Intravascular ultrasound (IVUS)-guided percutaneous coronary intervention (PCI) has been established as a class 1 recommendation for complex lesions due to its proven benefits in improving outcomes.1 Growing evidence also supports its use in less complex lesions, yet the global adoption of IVUS remains low.2 This is primarily attributed to the additional costs, procedural time and specialised skills required for image interpretation.3 Reducing procedural time is one of the factors in less invasive coronary interventions, as it can improve catheterisation laboratory efficiency, and may reduce patient exposure to radiation and contrast media, particularly in high-risk patients, such as older adults or those with chronic kidney disease. However, IVUS guidance often adds to the procedure duration due to the time required for device preparation, catheter manipulation, and image acquisition and interpretation, especially for less experienced operators.
Cardiac CT has become a standard modality for evaluating coronary artery disease, often serving as the first-step diagnostic tool. Recent advancements in CT technology have enabled detailed assessments of lesion anatomy and characteristics, providing information comparable to that of intravascular imaging, such as IVUS or optical coherence tomography.4,5 These capabilities have led to the proposal of a CT-guided PCI strategy. A major advantage of this approach is that it leverages a CT scan already performed for diagnosis, thereby eliminating the need for additional intracoronary imaging devices, and potentially reducing procedural time and cost.
This study aimed to compare a cardiac CT-guided PCI strategy with the conventional IVUS-guided approach in patients with less complex coronary lesions. We hypothesised that CT-guided PCI could shorten procedural time while maintaining comparable safety and procedural success.
Methods
Our study complied with the Declaration of Helsinki and the Japanese Ethical Guideline for Medical and Health Research Involving Human Subjects. The institutional ethical committee approved the study protocol. All patients gave their informed consent by an opt-out procedure on our institutional website.
We retrospectively collected data on 135 consecutive patients with coronary artery disease aged >18 years who underwent drug-eluting stent deployment with preprocedural planning by CT between January 2021 and July 2025 at Omuta City Hospital, Omuta, Japan.
CT Image Analysis and Preprocedural Planning
All target coronary lesions were carefully examined with CT images and, when available, diagnostic coronary angiography. Detailed CT analysis has been reported previously, but is summarised below.5 All CT datasets were reconstructed and analysed on an ordinary workstation (Synapse Vincent; Fujifilm). Lesion characteristics were assessed using axial, longitudinal and curved multiplanar reformat images for the quantitative measurements of lesion length and lumen areas at the proximal and distal reference sites. Moreover, we analysed target lesions with thin-slab maximum intensity projection (MIP) images, which typically create 5 mm thick slab projections that consolidate multiple thin slices into a single image. Its cross-sectional view can also be displayed side by side. As a result, this technique provides simultaneous longitudinal views similar to coronary angiography and cross-sectional views similar to intravascular imaging. Lesion lengths were measured on longitudinal images in which the lesions could be clearly delineated. Proximal and distal reference lumen diameters were measured on the cross-sectional views. A representative case is presented in Figure 1.
The initial treatment plan, including stent size, was generated for all patients in both the CT-guided and IVUS-guided groups based on the CT measurements. The suggested nominal stent diameter was the closest available size (in 0.5-mm increments) to the mean diameter of the proximal and distal reference lumen diameters when the mean diameter was ≥2.5 mm. When the mean diameter was <2.5 mm, a 2.5-mm stent was selected. The suggested stent length was the closest size that was longer than the distance between the proximal and distal references. This constituted the initial, CT-based plan for every patient.
Importantly, all patients in both groups underwent preprocedural cardiac CT as part of our institutional diagnostic protocol, ensuring comparable baseline imaging assessment and diagnostic time investment across both study arms.
PCI Protocol and Final Stent Selection
Our standard strategy involved a distal radial approach, using a 5 or 6 Fr guiding catheter, and performing IVUS-guided PCI regardless of lesion complexity, which is common in Japan.6,7 We implemented CT-guided PCI as our primary strategy from May 2023.
