Distraction-aware hierarchical learning for vascular structure segmentation in intravascular ultrasound images

Vascular structure segmentation in intravascular ultrasound (IVUS) images plays an important role in pre-procedural evaluation of percutaneous coronary intervention (PCI). However, vascular structure segmentation in IVUS images has the challenge of structure-dependent distractions. Structure-dependent distractions are categorized into two cases, structural intrinsic distractions and inter-structural distractions. Traditional machine learning methods often rely solely on low-level features, overlooking high-level features. This way limits the generalization of these methods. The existing semantic segmentation methods integrate low-level and high-level features to enhance generalization performance. But these methods also introduce additional interference, which is harmful to solving structural intrinsic distractions. Distraction cue methods attempt to address structural intrinsic distractions by removing interference from the features through a unique decoder. However, they tend to overlook the problem of inter-structural distractions. In this paper, we propose distraction-aware hierarchical learning (DHL) for vascular structure segmentation in IVUS images. Inspired by distraction cue methods for removing interference in a decoder, the DHL is designed as a hierarchical decoder that gradually removes structure-dependent distractions. The DHL includes global perception process, distraction perception process and structural perception process. The global perception process and distraction perception process remove structural intrinsic distractions then the structural perception process removes inter-structural distractions. In the global perception process, the DHL searches for the coarse structural region of the vascular structures on the slice of IVUS sequence. In the distraction perception process, the DHL progressively refines the coarse structural region of the vascular structures to remove structural distractions. In the structural perception process, the DHL detects regions of inter-structural distractions in fused structure features then separates them. Extensive experiments on 361 subjects show that the DHL is effective (e.g., the average Dice is greater than 0.95), and superior to ten state-of-the-art IVUS vascular structure segmentation methods.

Effect of the pipeline embolization device placement on branching vessels in anterior circulation: a systematic review and meta-analysis

BACKGROUND AND PURPOSE

Pipeline embolization device (PED) is widely used in intracranial aneurysms, and the scope of applications for the PED, which is frequently used to treat cerebral aneurysms, is also growing. It has some effect on branching vessels as a result of its inherent properties. The effects of PED on the complications rate and branching vessels blockage have not yet been thoroughly investigated.

OBJECTIVE

We conducted a systematic review searching reports from multiple databases on PED use for intracranial aneurysms, and analyzed the influence of PED on the occlusion rate of different branching vessels, and the influence of the amount of PED on the occlusion rate of branching vessels by meta-analysis.

METHODS

We searched the literature using PUBMED, Web of Science, and OVID databases until August 2023. Inclusion criteria were that the study used only PED, included at least 10 patients, and recorded branching vessels occlusion rates, mortality, and neurological complications.

RESULTS

Nine studies were analyzed consisting of 706 patients with 986 side branches. The results of the meta-analysis showed that application of more than one PED did not significantly elevate the rate of branching vessels occlusion compared to application of one PED (OR = 0.70; 95% CI: 0.34 to 1.43; P = 0.33). In the comparison of branching vessels occlusion rates in the anterior circulation, the anterior cerebral artery (ACA) had a significantly higher occlusion rate compared to the ophthalmic artery (OphA) (OR = 6.54; 95% CI: 3.05 to 14.01; P < 0.01), ACA also had a higher occlusion rate compared to the anterior choroidal artery (AchA) (OR = 15.44; 95% CI: 4.11 to 57.94 P < 0.01), ACA versus posterior communicating artery (PcomA) occlusion rate difference was not statistically significant (OR = 2.58; 95% CI: 0.63 to 12.82; P = 0.17), OphA versus AchA occlusion rate difference was not statistically significant (OR = 2.56; 95% CI: 0.89 to 7.38; P = 0.08), and the occlusion rate was significantly higher for PcomA compared to AchA (OR = 7.22; 95% CI: 2.49 to 20.95; P < 0.01) and lower for OphA compared to PcomA (OR = 0.33; 95% CI: 0.19 to 0.55; P < 0.01).

