The current study is the first single-unit comparative study evaluating the impact of the range of MIS approaches on conversion to open surgery during restorative TME for rectal cancer. Our results show that a robotic approach and taTME are both associated with a decreased conversion rate compared to conventional laparoscopic surgery. However, after performing a multivariate analysis, only taTME was associated with a decreased conversion rate, whilst obesity was an independent risk factor for conversion.
The analysis of conversion as an outcome is relevant since the literature shows that conversion is associated with impaired outcomes in MIS rectal cancer surgery[6, 7, 9]. In this study, conversion had a negative impact on postoperative outcomes, as reflected by a higher overall and surgical complication rate, higher wound infection rate and an increased length of hospital stay. In their systematic review meta-analysis examining the impact of conversion in laparoscopic rectal cancer surgery, Gouvas et at.[6] showed that conversion was associated with a longer duration of surgery, hospital stay, and higher rates of wound infection compared with completed laparoscopic cases. Additionally, others have shown that conversion may influence long-term outcomes of patients undergoing MIS rectal cancer surgery. Clancy et al.[8] published a systematic review and meta-analysis evaluating the influence of conversion on the oncologic outcomes of colorectal cancer surgery. In their study, a successful laparoscopic surgery was associated with a lower risk of disease recurrence (OR 0.634, 95% CI 0.429–0.938, P = 0.023) and lower overall mortality risk (OR 0.512, 95% CI 0.417–0.629, P < 0.0001). Similarly, a large Dutch national cohort study recently analysed the influence of conversion and anastomotic leakage on long-term oncologic outcomes in rectal cancer surgery[7]. Of 745 attempted laparoscopic cases, 14.4% were converted. OS and DFS were significantly shorter in the conversion compared to the laparoscopic group (p = 0.025 and p = 0.001, respectively) as well as in anastomotic leakage compared to patients without anastomotic leakage (P = 0.002 and 0.024, respectively). In the multivariable analysis, anastomotic leakage was an independent predictor of OS and DFS. Although conversion did not influence OS, it was an independent predictor of DFS (1.525, 1.071–2.172). The authors concluded that technical difficulties during laparoscopic rectal cancer surgery, as reflected by conversion and anastomotic leakage have a negative prognostic impact on these patients. Whilst more studies on this subject may further clarify the impact of conversion in the short and long-term outcomes of rectal cancer, it seems that tailoring the utilisation of the different MIS alternatives to diminish conversion rates is sensible and may translate into postoperative and long-term benefits in this scenario.
Our results are in keeping with the literature showing that robotic TME and taTME are associated with decreased conversion rates compared to conventional laparoscopic surgery[18, 19]. Nevertheless, few studies compare the outcomes of the full range of MIS approaches to TME. Two network metanalyses examining open, laparoscopic, robotic and transanal TME suggest that MIS approaches would provide similar oncologic outcomes and an enhanced postoperative recovery compared to open TME [20, 21]. Conversion, however, was absent as an outcome or not compared amongst MIS approaches in these studies. Moreover, the small proportion of robotic and taTME cases compared to open and laparoscopic TME limit the extrapolation of these data. A recent multicentric Dutch study compared the outcomes of laparoscopic, robotic and transanal rectal cancer surgery[22]. After propensity score-matching their cohort, 108 patients were compared in each group. Conversion rates did not differ between approaches. The authors concluded that robotic and taTME expert centres perform more anastomoses in rectal cancer surgery. Although these publications reflect an effort to better understand the outcomes of the various MIS alternatives to TME, potential selection biases and the comparison of outcomes from multiple centres with different skill sets hamper the ability to draw clear conclusions about the real value of one approach over the other.
To our best knowledge, this is the first single-unit experience evaluating the impact of the whole spectrum of MIS approaches to restorative TME on conversion. The limitations of this study, however, are multiple and essentially related to its retrospective design. Statistical type 2 errors cannot be excluded. It is also worth noting that our series includes the institutional and surgeons’ learning curves for robotic and taTME which could underestimate their benefit. Moreover, we do not provide long-term data to assess oncologic equivalency amongst the different approaches, especially in the light of recently published studies pointing to adverse oncological recurrence patterns associated with taTME[23, 24]. Our group along with other contributors from Australasia have previously published acceptable short-term outcomes as well as sound oncological outcomes associated with taTME[25]. These findings are similar to those reported in quaternary centres in the Netherlands[26], and combined series beyond the learning curve[27], but in contradiction to the previously cited series from the Netherlands and Norway[23, 24]. Further data on the learning curves and oncologic outcomes of taTME will be needed to validate this approach and tailor its utilisation in rectal cancer surgery.
Navigating the learning curve for new technology and techniques can be challenging. To overcome the learning curve for taTME, several studies have indicated that a case number between 30 and 70 cases is needed[28]. Our Australasian series suggests that 40 cases are required to navigate through the learning curve[25]. Similarly, data suggests that 40 to 50 cases are needed to overcome the robotic learning curve for low anterior resections[29]. Whilst this study evaluates the merits of the various MIS techniques within a quaternary setup, it is worth mentioning that we believe many to be complementary. For example, a low or very low rectal cancer in an obese male patient may be best served as a taTME in combination with a robotic transabdominal approach to enable safe reconstructive options and maximise the ability to dissect along correct oncologic planes. The current authors are strong advocates for appropriate training pathways for the adoption of novel MIS techniques such as robotic TME and taTME. Proctorship, case observations and finally appropriate case volume are required to allow such techniques to be safely implemented and skills to be maintained.