All procedures were performed by cardiology fellows and a certified operator (KS) using the standard technique, including mandatory predilatation before deploying a drug-eluting stent. The stent was initially selected according to the preprocedural CT-based plan. The final stent size, however, was determined differently between the groups during the procedure. For the CT-guided PCI group, the initial stent and predilatation balloon sizes were selected according to the preprocedural CT-based plan. The final stent size was then confirmed and minimally adjusted, if necessary, based on the angiographic findings of the fully expanded predilatation balloon to ensure optimal wall apposition and safety. This final angiographic check was considered supplementary to the primary CT-based sizing. When angiographic complications, such as flow limitation, dissection or haematoma, were suspected, IVUS evaluation was performed, which was necessary in only one case due to no-reflow phenomenon. In the IVUS-guided PCI group, the final treatment plans, including the definitive stent size, were constructed based on IVUS findings. In all cases, the interventional strategy and instrumentation used were at the operator’s discretion.
Patient Selection
This study compared two sequential cohorts reflecting our institutional transition from IVUS-guided to CT-guided PCI in May 2023. The IVUS-guided group consisted of consecutive patients treated from January 2021 to April 2023, while the CT-guided group included consecutive patients treated from May 2023 to July 2025. We established exclusion criteria for CT-guided PCI and retrospectively applied these same criteria to the IVUS-guided cohort to ensure reduced bias between the two groups and fair comparison. The exclusion criteria were: severe calcified lesions requiring modification devices or calcified lesions where stent landing points could not evaluated (9 patients), in-stent restenosis lesions (3 patients), bifurcation lesions requiring two stents (0 patients), ostial lesions (5 patients), diffuse lesions requiring multiple stents or a very long (>40 mm) stent (9 patients), left main lesions (9 patients), chronic total occlusions (21 patients), coronary bypass grafting patients (1 patient), acute coronary syndrome patients (16 patients), multiple vessel intervention at one stage (3 patients), drug-eluting balloon or only plain balloon treatments (19 patients) and unclear visualisation of lesions on coronary angiography (2 patients). In total, 38 patients were included in the final analysis after excluding these patients from the 135 enrolled, with 23 patients in the CT-guided group and 15 in the IVUS-guided group.
The target lesions were evaluated by conventional 2D, single-vessel, off-line with a validated electronic-assisted system (CAAS workstation 8.5.2; Pie Medical Imaging). To standardise the measurement and avoid confounding factors, such as diagnostic angiography, in ad hoc PCI cases, arterial access difficulties or guiding catheter engagement challenges, we defined procedure time as the duration from the first contrast injection through the guiding catheter (excluding test shots for catheter positioning) to the final angiographic run confirming satisfactory stent deployment and vessel patency. This definition specifically captures the time required for the interventional procedure itself, including lesion preparation, stent deployment and final assessment, while excluding time variations related to access or diagnostic procedures.
Statistical Analysis
Continuous data are presented as the median with interquartile range (IQR). Categorical data are presented as absolute numbers and percentages. Continuous data were compared using the Mann–Whitney U test, and categorical data were compared using Fisher’s exact test. The correlation analyses were performed using Spearman’s rank correlation test. Specifically, the correlation between the CT-measured reference diameter (the mean of proximal and distal reference lumen diameters) and the actual size of the stent used was analysed in the CT-guided group. p-values of <0.05 were considered to be statistically significant. The statistical analyses were performed using R 4.3.1 (The R Foundation for Statistical Computing) with the EZR software package (Jichi Medical University).8
Results
Patient Characteristics
The baseline patient characteristics are summarised in Table 1. The age was significantly higher in the CT-guided group compared with the IVUS-guided group (77 [74.5–81.5] versus 73 [73–76.5] years; p=0.048). There was no significant difference between the two groups in terms of sex, body mass index and other cardiovascular risk factors. Creatinine and estimated glomerular filtration rate levels were also comparable between the groups.