CONCLUSION

The meta-analysis shows that use of multiple PEDs did not significantly increase the occlusion rate of branching vessels, and the larger the diameter of branching vessels covered by PED, the higher the occlusion rate of branching vessels. However, the incidence of complications is low after branching vessels occlusion in anterior circulation, which is related to the collateral circulation compensation of the branching vessels.

Establishing correlations between normal pancreatic and submandibular gland ducts

Background

The objectives of this study were to evaluate the relationship between ductal morphometry and ramification patterns in the submandibular gland and pancreas in order to validate their common fractal dimension.

Methods

X-ray ductography with software-aided morphometry were obtained by injecting barium sulphate in the ducts of post-mortem submandibular gland and pancreas specimens harvested from 42 adult individuals.

Results

Three cases were excluded from the study because of underlying pathology. There was a significant correlation between the length of the main pancreatic duct (MPD) and the intraglandular portion of the right submandibular duct (SMD) (r = 0.3616; p = 0.028), and left SMD (r = 0.595; p < 0.01), respectively, but their maximal diameters did not correlate (r = 0.139—0.311; p > 0.05). Both dimensions of the SMD showed a significant right-left correlation (p < 0.05). The number of MPD side branches (mean = 37) correlated with the number of side branches of left SMD, but not with the right one (mean = 9). Tortuosity was observed in 54% of the MPD, 32% of the right SMD, and 24% of the left SMD, with mutual association only between the two salivary glands.

Conclusions

Although the length of intraglandular SMD and MPD correlate, other morphometric ductal features do not, thus suggesting a more complex relationship between the two digestive glands.

Accessory Vein Obliteration and Balloon-Assisted Maturation for Immature Arteriovenous Fistulas for Haemodialysis: A Systematic Review and Meta-Analysis

PURPOSE

This study aims to examine outcomes of immature arteriovenous fistula salvage using balloon angioplasty (PTA) without and with accessory vein obliteration (PTA + VO).

MATERIALS AND METHODS

PubMed and Embase were accessed on 21 September 2020 to retrieve cohort studies on adult patients with end-stage renal failure (ESRF) requiring dialysis. Risk of bias was assessed using Newcastle-Ottawa Scale. Studies were pooled into PTA or PTA + VO arms, with outcomes (technical/clinical success, primary/secondary post-intervention patency until 12 months) reported as event rates with 95% confidence intervals. Random-effects model and maximum likelihood meta-regression were used for meta-analysis.

RESULTS

Fourteen studies (1030 participants) were included. The between-subgroup difference in outcomes was largely non-significant (p > 0.050).

CONCLUSION

The evidence does not support balloon angioplasty with concomitant accessory vein obliteration for immature fistula salvage.

The importance of side branch preservation in the treatment of chronic total occlusions with bifurcation lesions

Objectives

The aim of this study was to identify the predictive factors for suboptimal result in side branch (SB) in chronic total occlusion (CTO) bifurcation lesions and clinical outcomes of patients with suboptimal result in SB.

Background

There is little evidence on the optimal strategy for bifurcation lesions in CTO.

Methods

We retrospectively reviewed 314 consecutive bifurcation lesions in CTO with SB ≥ 2.5 mm in 3 hospitals from March 2010 to June 2018. Patients were divided into the two groups based on the suboptimal SB treatment (SST) and optimal SB treatment (OST) groups. The baseline characteristics, procedural and clinical outcomes were compared between the two groups. This study also evaluated the predictors of suboptimal result in SB.

Results

Suboptimal result in SB occurred in 47 cases. Presence of stenosis in SB, bifurcations located within the occluded segment and sub-intimal tracking at SB ostium was an independent predictor of suboptimal result in SB. The cumulative incidence of target lesion revascularization (TLR) in all lesions was not significantly different between the two groups, however, TLR in right coronary artery (RCA) was significantly higher in the SST group. In the Cox regression analysis, suboptimal result in SB in RCA and sub-intimal tracking were independent predictors of TLR for MB. In patients with bifurcations located within the occluded segment, usage of two-stent technique was significantly lower in the SST group.

Conclusions

Meticulous procedures are required for SB preservation to improve not only SB prognosis but also MB, especially in RCA.

