Robot-assisted general surgery in Aotearoa New Zealand

Robot-assisted surgery (RAS) refers to a surgeon controlling a robotic device that performs an operation. In its simplest iteration, it is an extension of surgical instruments and is not autonomous as it remains under the complete control of the operating surgeon. The first approved robotic surgical system (RSS) for clinical use in general surgery was a robot-assisted camera holder for laparoscopic surgery in 1993.¹ The da Vinci Surgical System (dVSS) (Intuitive Surgical, Sunnyvale, California, United States [US]) received approval from the US Food and Drug Administration (FDA) in 2000 and has been the dominant RSS used in general and abdominal surgery.¹

Robotic surgical systems for general surgery in Aotearoa New Zealand

There are various other RSS for general surgery available,¹ such as the Hugo (Medtronic, Dublin, Ireland) and Versius (CMR Surgical, Cambridge, United Kingdom) in Australia; however, to the best of our knowledge, these are not yet currently available in Aotearoa New Zealand. The first surgery using the dVSS in Aotearoa New Zealand was robot-assisted radical prostatectomy performed in 2007,² and there are currently seven dVSS in operation in the country. North Shore Hospital is the only public hospital with a dVSS, and its first robot-assisted general surgery procedure was performed in late 2022.

The RAS-specific code was only introduced to the National Minimum Dataset for hospital events in 2019, and despite its implementation it has been variably applied (communications with National Collections and Reporting, Manatū Hauora – Ministry of Health). Therefore, the data presented here utilise anonymous procedure-only information from the Aotearoa New Zealand distributor of the dVSS (Device Technologies, Auckland, New Zealand) akin to other published work in this area.^3,4

A total of 4,709 operations using the dVSS have occurred in private hospitals in Aotearoa New Zealand from 2007–2022. The number per year increased almost sevenfold, from 110 in 2008 to 743 in 2022 (Figure 1). For an initial 7 years, from 2007–2013, the dVSS was solely used for urological surgery, until the first cases of gynaecology and general surgery were recorded in 2014, and head and neck in 2016.

View Figure 1, Table 1–2.

The numbers of procedures and proportions in each of the defined categories are shown in Table 1, along with the five most frequently performed surgical procedures overall and their proportions in their respective category.

Among the 16 recorded general surgery procedures are rectopexy, cholecystectomy, distal pancreatectomy, liver resection, liver cystectomy and ventral/incisional hernia repair.

RAS with the dVSS in Australia and globally

In Australia, the dVSS was the only robotic platform to perform soft tissue operations until the limited entry of other platforms in 2018.³ Using the same data source,⁴ Table 2 compares the number of cases and systems available in Australia and Aotearoa New Zealand from 2015 to 2020. There was a similar rate of annual increase in the number of cases between the countries. In 2020, Australia had over double the number of dVSS per capita (2.6 vs 1.0 per 1 million) but performed almost five times the number of cases per capita (543 vs 109 per 1 million) as a result of the almost twice as many cases performed per RSS (208 vs 111).

Australia has seen a decreasing proportion of urology cases due to the expansion to other specialities, with urology accounting for 68% and gynaecology for 15% of cases in 2020, compared with 88% and 10% in Aotearoa New Zealand, respectively.^4,5 No detailed analysis of the numbers and types of all general surgery cases in Australia has been published, except pertaining to robot-assisted colorectal surgery.³ There were 6,110 robot-assisted general surgery cases using the dVSS in Australia between 2010 and 2019 with colorectal procedures accounting for 57.6%.³

World-wide, there were over 1.8 million procedures done utilising over 7,500 dVSS in 2022, with general surgery being the most rapidly growing and largest category—comprising almost half of all procedures—followed by urology and then gynaecology.⁶