Angiographic and CT Finding
Angiographic and CT findings are detailed in Table 2. The angiographic lesion length was significantly shorter in the CT-guided group (11.68 [9.49–18.19] versus 19.86 [17.08–27.52] mm; p=0.004). Similarly, the CT-based lesion length was shorter; however, the lesion length measured on the thin-slab MIP image was not statistically significant. There were no significant differences in reference vessel diameter, minimum lumen diameter on quantitative coronary angiography or proximal and distal reference parameters on CT.
Procedural Characteristics
All procedures were successful. Procedural characteristics are presented in Table 3. The length of the used stent was also significantly shorter in the CT-guided group (18 [16–24] versus 24 [21–32] mm; p=0.037). No significant differences were observed in fluoroscopy time or contrast volume. The procedure time was significantly shorter in the CT-guided group compared with the IVUS-guided group (25 [22–48] versus 43 minutes [35–50]; p=0.028; Figure 2 ). The study found no significant differences in the rate of side branch wiring, modified jailed balloon technique, post-balloon dilatation or the size of the guiding catheter. One patient in each group experienced a periprocedural MI. In the CT-guided group, this patient was the only one to undergo additional intravascular imaging, requiring IVUS examination for haematoma evaluation (Supplementary Figure 1 ). Minor bleeding (Bleeding Academic Research Consortium type 2) was observed in one patient in the CT-guided group. A Spearman’s rank correlation test was performed to examine the relationship between procedure time and lesion length. The results showed no significant correlation (rho=0.0817; p=0.626).
Comparative Analysis of Stent Sizing
The analysis examined the difference between the diameterof the stent used and the predicted diameter, which was calculated as the average of the proximal and distal CT reference diameters. Descriptive statistics showed a median positive difference in both groups, indicating a consistent clinical preference for oversizing relative to the CT-derived mean diameter. The CT-guided group had a median difference of 0.1 mm (IQR −0.10–0.45 mm), whereas the IVUS-guided group showed a larger median difference of 0.5 mm (IQR 0.15–0.53 mm). A Mann–Whitney U test confirmed a statistically meaningful disparity in the distribution of these sizing errors between the two groups (W=106; p=0.048). This finding indicates that stents sized with IVUS guidance resulted in a significantly greater degree of oversizing relative to the initial CT-based prediction than those sized with CT guidance alone. Furthermore, both prediction methods showed a strong positive correlation with the final used size (Spearman’s ρ: CT-guided 0.682, p<0.001; IVUS-guided 0.761, p<0.001).
Discussion
In the modern era of PCI, IVUS guidance has been established as a class 1 recommendation for complex lesions, and its proven benefits in improving outcomes are now being extended to less complex lesions.9 However, the associated procedural time and cost remain significant barriers to its widespread adoption. This study, to our knowledge, is the first to directly compare a cardiac CT-guided PCI strategy with the conventional IVUS-guided approach, specifically evaluating procedural time in patients with less complex lesions. Given the single-centre, retrospective design with a limited sample size, our findings should be interpreted as a preliminary investigation of feasibility rather than a definitive comparison of clinical effectiveness.
Our findings suggest that a cardiac CT-guided PCI strategy may be feasible for reducing procedural time in selected noncomplex lesions, although this requires validation in larger studies. The observed reduction of approximately 18 minutes per case suggests potential benefits, as a decrease in procedure duration is one of the factors in less invasive coronary interventions. Shorter procedural times can enhance catheterisation laboratory efficiency, and may potentially reduce patient exposure to radiation and contrast media, particularly in high-risk patients. Prolongation of procedure time during IVUS-guided PCI is a major barrier to adhering to guideline-recommended practices and is significantly influenced by lesion complexity, although studies detailing the actual duration of the prolongation are limited. In contrast, a CT-guided study on RCA ostial lesions demonstrated a reduction in procedural time.10
Our findings are particularly noteworthy because they demonstrate this time-saving effect in noncomplex lesions, where the added value of IVUS in routine clinical practice may be limited. Despite the limited sample size, this proof-of-concept investigation provides preliminary evidence of the feasibility of CT-guided PCI in selected patients. The substantial time savings observed (median reduction of approximately 18 minutes) suggest clinically meaningful benefits that warrant larger-scale validation studies.