Percutaneous coronary intervention in side branch coronary arteries: Insights from the Japanese nationwide registry

Background

Performance of percutaneous coronary intervention (PCI) in side-branch vessels (SB-PCI) has not been fully investigated despite the technical advancement of PCI.

Methods

We investigated 257,492 patients registered in the Japanese nationwide PCI registry from January to December 2018; 199,767 (78%) underwent PCI for major vessel PCI (MV-PCI), 21,555 (8.4%) underwent SB-PCI, and 24,862 (9.6%) underwent PCI for both vessels (SB + MV-PCI). The frequencies of primary composite adverse events, defined as in-hospital mortality and procedural complications (i.e., peri-procedural myocardial infarction, tamponade, new-onset cardiogenic shock, stent thrombosis, emergent surgery, and bleeding), and PCI for restenotic lesions were investigated. Their association with institutional frequency of each PCI was also investigated.

Results

Fewer drug-eluting stents (66% vs. 86%) and more drug-coated balloons (23% vs. 9%) were used in SB-PCI than in MV-PCI (p < 0.001). Pre-procedure non-invasive testing was similarly performed in SB-PCI and MV-PCI (57% vs. 61%). The composite endpoint was observed in 0.7%, 1.9%, and 2.2% of the SB-PCI, SB + MV-PCI, and MV-PCI groups, respectively (p < 0.001). Institutional frequency of SB-PCI was inversely associated with the composite-endpoint risk for all PCI procedures (odds ratio 1.37, 95% confidence interval 1.04–1.81 in the lowest tertile, with reference to the middle tertile, p = 0.02). Frequency of PCI for restenotic lesions was also inversely associated with the institutional frequency of MV-PCI (p < 0.001).

Conclusion

SB-PCI was performed safely with a low frequency of acute complications, and higher SB-PCI frequency presented a lower risk of in-hospital adverse events, albeit with a cost of an increase in PCI for restenotic lesions.

Optimal strategy for side branch treatment in patients with left main coronary bifurcation lesions

INTRODUCTION AND OBJECTIVES

There are no guidelines regarding the most appropriate approach for provisional side branch (SB) intervention in left main (LM) bifurcation lesions.

METHODS

The present prospective, randomized, open-label, multicenter trial compared conservative vs aggressive strategies for provisional SB intervention during LM bifurcation treatment. Although the trial was designed to enroll 700 patients, it was prematurely terminated due to slow enrollment. For 160 non-true bifurcation lesions, a 1-stent technique without kissing balloon inflation was applied in the conservative strategy, whereas a 1-stent technique with mandatory kissing balloon inflation was applied in the aggressive strategy. For 46 true bifurcation lesions, a stepwise approach was applied in the conservative strategy (after main vessel stenting, SB ballooning when residual stenosis> 75%; then, SB stenting if residual stenosis> 50% or there was a dissection). An elective 2-stent technique was applied in the aggressive strategy. The primary outcome was a 1-year target lesion failure (TLF) composite of cardiac death, myocardial infarction, or target lesion revascularization.

RESULTS

Among non-true bifurcation lesions, the conservative strategy group used a smaller amount of contrast dye than the aggressive strategy group. There were no significant differences in 1-year TLF between the 2 strategies among non-true bifurcation lesions (6.5% vs 4.9%; HR, 1.31; 95%CI, 0.35-4.88; P=.687) and true bifurcation lesions (17.6% vs 21.7%; HR, 0.76; 95%CI, 0.20-2.83; P=.683).

CONCLUSIONS

In patients with a LM bifurcation lesion, conservative and aggressive strategies for a provisional SB approach have similar 1-year TLF rates.