Status and current evidence on robot-assisted general surgery

The robot-assisted approach has been applied for almost all procedures in general surgery (colorectal,⁷ oesophagogastric,⁸ hepatopancreatobiliary,⁹ breast,¹⁰ endocrine,¹¹ hernia¹² and transplant¹³). The diversity of procedures in general surgery and the well-established role of laparoscopy as a minimal access technique for common procedures have resulted in RAS only comprising a relatively small proportion of all general surgery procedures despite the significant rate of growth. For example, in the US state of Michigan, the proportion of RAS for general surgery increased from 1.8% in 2012 to 15.1% in 2018, with RAS comprising 7.5% of all cholecystectomies in 2018.¹⁴ At US community hospitals, which make up almost 90% of all general surgical RAS using the dVSS, it is estimated that about two general surgery procedures were done per dVSS per week in 2021.¹⁵

While the feasibility, safety and efficacy of RAS for numerous general surgery procedures have been demonstrated, contemporary evidence comparing its efficacy against the next best alternative (laparoscopic or open surgery) in randomised controlled trials (RCTs) is only recently emerging.¹⁶ These suggest the value of robotic assistance for surgical procedures manifests in complex procedures, wherein conventional laparoscopy as the other alternative to the minimally invasive approach is technically challenging or inexpedient. For example, in the largest (n=1,171) and most recent multi-centre (11 hospitals) RCT, total mesorectal excision for rectal cancer using RAS compared with laparoscopic surgery resulted in significantly reduced intra- (5.5% vs 8.7%) and post-operative (16.2% vs 23.1%) complications, fewer conversions to open surgery (1.7% vs 3.9%), shorter length of stay (7 vs 8 days) and better oncological quality of resection.⁷ Similar improvements in post-operative complications (13.2% vs 23.7%), open conversion (0% vs 2.9%) and post-operative length of stay (5 vs 7 days) have been observed in RAS compared with laparoscopy for abdominoperineal resections for low rectal cancer in a single-centre RCT (n=347), with additional improvements in 30-day readmission rate (2.3% vs 6.9%) and in urinary and sexual function without a difference in long-term oncological outcomes.¹⁷ A lower rate of post-operative complications was also observed in gastric cancer comparing RAS with laparoscopy for gastrectomy (8.5% vs 19.3%, two-centre RCT, n=236)¹⁸ and distal gastrectomy (9.2% and 17.6%, single-centre RCT, n=283).¹⁹ Further RCTs comparing RAS with thoraco-laparoscopic oesophagectomy for oesophageal cancer^8,20 and RAS with open pancreaticoduodenectomy for pancreatic and periampullary tumours^21,22 are ongoing.

Well-designed and conducted multi-centre RCTs provide the highest level of evidence regarding the efficacy of surgical therapeutic interventions.²³ Such trials are difficult to complete, with numerous challenges well described.²⁴ Although many established surgical procedures are not underpinned by multi-centre RCTs (for example, appendicectomy for uncomplicated acute appendicitis²⁵ and laparoscopic cholecystectomy²⁶), its value has been highlighted by a multi-centre RCT comparing minimally invasive to open radical hysterectomy for early cervical cancer.²⁷ Those results in gynaecologic oncology contravened the other retrospective and non-randomised evidence at the time to show an increased risk of death and recurrence with minimally invasive radical hysterectomy. The decreased overall survival in cervical cancer associated with RAS compared to open radical hysterectomy has since been corroborated in a recent systematic review and meta-analysis of matched or adjusted studies.²⁸

In addition to the general considerations of the applicability of trial populations (e.g., rates of obesity and comorbidities), a special consideration of trials involving surgical procedures is that the results are significantly influenced by the surgeons’ performance of the procedure.²⁹ The concept of a learning curve for surgical procedures is well recognised, but how to define and measure it for a specific procedure is variably established, let alone for a specific surgeon.³⁰ When comparing new surgical procedures with an established alternative there is a risk that trials earlier in the learning curve may not represent its true effectiveness, as was the case for laparoscopic inguinal hernia repair.³¹

The current literature reveals a significant monetary cost associated with RAS, especially in the context of a monopolistic RSS vendor.¹⁶ Despite the recent and future introduction of numerous other RSS vendors to the market¹ it is extremely unlikely that the direct costs of RAS will be lower than laparoscopic or open surgery due to the requirement of extra equipment to enable robotic assistance. It is very seldom that an advancement in technology, whether in telecommunications, homeware or medical devices, is associated with a reduction in direct equipment costs. Hence, RAS must demonstrate robust clinical benefits to be determined cost effective. Evidence from multi-centre RCTs suggested no clinical benefits for less complex procedures such as inguinal¹² and simple ventral hernia³² repair compared with laparoscopy, and instead demonstrated increased operative time, healthcare costs and surgeon frustration.