The usefulness of cardiac CT has been widespread not only in the diagnosis of coronary artery disease, but also in preprocedural planning of PCI. CT-guided PCI has limitations, as its assessment is only preprocedural, whereas IVUS guidance is beneficial for pre-, peri- and postprocedural evaluation. Although a growing body of evidence suggests its advantages even in less complex lesions, the clinical relevance of IVUS in noncomplex lesions remains a matter of debate.9 When the preprocedural assessment is thorough, most remaining issues in noncomplex lesions are flow-limiting edge dissection or haematoma, as seen in the present study. Edge dissection and haematoma can be identified on angiography, while larger stent sizing and stent edge dissections are characteristic features observed with IVUS guidance.11 Thus, the added cost and time for IVUS in noncomplex lesions may not be easy to justify in real-world clinical practice. Cardiac CT has become the primary diagnostic tool for assessing coronary artery disease. As a result, CT data is typically acquired during the initial diagnostic phase. While CT-guided PCI has not been established as a substitute for IVUS, it provides a significant advantage by offering detailed preprocedural planning without increasing procedural time or necessitating additional equipment.
Although this study does not directly compare CT guidance with angiography-only PCI, the CT-guided approach offers a more precise preprocedural assessment, enabling accurate selection of stent size and optimal landing zone.12 The present study demonstrated a median difference of only 0.1 mm between the used stent diameter and the predicted diameter, which was calculated as the average of the proximal and distal CT reference diameters, with a high degree of correlation. This precision, even without intraprocedural IVUS, could lead to better acute and long-term outcomes, and potentially reduce adverse events compared with angiography-only procedures, even in noncomplex lesions.
A recent randomised clinical trial found comparable results between quantitative coronary angiography-guided and IVUS-guided PCI.13 Quantitative coronary angiography does not require intravascular imaging; however, the study used a specific stent size selection algorithm and PCI protocol. The 5–10% larger stent sizing and mandatory postdilatation at high pressure may raise concerns about coronary dissections and perforations for operators, potentially leading to undersizing, similar to angiography-guided PCI. Visual confirmation of lumen sizes is an advantage in CT-guided PCI. The ongoing large-scale clinical trials comparing IVUS-guided and CT-guided PCI will provide more definitive evidence regarding the impact of CT guidance on clinical outcomes.
The choice of imaging technique for preprocedural planning may influence procedural efficiency. For stent sizing, we specifically used thin-slab MIP images, as they offer a unique advantage in identifying optimal stent landing zones compared with traditional straightened multiplanar reformat images. While this thin-slab MIP analysis is a routine part of our preprocedural assessment, the time required is minimal. Thin-slab MIP images bridge the gap between CT and angiography by providing an angiography-like view that clearly visualises the coronary anatomy. This approach allows operators to select appropriate landing points and quickly obtain accurate measurements from the corresponding axial CT images, typically requiring <10 minutes, and is performed well in advance of the catheterisation procedure. Therefore, the combination of angiography-mimicking morphology from thin-slab MIP and rapid, accurate measurement from axial CT makes this technique a superior guide for efficient and precise PCI without adding significant time to the overall workflow.
Our study specifically evaluated the efficiency of the interventional phase of PCI, as reflected in our procedural time definition (from first contrast injection through the guiding catheter to final angiography). This focused measurement was designed to isolate the impact of CT-guided versus IVUS-guided strategies on the execution of the coronary intervention itself, independent of access-related variations or diagnostic procedures. While this approach provides a clear assessment of interventional efficiency, our findings should be interpreted as reflecting the time required for the coronary intervention proper, rather than total catheterisation laboratory time from puncture to completion.
