In the last few decades, there has been a rapid succession of technological advances, marking a radical shift from the open to the minimally invasive surgical (MIS) approach [1]. The advantages of laparoscopy over laparotomy are now widely acknowledged [2]. Over the past twenty-five years, robotic surgery has experienced a raise and today, with the availability of several platforms alongside the continuously leading da Vinci systems (Intuitive Surgical Inc., Sunnyvale, CA, USA), robotic approaches are playing an increasingly crucial role [3, 4]. Despite observed advantages for certain patient characteristics (e.g., BMI > 30), challenges such as the lack of dedicated reimbursement, high costs and often longer operating times still limit the widespread use of robotic platforms worldwide [5]. Open surgery provides direct visual and tactile information of the explored regions. In contrast, MIS comes at the cost of predominantly two-dimensional view and limited tactile assessment. Intraoperative ultrasound (IOUS) is commonly utilized during open surgery with linear or finger probes, particularly in the hepatobiliary (HPB) and urological fields [6, 7]. In laparoscopic setting, ultrasound probes for guidance in MIS are more challenging to handle [8]. To overcome this limitation, innovative approaches for robotic platforms integrate ultrasound imaging to facilitate its use in MIS [9]. Image-guided robotic approaches, particularly those based on three-dimensional (3D) imaging, augmented reality (AR), and machine learning algorithms, offer advantages in the era of digital surgery [10]. Real-time, non-invasive, cost-effective and dynamic intraoperative imaging of complex anatomy are the main benefits of computer-assisted surgery. In this context, IOUS has emerged as the imaging modality of choice facilitated by the introduction of articulated robotic instruments to handle ultrasound probes [11]. The augmentation and fusion of imaging modalities are especially beneficial for delineating healthy and neoplastic tissue in oncological surgery [12]. The navigation of drop-in ultrasound probes manoeuvred by articulated robotic graspers provides access to anatomical spaces and angles that are inconvenient for relatively rigid laparoscopic probes. While initial reports of applications of intraoperative ultrasound during robotic surgery (RIOUS) have been published in the fields traditionally managed by open surgery, with encouraging results., pooled data are lacking [6, 13]. Therefore, the aim of this systematic review is to assess the reported applications of intraoperative ultrasound-guided robotic surgery and to outline future perspectives.

Materials and methods

Search strategy

The systematic review was conducted according to Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines [14] and registered with the International Prospective Register of Systematic Reviews PROSPERO (n CRD42023494430) prior to data extraction. Articles were obtained by querying the PubMed database, Google Scholar, ScienceDirect and ClinicalTrial.gov filtered by the English language up to October 2023 without additional restrictions. The database was retrieved through title and abstract screening using the following search terms: “intraoperative”, “robotic”, “surgery”, “ultrasound”, “laparoscopic”, “probe”.

Data extraction

After removing duplicate publications, titles, abstracts, and keywords were independently reviewed by M.P and E.T. for inclusion, followed by full text review of eligible articles. In case of discrepancies, a consensus was reached through agreement with a third author (M.G.). The inclusion criterion was the description of ultrasound-assisted robotic surgical procedures. Excluded were articles without robotic use of the probe, as well as abstracts, reviews, meta-analyses, letters, and editorials. Studies reporting robotic ultrasound imaging independently of a surgical procedure, or those focused on percutaneous ultrasound-guided techniques and biopsies were also excluded. Data about the authors, surgical procedures, probes specifics and ultrasound-assisted robotic procedures were extracted for further analysis.

Results

The search strategy identified studies reporting intraoperative ultrasound imaging during robotic surgery. Initially, 781 studies were identified, and 68 full texts were selected through title and abstract screening. Finally, 20 studies met the inclusion criteria for the systematic review (Fig. 1). Due to the low number of reports, a qualitative analysis was performed [6,7,8, 11, 13, 15,16,17,18,19,20,21,22,23,24,25,26,27,28].

Fig. 1
figure 1

PRISMA flow diagram of study selection

Full size image

Among the included studies, two were prospective (10%), fifteen teen were retrospective (75%), three were experimental (15%), involving laboratory tests in vivo (porcine models) or on ex vivo phantoms. The studies were mainly (53%) from the field of general surgery during liver, pancreas, spleen, gallbladder/bile duct, vascular and rectal surgery [6, 8, 11, 16,17,18,19,20,21,22, 28]. The remaining studies(42%) covered other fields of oncological surgery including urology [7, 13, 23,24,25,26], lung surgery [27], and retroperitoneal lymphadenectomy for metastases [15].