Cost effectiveness is an important consideration encompassed in assessing the value of an intervention. All healthcare systems, including our own, will continue to face multiple demands in weighing up  investment opportunity costs. In addition to the possible clinical benefits pertaining to complex surgical procedures previously evidenced, we believe the value of RAS in the public healthcare system will manifest through engendering equitable access, quality improvement and workforce development to futureproof surgical care for our population.

The value of robot-assisted general surgery in the Aotearoa New Zealand context

As new RSS vendors enter the market, it is salient to note that Aotearoa New Zealand does not have a pre-market approval process for medical devices under the Medicine Act 1981. RSS are multi-speciality technology that facilitate diverse procedures and indications. Specialists must consider the value of a specific procedure for a specific patient in their hands with the best available evidence. For instance, robot-assisted cholecystectomy may provide superior outcomes for certain indications (e.g., Mirizzi syndrome) and populations (e.g., chronic liver disease), which are not amenable to RCTs, by an experienced RAS surgeon.³³ Therefore, the assessment of the value of RSS for the health system is perhaps more complex than a particular medical device designed for a specifically defined indication.

Value assessments must also incorporate a focus on equity rather than a singular focus on cost effectiveness, as interventions that reduce inequity of health outcomes may cost more but be more valuable. Private healthcare in Aotearoa New Zealand is following regional and global trends in RAS, with an established practice in urology and a nascent practice in gynaecology. Most recent available figures show robot-assisted radical prostatectomy for prostate cancer comprised 28% of all radical prostatectomies in Aotearoa New Zealand for the 2019/2020 year, compared to only 11% in 2010/2011.³⁴ General surgery in Aotearoa New Zealand appears to be on the precipice, and international experience suggests that it is not only the fastest-growing category but also the highest volume. Until recently, access to RAS has only been available via private healthcare through the ability to pay and through having private health insurance. That inevitability results in disparities in access by wealth, and only 38% of the population report being covered by private health insurance.³⁵ This disproportionally affects Māori and Pacific peoples, who have an average annual household equivalised disposable income of 16–21% ($9,000–$12,000) less than NZ European s³⁶ and lower rates of private health insurance—22% of Māori and 17% of Pacific peoples compared to 40% of NZ European/Other.³⁵ The implementation of robot-assisted general surgery in the public healthcare system at the current opportunity, when it is not prevalent in private healthcare, may mitigate against disparities in access seen in other specialities.

Robot-assisted general surgery may also promote health equity by improving outcomes related to patient and disease-specific factors. For example, one of the Te Aho o Te Kahu quality improvement indicators for rectal cancer is the rate of abdominoperineal resection, which is associated with the rate of permanent stomas.³⁷ Māori have a higher rate than NZ European/Other (25.5% vs 21.9%),³⁷ and evidence from the most recent multi-centre RCT comparing RAS to laparoscopy for middle and low rectal cancer suggests a significantly lower rate for RAS (16.9% vs 22.7%).⁷  In addition, the benefits of RAS for gastric cancer^18,19 are particularly relevant for Māori, for whom it is the fourth most common cause of cancer death, and, compared with NZ European/Other, have a higher age–sex-standardised incidence and are more likely to be diagnosed with local and regional disease amenable to surgery.^38,39 Thus, the implementation of robot-assisted general surgery in public hospitals aligns with the New Zealand Health Strategy’s vision of pae ora, a healthy future for all, in “harnessing the benefits of innovation, technology and practice that improves how care is delivered, reduces variation and tackles inequity in outcomes. … and support[ing] access for the most under-served communities”.⁴⁰

What role Pharmac may have in determining the availability of RSS in public hospitals as it establishes a national list of all hospital medical devices by 2025 is yet to be defined. Traditional health technology assessments have been shown to be inadequate when exploring the context of application, such as patient-related and socio-organisational factors.⁴¹ Therefore, there is also an imperative for clinicians to lead and be involved in the evaluation to generate evidence specific to the Aotearoa New Zealand context. Such are the limitations of the currently presented and available data, devoid of clinical characteristics.