Prognostic significance of the Medina classification in bifurcation lesion percutaneous coronary intervention with second-generation drug-eluting stents

The Medina classification is the most widespread method to describe bifurcation lesions. However, little is known regarding its prognostic impact. Therefore, the aim of this study is to assess the prognostic significance of the Medina classification following percutaneous coronary intervention (PCI). From a prospective registry of 738 consecutive patients undergoing PCI for bifurcation lesions, 505 were treated with second-generation drug-eluting stents (DES). Of these, 407 (80.6%) presented with "true bifurcation" (TB) lesions (Medina class 1.0.1, 1.1.1, 0.1.1) and 98 (19.4%) in all other categories ("non-true bifurcation" = NTB). We compared rates of death and major adverse cardiac events (MACE: cardiac death, myocardial infarction, or target vessel revascularization) at 12 months and 3 years. Patients with TB had lower rates of previous bypass surgery (7.4% vs. 11.2%, p = 0.043). TB lesions were more likely to be calcified (33.9% vs. 28.6%, p = 0.003) and ulcerated (8.8% vs. 4.1%, p < 0.01). At 12 months, mortality was numerically higher for TB PCI (4.1% vs. 2.1%, p = 0.052) and MACE rates were higher (19.2% vs. 10.2%, p < 0.001). At 3 years, both all-cause death (10.1% vs. 5.1%, p = 0.002) and rates of MACE (37.2% vs. 17.6%, p < 0.001) were higher for TB PCI. After performing regression analysis, TB remained an independent predictor for poor outcomes (OR-2.28 at 12 months, CI 1.45-9.50, p = 0.007, OR-3.75 at 3 years, CI 1.52-6.77, p = 0.001 for MACE). In conclusion, TB lesions, according to the Medina classification, portend worse prognosis for patients undergoing bifurcation PCI. This may guide prognostication and decision-making in treatment.

The Relation Between Optical Coherence Tomography-Detected Layered Pattern and Acute Side Branch Occlusion after Provisional Stenting of Coronary Bifurcation Lesions

BACKGROUND/PURPOSE

Layered pattern (presumed to be healed plaque after a thrombotic event) can be observed by optical coherence tomography (OCT). We sought to assess the ability of OCT-detected plaque composition to predict acute side branch (SB) occlusion after provisional bifurcation stenting.

METHODS

This is a retrospective observational study using pre-intervention OCT in the main vessel to predict Thrombolysis in Myocardial Infarction (TIMI) flow grade ≤1 in a SB (diameter ≥ 1.5 mm) after provisional bifurcation stenting. OCT-detected layered pattern was defined as plaque with a superficial layer that had a different optical intensity and a clear demarcation from underlying tissue.

RESULTS

Overall, 207 patients with stable coronary disease were included. SB occlusion occurred in 26/207 (12.6%) bifurcation lesions. Operators decided not to perform additional treatment, and TIMI flow did not improve to ≥2 in cases with SB occlusion. The prevalence of OCT-detected layered pattern was more common in lesions with versus without SB occlusion (88.5% versus 33.7%, p < 0.0001); OCT-detected layered pattern was more often located on the same side of the SB (73.9% versus 21.3%, p < 0.0001) circumferentially compared to lesions without SB occlusion. Multivariable analysis showed that OCT-detected layered pattern was an independent predictor of SB occlusion (odds ratio 18.8, 95% confidence interval 5.1-68.8, p < 0.0001) along with true bifurcation lesion and wider angiographic bifurcation angle.

CONCLUSIONS

The presence of an OCT-detected layered pattern near its ostium was a strong predictor of SB occlusion after provisional bifurcation stenting.

Serial changes in the three-dimensional aspect of the side-branch ostium jailed by a drug-eluting stent assessed by optical coherence tomography

The present study investigated serial changes in the three-dimensional (3D) aspect of the jailed side-branch (SB) ostium. We evaluated 32 patients who underwent examination with optical coherence tomography (OCT) both at baseline and at follow-up. After reconstruction of the 3D images, we classified the configuration of overhanging struts at the SB orifice into three groups according to the 3D aspect of the jailing configuration. The number of compartments divided by the stent strut was counted. The side-branch flow area (SBFA), i.e., the area of the SB ostium except for jailing struts, was measured by cut-plane analysis. Forty-eight SBs of 25 patients were analyzed. Thirteen SBs were classified as the No-jail type (N-type), 19 as the Simple-jail type (S-type; no longitudinal link at the carina), and 16 as the Complex-jail type (C-type; had a link at the carina). In the N-type, the SBFA was significantly increased at follow-up (P = 0.018). In the C-type, the SBFA was significantly decreased at follow-up (P = 0.002). Percent reduction of SBFA in the C-type group was significantly greater than that in the N-type or S-type groups (S-type vs. C-type P = 0.002, N-type vs. C-type P < 0.001). 3D-OCT images showed that some of the compartments were filled with tissue. The number of compartments was significantly decreased at follow-up (P < 0.001). In the C-type group, the SBFA was significantly decreased and small compartments were filled with tissue. These findings suggest that stent jail complexity is associated with the progression of SB ostial stenosis.