Limitations
This study has several important limitations that warrant careful consideration. First, this sequential cohort comparison may be subject to temporal bias due to our institutional transition from IVUS-guided to CT-guided PCI. Although a single certified operator (KS) supervised all procedures and identical exclusion criteria were applied to minimise this concern, multiple rotating cardiology fellows participated throughout the study period, and we cannot entirely exclude potential contributions from operator-dependent variability.
Second, it is a single-centre, retrospective study with a small sample size, which limits the generalisability of our findings.
Third, the CT-guided group had a significantly shorter lesion length. Although we found no significant correlation between lesion length and procedure time, suggesting this difference may not fully explain the observed time reduction, we cannot definitively exclude it as a contributing factor given the study’s limited sample size.
Fourth, we did not systematically compare CT and IVUS measurements within the same patients, as the anatomical locations of IVUS measurements were not systematically documented with fixed landmarks during IVUS-guided procedures, precluding reliable anatomical correspondence. This limitation reflects our institutional documentation practices at the time and underscores the importance of systematic measurement site recording. Our study focused on comparing procedural strategies rather than validating CT measurement accuracy, which has been established in prior dedicated studies.
Fifth, this study did not assess whether the CT-guided approach provides a similar outcome improvement as IVUS. However, previous reports have suggested that CT-based stent sizing is comparable to that of optical coherence tomography.12 The current study showed no significant differences in quantitative coronary angiography parameters or final stent sizes between the two groups.
Sixth, our study did not assess long-term clinical outcomes, which remain the ultimate measure of PCI success. The learning curve for CT image interpretation and the availability of high-quality CT scans may also influence the generalisability of our findings.
Finally, our procedure time measurement focused specifically on catheterisation laboratory efficiency and did not include the time for preprocedural CT acquisition, which was performed identically in both groups as part of standard diagnostic workup. This focused definition allows for a direct comparison of the two intraprocedural guidance strategies.
A further limitation relates to the interpretation of the procedural time difference. The additional 18 minutes observed in the IVUS-guided group is a complex result influenced by the distinct procedural protocols (predetermined planning in the CT group versus real-time strategic optimisation in the IVUS group). Furthermore, as our study was conducted in a training environment involving cardiology fellows and is not a high-volume PCI centre, subtle contributions from instruction time and potentially less routine proficiency among allied medical staff in IVUS operation may have contributed to the observed difference when compared with external reports.
Conclusion
In this preliminary single-centre study, we demonstrated that a cardiac CT-guided PCI strategy was feasible and associated with shorter procedural time in patients with noncomplex coronary lesions compared with the conventional IVUS-guided approach. Safety and procedural success were comparable between the two strategies. While a shorter procedural time may contribute to a less invasive approach for patients and improved catheterisation lab efficiency, these promising preliminary results provide the foundation for planning adequately powered randomised controlled trials to definitively establish the comparative effectiveness of CT-guided versus IVUS-guided PCI strategies. Therefore, these findings should be considered hypothesis-generating and require validation in larger, prospective, multicentre studies before CT-guided PCI can be broadly recommended as an alternative to IVUS guidance in noncomplex lesions.
Clinical Perspective
- In this preliminary study of noncomplex coronary lesions, CT-guided percutaneous coronary intervention was associated with shorter procedure time compared with intravascular ultrasound guidance (median 25 versus 43 minutes; p=0.028), although baseline lesion characteristics differed between groups.
- CT-measured reference diameters showed good correlation with actual stent sizing (Spearman’s ρ=0.682; p<0.001), suggesting that preprocedural CT planning may enable reasonable stent selection without intravascular imaging in selected cases.
- These hypothesis-generating findings require validation in larger studies with matched cohorts or adjusted analyses before CT guidance can be considered as an alternative to intravascular ultrasound in clinical practice.