Ten studies (53%) were focused on locating tumoral lesions and defining resection margins [7, 11, 13, 18,19,20, 22, 25, 27, 29]. Additionally, four studies (15%) were designed to assess the feasibility of robotic ultrasound-guided surgery [8, 15, 23], two (10.5%) aimed to compare robotic and laparoscopic ultrasound probes [6, 24], another two (10.5%) were conducted to evaluate the robotic drop-in probe [13, 16], one study (5%) focused on assessing hepatic tissue consistency [17] and another (5%) aimed to visualize the blood flow in the splenic artery [21].

In eleven articles (55%), a miniaturized linear drop-in probe was used [6, 7, 11, 13, 15, 16, 23,24,25, 27, 28]. These probes can be introduced via a 10–12 mm accessory trocar and steered from the surgeon’s console using robotic graspers (Fig. 2). Five manuscripts reported the use of rigid probes, which can be docked to the robotic arm (12 mm trocar) as prototypes corresponding to da Vinci robotic instruments (Fig. 2) [8, 21]. Alternatively, a laparoscopic articulated probe can be used during robotic surgery, introduced via the 10 mm accessory port and manipulated by the bedside assistant (Fig. 2). The ultrasound frequencies of the probes used in the included studies ranged from 3 to 13 MHz. All reported procedures were performed with the da Vinci robotic platforms. Table 1 summarizes the characteristics of the included articles, and details about ultrasound probes and surgical applications. No clinical trials on the use of RIOUS were registered at the timepoint of the database query. Due to the heterogeneity of data concerning probes, frequencies, procedures and study outcomes, a quantitative analysis of the results was deemed inappropriate.

Fig. 2
figure 2

Timeline of intraoperative ultrasound techniques

Full size image
Table 1 Study details and ultrasound
Full size table

Discussion

In this systematic review, we present a a comprehensive analysis that sheds light on the current state of intraoperative ultrasound for guidance in robotic procedures.

Summary of main results

Despite the high quality level of evidence supporting laparoscopic ultrasound in various thoraco-abdominal pathologies [30] and the desire to implement RIOUS for over two decades [31], the literature still reflects limited evidence regarding ultrasound guidance during robotic surgery, with relatively small cohort sizes.

All included studies, however, consistently report satisfactory performance of RIOUS. To facilitate the widespread adoption of RIOUS, there is a need for increased adoption of robotic surgical procedures and training for surgeons in IOUS. The utilization of computer assistance for image acquisition and interpretation, through the development and training of machine learning algorithms, could contribute to overcoming operator dependency in ultrasound examinations [32]. In line with the findings of this analysis on RIOUS, image guidance for identifying resection margins by differentiating between healthy and neoplastic tissues has proven particularly useful in oncological diseases [33]. The foremost beneficiary of (R)IOUS thus far is the hepatobiliary field, particularly for the comprehensive anatomical assessment of the biliary and vascular trees [34]. In liver surgery, IOUS plays a well-established role as an intraoperative guidance tool in combination with preoperative CT and MRI imaging. Surgical radicality depends on the detectability of lesions in the different imaging modalities. Techniques such as image fusion of CT/MRI and US, multimodal registration of 2D and 3D imaging modalities as well as (contrast-enhanced) ultrasound contribute to identifying known and preoperatively undetected lesions in order to intraoperatively tailor the surgical strategy [35, 36]. Furthermore, RIOUS was demonstrated to have superior performance compared to conventional LIOUS with a success rate exceeding the one of LIOUS in liver surface exploration (85% vs. 73%, P = 0.030) and tool manipulation (79% vs. 57%, P = 0.028) [8]. Post-task questionnaires completed by participating surgeons revealed that robotic ultrasound significantly improved probe positioning (80%), reduced fatigue (90%), and was overall more useful than LIOUS (90%) [8]. Facilitating precise probe positioning in RIOUS not only enhances surgical precision but also reduces the physical strain on surgeons during complex procedures [6]. An even more significant benefit is the opportunity to identify otherwise undetected lesions, such as in pancreatic lesions [18]. In benign disease of the biliary tract, IOUS has demonstrated comparable efficacy with intraoperative cholangiography in diagnosing choledocholithiasis, surpassing it in terms of speed and completion rates. This is achieved without the need for a contrast agent, with reduced invasiveness and a decreased risk of infection.The comprehensive assessment of the intra- and extrahepatic biliary tree can be accomplished in an average time of 164.1 s using RIOUS and can be complemented by Doppler ultrasound for assessing the porta hepatis. Precise measurements of the biliary tree and ductal stones enable intraoperative decision-making and management of ductal pathologies, including hybrid approaches [6, 37].