The Sendai and Fukuoka consensus criteria for the management of branch duct IPMN - A meta-analysis on their accuracy

INTRODUCTION

The risk of malignancy in branch duct intraductal papillary mucinous neoplasia of the pancreas (BD-IPMN) is controversially debated. An increasing number of studies report on outcomes using the Sendai or Fukuoka consensus criteria for treatment decision-making. The objective of this work was to evaluate the diagnostic accuracy of the Sendai and Fukuoka criteria.

METHODS

We systematically reviewed studies on Sendai or Fukuoka criteria-guided management of BD-IPMN. Pooled sensitivity, specificity and diagnostic odds ratios as compound measures of diagnostic accuracy were calculated from studies matching the inclusion criteria. The meta-analysis was performed using a random effects model.

RESULTS

Fifteen studies with a total of 2710 patients were included. Twelve of these used the Sendai criteria. In these studies, 23% of Sendai-negative patients had a high grade dysplastic lesion or an invasive carcinoma in final histology. Pooled sensitivity was 56%, specificity was 74% and the diagnostic odds ratio for malignancy in Sendai-positive lesions was 7.45. When the results of follow-up examinations were included, diagnostic accuracy improved significantly (14.66, p < 0.001). Three studies were identified that used the Fukuoka criteria for decision making. Of 200 patients with Fukuoka-negative lesions who underwent surgery, 22 had a malignant lesion in final histology (11%). Pooled sensitivity was 83%, specificity was 53% and the diagnostic odds ratio was 8.76.

CONCLUSION

The Fukuoka criteria have considerably improved sensitivity but still lack adequate specificity. For further reduction of a potential surgical overtreatment of BD-IPMN, the development of criteria with an increased specificity is required.

A novel side branch protection technique in coronary stent implantation: Jailed Corsair technique

Side branch occlusion, which was one of the common complications in percutaneous coronary interventions, was closely associated with cardiac death and myocardial infarction. Clinical guidelines also support the importance of preservation of physiologic blood flow in SB during PCI to bifurcation lesions. In order to avoid side branch occlusion during stent implantation, we often performed the jailed wire technique, in which a conventional guide wire was inserted to the side branch before stent implantation to the main vessel. However, the jailed wire technique could not always prevent side branch occlusion. In this case report, we described a case of 72-year-old male suffering from angina pectoris. Coronary angiography revealed the diffuse calcified stenosis in the proximal and middle of left anterior descending coronary artery, and the large diagonal branch originated from the middle of the stenosis. To prevent side branch occlusion, we performed a novel side branch protection technique by using the Corsair microcatheter (Asahi Intecc, Nagoya, Japan). In this case report, we illustrated this "Jailed Corsair technique", and discussed the advantage compared to other side branch protection techniques such as the jailed balloon technique.

Differential effect of side branch intervention on long-term clinical outcomes according to side branch stenosis after main vessel stenting: Results from the COBIS (Coronary Bifurcation Stenting) Registry II

BACKGROUND

Indication of side branch (SB) intervention after main vessel (MV) stenting is not established for coronary bifurcation lesions.

METHODS

We evaluated 2017 patients who were treated with 1-stent technique or MV stenting as a first strategy. Patients undergoing SB intervention after MV stenting (SB intervention group, n=929) were compared to those treated with MV stenting only (no-SB intervention group, n=1088).

