better approached with open conventional surgery based on the incidence of complications.
On the ground of these considerations, we suggest that in certain instances RS should represent the true alternative to open conventional surgery and that the comparisons for both costs and outcome analysis should be performed between these two approaches. Of note, 8 out of the 77 RMAS procedures reported in our series (roughly 10%) would have been approached with open conventional surgery (see Table 1) and this subgroup of procedures should be considered as those that would benefit the most from RS in terms of both feasibility and safety. Noteworthy, short-term outcomes of these latter procedures resemble those of open conventional surgery, with the well-known added advantages of minimally invasive approaches (no intraoperative complications, no need for conversion, reduced risk of adhesions, etc).38,39
One of the key goals of every new and advanced technology is represented by the improvement of quality of care, taking into account the initial unavoidable increased costs. Conventional MAS instruments are not disposable and are being re-used many times, with a reported instrument-malfunction rate of 17%. 42 On the other hand, RS disposable instruments can be reused for a median of ten utilizations for each instrument (forceps, needle driver, hook, etc.) and have a reported rate of malfunction and/or failure of 0.5% of either robot or instrumentation. 15 Although the malfunction rate based on available literature data seems trivial in RS, given the lack of adequate prospective observational blinded studies as those performed for conventional MAS, 42 a reliable incidence of malfunctions or failure is not available and deserves further studies.
One of the limits of our study is the lack of homogeneity among procedures included in both groups (RMAS and no-RMAS), which is a direct consequence of the lack of randomization. Even so, the primary aim of this work was to present the setting of RS in a tertiary-care children's research hospital more than to compare approaches.
However, what comes clear from our study is that RS is extremely versatile and that this new technology can be applied to a vast area of pediatric surgical procedures, including some that would only be approached with conventional open surgery. This is one of the most intriguing take-home messages of our study that underlines how RS allows surgeons to apply MAS concepts to surgical procedures that are usually approached with conventional open surgery.
Further multicenter prospective studies and more appropriate technologies, adequate and tailored for small babies, are still required. It is necessary to understand which is the best setting to assure the best quality of care to our children as well as to develop a surgical network between hospitals (research, teaching, and clinical practice).
This preliminary experience proved to be extremely positive, confirming that the involved staff members are very keen in investing in high-level technology that extended the horizon of the applicability of MAS and its well-known fast-track concepts of care. The Da Vinci Robotic System is an intuitive, easy, and reproducible technology that offers the MAS approach to many children, in particular those who require complex reconstructive surgery. Although cost-effectiveness is still a matter of debate, a number of pediatric patients would surely benefit from this highly innovative approach.
Footnotes
Acknowledgments
The authors thank Prof. Mario Lima, University of Bologna, Prof. Gloria Pellizzo, University of Pavia, and Prof. Gian Antonio Manzoni, University of Milan, for their friendly suggestions toward the implementation of this study due to their experience with MAS and robotics.
Disclosure Statement
No competing financial interests exist.