Similarly, rectal tumours were successfully detected using RIOUS, showing its effectiveness in determining the optimal transection line for rectal surgeries, especially in cases where tumours are too high for transanal palpation [16]. Furthermore, in obese patient with rectal cancer RIOUS has been proved to be useful to safely guide vascular dissection [28]. Nephron-sparing surgery, as an alternative to radical nephrectomy, is gaining support as an oncologically equivalent procedure while preserving renal functional capacity [9]. The evolution of robot-assisted partial nephrectomy techniques has ushered in a progressive refinement of tools aiding surgeons in the identification of masses and their vascular networks. A remarkable 100% success rate was demonstrated in identifying kidney lesions with RIOUS [25], optimizing tumour identification, enhancing renal tissue preservation through partial nephrectomy, and ensuring oncological safety [6, 9, 37]. In transoral robotic tongue base resection for obstructive sleep apnoea RIOUS has emerged as an invaluable tool for locating the lingual artery and assessing laryngeal tissues. The integration of RIOUS significantly enhances efficiency by substantially reducing the risk of detrimental intraoperative bleeding complications [29].

Despite the numerous advantages observed across various surgical domains, the integration of intraoperative ultrasound in the robotic field remains underused due to costs consideration, lack of expertise, and the necessity for highly skilled minimally invasive surgeons trained in both robotics and ultrasound techniques [11]. Moreover, although rigid prototypes compatible with robotic arms have been developed [8, 20], they are barely due to cost and the absence of a significant advantages over rigid laparoscopic probes, [6, 38]. In contrast, the adaptability of drop-in probes to all multi- and single-port robotic platforms offers high scalability in clinical applications [6, 8, 20, 38].

Results in the context of published literature

Applications of IOUS originated in 1960 for the identification of kidney stones in A-mode [34, 39]. Since 1980s, rapid innovations have progressed with applications in hepato-pancreato-biliary and gastrointestinal surgery [40, 41]. In the 1990s, attempts were made to extend the benefits of IOUS to minimally invasive surgery by creating dedicated probes for laparoscopic ultrasound [42]. When used in the robotic setting, these probes were operated by the bedside assistant. However, laparoscopic probes lack the flexibility of IOUS in open surgery (Fig. 2). As robotic platforms do not yet provide integrated ultrasound probes, a specific transducer known as the “drop-in-probe" was recently introduced for robotic surgery. This probe, with a dorsal fin to be grasped with a robotic instrument, can be steered from the console. The small transducer attached to a highly flexible cable, coupled with the motion range of the articulating instrument, facilitates access to anatomical areas that are hard to reach with standard laparoscopic probes.

Furthermore, dedicated robotic console software, such as TilePro (Intuitive Surgical Inc., Sunnyvale, CA, USA), enables the surgeon to create an in-console split-view with side-by-side intraoperative and ultrasound images, or switch between the minimally invasive 3D camera and ultrasound view directly from the console [22]. On platforms with open consoles, surgeons can switch from the integrated robotic display to the external ultrasound screen ideally positioned close to the console surgeon [43]. One of the known limitations of laparoscopy, and even more so in robotic surgery, is the reduced/absent tactile feedback, requiring considerable training is needed to learn to replace haptic with visual information. Consequently, the availability of additional information via RIOUS is particularly relevant in oncology, where achieving zero residual tumour is a major prognostic factor [44,45,46,47,48]. Exploration of the abdominal cavity with LIUOS can detect malignant deposits preventing conversions to open surgery when remaining disease can be excluded [44,45,46,47, 49]. Fertility-sparing surgery can be enhanced by IOUS assistance by discriminating healthy from cancerous tissues and to spare ovarian parenchyma [46]. Image-guided organ exploration during surgery could also impact the detection of undiagnosed masses, especially in pancreatic and splenic diseases [11, 18, 19]. Therefore, margin assessment and mapping resection guidance with IOUS are highly relevant in conservative oncologic surgery [50, 51]. However, large-scale future randomized controlled trials (RCTs) are necessary to demonstrate the utility of IOUS in assessing oncological outcomes.