There are also benefits that extend to education, training and quality assurance, some of which did not exist with open or laparoscopic surgery. It has been shown that early surgical trainees perform more competently with RAS than with laparoscopic surgery,⁴² and for surgeons performing complex oncological surgery the RAS learning curve may be less than open surgery for achieving adequate cancer control.⁴³ This is germane to the Aotearoa New Zealand context due to our relatively small population; we could be considered a low-volume country for many complex surgical procedures.⁴⁴ The advances in simulation, proficiency-based curricula coupled with artificial intelligence and novel feedback mechanisms have improved safety and outcome for patients.^45–47 This has particular implications for Aotearoa New Zealand’s public healthcare system, where patients do not usually have a choice of hospital or surgeon, and consumers have emphasised the importance of ensuring professional competence that is publicly demonstrated.⁴⁸

Furthermore, the provision of RAS in public hospitals is a prudent strategic investment in developing a skilled workforce capable of delivering high-quality care, a priority area in the New Zealand Health Strategy.⁴⁰ As the evidence on RAS matures it is likely that Aotearoa New Zealand will follow the trends of other advanced economies overseas that are increasingly utilising RAS for complex surgical oncology.^3,49,50 General surgery training predominantly takes place in public hospitals, where the only accredited training attachments are based. RAS in public hospitals provides equitable opportunities to upskill current advanced trainees for competitive overseas fellowships at academic centres, where RAS is increasingly used. It will also support the recruitment and retention of returning specialists to the public health system, where they may apply their expertise in advanced therapies for the benefit of our local populations and contribute to the education of colleagues, including trainees. This will build capacity to integrate RAS into the training curriculum and ultimately develop self-sufficient pathways for local trainees in the Aotearoa New Zealand context. At Te Whatu Ora – Health New Zealand’s Waitematā District we partnered with several stakeholders to deliver free minimally invasive surgery workshops for surgeons and trainees that involved laparoscopic box trainers, ex-vivo animal organ simulation and RAS training, including the use of virtual reality.

Frameworks in place to support ethical implementation in Aotearoa New Zealand

In addition to equity of access and outcomes discussed above, the adoption of RAS necessitates other ethical considerations regarding informed consent, biases and managing conflicts of interest, including advertising. Aotearoa New Zealand law (Health and Disability Commissioner Act 1994 and the Health and Disability Commissioner [Code of Health and Disability Services Consumers’ Rights] Regulations 1996) and the Commissioner’s decisions provide clear guidance on informed consent for innovative procedures.^51–53 Several cognitive and emotional biases exist when handling medical technology.⁵⁴ It is important to be aware of biases as they can influence clinical practice and patient outcomes.^55,56 An essential component of addressing biases is mitigating the effect of conflicts of interest.⁵⁷ The Royal Australasian College of Surgeons provides practical guidance in a position paper on interactions with the medical industry.⁵⁸ Te Kaunihera Rata o Aotearoa | Medical Council of New Zealand have a statement on advertising that sets a standard supported by the Fair Trading Act 1986.⁵⁹

At Te Whatu Ora – Waitematā District we have established a transdisciplinary committee of multi-speciality clinicians, hospital management and non-clinical representation that guides the implementation of RAS in line with suggested evidence-based practice.⁶⁰ We have also developed an independent credentialing process that recognises individual surgeon performance is context specific and is not necessarily portable from one setting to another.⁶¹

Conclusion

The introduction of RAS to general surgery in Aotearoa New Zealand has some parallels to the introduction of laparoscopy over two decades ago.⁶² Current evidence suggests that its value for patients is realised in complex procedures, and its value for the health system may be multi-faceted. To achieve optimal outcomes, educational and quality improvement initiatives should be embedded in clinical implementation. Aotearoa New Zealand is well placed with legal, ethical and professional frameworks to support evidence-based dissemination. Clinicians from multiple specialities within general surgery, along with patients, should be involved in defining the future role of robot-assisted general surgery in Aotearoa New Zealand.