RESULTS

During a median follow-up of 37months, cardiac death or myocardial infarction (MI) tended to occur less frequently in the SB intervention group than in the no-SB intervention group (1.8% versus 2.9%; adjusted hazard ratio [HR] 0.53; 95% confidence interval [CI] 0.25-1.11; P=0.09). There was a significant interaction between SB intervention and SB stenosis after MV stenting (P for interaction <0.01). Among 1077 patients with diameter stenosis of SB ≥50% after MV stenting, SB intervention was associated with a lower risk of cardiac death or MI (1.2% versus 4.2%; adjusted HR 0.22; 95% CI 0.09-0.52; P<0.01). However, among 940 patients with diameter stenosis of SB <50%, there was no significant difference in cardiac death or MI between the SB intervention group and the no-SB intervention group (3.5% versus 2.2%; adjusted HR 1.36; 95% CI 0.58-3.20; P=0.48).

CONCLUSIONS

The effect of SB intervention differed according to SB stenosis after MV stenting. SB intervention may reduce cardiac death or MI in bifurcation lesions with diameter stenosis of SB ≥50% after MV stenting.

The fate of small side branches following drug eluting stent implantation

OBJECTIVES

Although drug eluting stents (DES) have documented convenience in bifurcation lesions, possible unfavorable effects on small side branch ostium (SBO) remain a question. We aimed to explore the effects of DES on small jailed SBs (1.5-2.25 mm) which originated from the lesion on the main vessel and were not treated with either stenting or balloon dilatation.

METHODS

Angiographic data of 107 consecutive patients (129 SB) with Medina 1,1,1 or 1,1,0 lesions were evaluated at the time of procedure and at the follow-up.

RESULTS

Of all DES used, 70 (54.7%) was sirolimus-eluting, 39 (30.5%) was paclitaxel-eluting and 20 (14.8%) was zotarolimus-eluting. The diameter of SBs was 1.84 ± 0.41 mm with a stenosis of 20.7 ± 26.6% at SBO at baseline. The lesion at the SBO had progressed after the procedure when the pre vs postprocedure values and follow-up vs pre-procedure values are compared (20.7 ± 26.6% vs 29.4 ± 27.4%; p < 0.0001 and 25.4 ± 25.1 vs 20.7 ± 26.6%; p = 0.004 respectively). A significant reduction in stenosis was revealed over the follow-up (29.4 ± 27.4 vs 25.4 ± 25.1 respectively; p = 0.013). The severity of the disease at the SBO at baseline was the only parameter that affected the severity of SB stenosis in acute, longterm and follow-up. Additional parameters with influence on SB patency at different times were female gender, stent deployment with low pressure, cTFC of the main lesion, age, cTFC of the lesion, late loss index and the preprocedure TIMI flow grade of the SB.

CONCLUSIONS

Although there was a significant deterioration of SBO immediately after stenting, follow-up data showed that the lesion at SBO improved but remained worse than baseline.

Instructive percutaneous coronary intervention to avoid the risk of side branch occlusion at a lesion with a lotus root appearance: a case report

Background

A lotus root appearance is a rare entity, and there is little opportunity to perform coronary intervention for this kind of lesion. Because of its peculiar anatomical characteristics, one of the problems regarding percutaneous coronary intervention (PCI) for these lesions is related to the involvement of branch vessels.

Case presentation

We encountered a case of PCI for a stenotic lesion with a lotus root appearance in the mid-portion of the right coronary artery (RCA). To avoid the risk of right ventricular (RV) branch occlusion due to stent deployment in the main RCA, we re-crossed the third guidewire into the main RCA via the nearest point to the RV branch ostium through the communicating vascular lumen. Thereafter, we deployed a drug-eluting stent in the main RCA crossing over the RV branch, and the ostium of the RV branch remained intact, as we expected.

Conclusions

This case is the first report in the world describing the details of how to maintain the patency of the side branch bifurcating from a lesion with a lotus root appearance under optical coherence tomography guidance.

Bailout technique to rescue the abruptly occluded side branch with collapsed true lumen after main vessel stenting

Guidewire recrossing into the abruptly occluded side branch (SB) after main vessel (MV) stenting in the coronary bifurcation is difficult, particularly if the SB has a dissection because the true lumen of SB is collapsed by a hematoma and the second guidewire easily goes into the false lumen. This paper reports a bailout technique to rescue the occluded SB that was complicated by a hematoma because of an unsuccessful guidewire recrossing after MV stenting using a small balloon dilation in the collapsed SB true lumen behind the stent strut and wire penetration.