Although the limited number of publications and the presence of heterogeneity among the included studies, mostly consisting of case reports and case series which have been included to report comprehensively the literature evidence, this systematic review on RIOUS procedures highlights the relevance of the technical advances in robotic surgery which underline its expected impact in the field of image-guided surgery.

Implications for practice and future research

In recent years, an increasing number of robotic platforms has entered the marketplace, a trend expected to persist with decreasing costs and user-friendly platforms for a variety of procedures [3]. However, as the integration of advanced technology based on artificial intelligence and augmented reality is not yet fully automated, making the inclusion of real-time 3D image information into MIS a crucial step in advancing surgical care [10]. Ultrasound-assisted procedures are poised to play a pivotal role in filling this technological gap and are anticipated to grow in parallel with ongoing advancements.

Beyond 3D macroscopic guidance, there is a growing demand for real-time intraoperative tissue analysis, particularly for tailoring the radicality of resection in oncological diseases. In vivo 3D tissue analysis would be ideal for guiding surgery intraoperatively. A variety of intraoperative optical imaging techniques are currently under assessment to complement or potentially replace extemporaneous histopathological analysis [23, 52]. For in vivo tissue, 3D high resolution ultrasound represents a significant step forward in intraoperative analysis within the anatomical context, aiding decision-making on whether resection is required, such as in lymph node metastasis [50]. High (up to 70 MHz) and ultra-high (up to 100 MHz) frequency probes are considered candidates to achieve a resolution of 30 µm, similar to histopathology [53]. An immediate ex vivo imaging system that does not require dedicated sample preparation is full-field optical coherence tomography (FF-OCT), showing a rapid learning curve and analysis of tissue sections similar to [54, 55]. On resected specimens, whole-slide imaging can be used for digital reconstruction as a 3D volume preventing missed lesions for skipped depth slide [56]. In the era of digital surgery, robotic platforms represent computer interfaces capable of integrating multiple modalities of real-time data analysis [10] (Fig. 3). The integration of surgical and imaging sciences will need interdisciplinary training and specific core curricula such as the Master in Image-Guided Surgery, teaching surgeons to perform IOUS, particularly in MIS [57]. Moreover, ongoing studies in deep learning applied to new diagnostic technologies will address the need for standardised IOUS performance and data interpretation by surgeons who may lack adequate radiological expertise [58,59,60].

Fig. 3
figure 3

(Copyright Barbara Seeliger/ Carlos Amato; Chengyuan Yang; Niloofar Badihi; IHU Strasbourg and Cannon Design USA)

The next-generation hybrid operating room integrating artificial intelligence and robotics for diagnostic imaging, procedure planning and execution: the operating room of the future is envisioned as the centre of a technology ecosystem. Illustrated technology include advanced interactive digital displays with real-time connectivity and AI analytics, mixed-reality environments, and robotic applications for various interventions, imaging (ultrasound, cone-beam CT, intraoperative CT/MRI, etc.), nursing assistance and sterile instrument management, as well as a predictive logistics supply system with Automatic Guided Vehicles [61]

Full size image

Conclusions

Robotic surgery has become increasingly common in routine clinical practice. Recent technological advancements have paved the way for new tools and equipment in robotic and image-guided surgery. The advantages of robotic instrumentation, including ergonomics, dexterity, and precision of movements, are particularly relevant for robotic intraoperative ultrasound. This systematic review demonstrates the virtue of RIOUS to support intraoperative decision-making and potentially reduce minimally invasive procedure times. Prospective studies, however, are needed to better understand its potential, including disciplines like gynaecologic oncology, where these procedures are not yet commonly performed.