Coronary heart disease (CHD) is the leading cause of mortality and morbidity worldwide, and arterial calcification is strongly associated with poor prognosis of CHD [1, 2]. Previous studies have shown that the higher coronary artery calcification (CAC) was associated with the higher risk for atherosclerotic cardiovascular disease, and the CAC score has been regarded as an effective tool for predicting cardiovascular risk [1, 3, 4, 5].

Percutaneous coronary intervention (PCI) is one important treatment for CHD, but calcified lesions have been found to increase the possibility of failure in stent delivery [2]. To avoid this problem, the rotational atherectomy (RA) adopted differential cutting to ablate atherosclerotic plaques by forward advancement of a rotating abrasive diamond-encrusted burr [6]. RA can effectively reduce the plaque volume and enlarge the lumen, which allows easier stent deployment [7, 8, 9], and plaque modification has been shown to decrease the risk of stent restenosis and malapposition [10, 11, 12, 13]. RA has been traditionally performed via the transfemoral approach, mostly utilizing large guide catheters [$\ge$7 French (Fr)], which can accommodate the passage of large burrs ($\ge$1.75 mm). However, contemporary data show that using smaller sheath and catheter sizes reduces the risk of procedural access-site related complications [6]. We seek to clarify further factors that might be associated with the RA patient prognosis, such as procedural indices including the number of burrs, the size of burrs, approach site, the size of guiding catheter.

Many studies have also examined the relationship between heart patients’ blood chemistry and their prognosis. Plasma lipoprotein(a) [Lp(a)], low density lipoprotein cholesterol (LDL-C), and inflammatory factor high sensitivity C-reactive protein (hs-CRP) have all been shown to be related to the prognosis of patients with atherosclerotic cardiovascular disease (ASCVD) [1, 14, 15]. Importantly, Lp(a) has been shown to be a causal risk factor for CAC [16, 17, 18, 19]. However, it remains uncertain whether the level of Lp(a), LDL-C or hsCRP is related to the prognosis of patients who undergo RA, a relationship we aim to clarify in our study.

CAC was defined as mineral deposition in the coronary artery wall, which was very common in the aging population of the world [20, 21]. In routine PCI procedure, CAC could impair stent delivery and expansion. Importantly, it was highly predictive for MACE. So heavily calcified lesions need to be treated by RA to facilitate subsequent procedures [1, 2]. In the European expert consensus on RA, they recommended during RA, the operators could choose burrs step by step through burrs upsizing or downsizing [6]. However, there is scant data on whether the number or size of burrs is associated with long-term outcomes for the patients who underwent RA.

In this study, we analyzed the association between RA operation procedural indices and the prognosis of patients undergoing RA after a median of 56.7 months follow-up period. In our study, RA procedures were performed according to the operators’ experience and the routine about RA of out hospital in both the groups, including that the operators checked the calcification lesion by IVUS before RA, or they decided to perform RA procedure after they failed to dilate the lesion by post-dilated balloon or cutting balloon. And they usually used cutting balloon or post-dilated balloon to dilate the lesion fully after RA. Under the abovementioned background about procedural details, the analysis showed that using single burr during RA was related with lower rate of MACE and higher event-free survival rate (Log-rank p = 0.0065). We speculated that the possible explanations included that optimization of choosing the single burr during RA was enough to facilitate the procedure of intervention about severe calcification lesions, and in addition, choosing the optimal single burr, not 2 or more burrs could shorten the whole procedural time, and had more possibility to choose transradial access and less puncture-site related complications. In previous studies, using large burrs, which was more possible in non-single-burr group, didn’t improve the outcomes of the patients undergoing RA. Moreover, the experience from Taiwan local hospital supplied convincing evidence supporting the use of a single-burr strategy for RA to treat complex calcified coronary lesions rather than a routine step-by-step strategy for RA [22].

This study finds that appropriate burr size is vital for treatment of CAC during RA. We divided burrs into small (1.25 mm and 1.5 mm) and large (1.75 mm and 2.0 mm) groups and saw that, a burr-to-artery ratio 0.7 was recommended. In our study, we found that the burr-to-artery ratio was similar between the single-burr group and the non-single burr group (0.52 $\pm$ 0.1 vs 0.51 $\pm$ 0.09, p = 0.46, Table 1), and the burr to artery ratio was not associated with the occurrence of MACE (HR 1.88, 95% CI 0.11–3.26, p = 0.46, Table 2). Burr upsizing and burr downsizing was also beneficial for the lesions which could not be treated by one burr. The aforementioned European expert consensus suggested the step-up approach starting with 1.25 mm up to 1.5 mm or 1.75 mm burr [6, 7]. Given that RA has changed from merely a debulking tool to now lesion modification before stent implantation, a single-burr strategy could be improved to disrupt the continuity of calcium rings and flatten the vessel lumen in order to enable balloon dilatation and stent implantation [6, 7].

Chiou et al. [22] find that the single-burr strategy had the same result compared to the step-up approach pursuant to its advantages of less procedure time and complications. Levi et al. [23] report that 151 patients with small burrs achieve successful procedure, in comparison with 58 patients received the large burr (93% vs 100%, p = 0.07). In a long-term of 2616 days follow-up study about RA from a single center, Bartuś et al. [24] found that both high-risk category and mean stent(s) length were identified as independent predictors of MACCE. EuroSCORE II was confirmed to be the only independent predictor of MACE after RA [24].

However, approach site also plays an important role in the RA procedure. As reported previously, the procedural success rate and long-term prognosis were found to be similar between the transfemoral approach (TFA) and transradial approach (TRA) in routine PCI without RA procedures, but radial access had a lower risk of in-hospital major bleeding, major access site complications and longer hospital stay [6, 25]. Watt et al. [25] find TRA has the same procedural success as TFA (95.2% vs 94.9%, p = 0.56) and that TRA has a lower incidence of major access site complications than TFA (0.04% vs 1.3%, p = 0.004). Similarly, Kübler et al. [26] suggest that TRA is associated with equivalent procedural success compared to TFA (95% vs 87%, p = 0.07). However, they find that TFA is prone to have major access site bleedings (13% vs 1%, p = 0.001). After 1 year, however they find that TRA maintains the same results in comparison to TFA (p = 0.41) [26]. In our study, we also find that there was no statistical significance for the long-term prognosis between TRA and TFA group (p = 0.69, Table 2), providing evidence that TRA is a useful alternative method to avoid vascular complications. Using a 7F system offers the availability of larger burr sizes, when it comes to plaque preparation, operators in our study found it adequate to use a maximum burr size of 1.75 mm, which can be accommodated in a 6F guide. And 6F guide, which were themselves usually used in transradial procedures. Larger guide catheter offers the more support for complex bifurcation [7, 27]. In our study, there was no difference in prognosis between the two catheter-size groups (p = 0.25, Table 2).

In previous studies, Lp(a) was found to be associated with CAC in patients with or without familiar hypercholesterolemia [17, 28]. Pechlivanis et al. [18] find that both log-transformed Lp(a) and categories of Lp(a) (Lp(a) $\ge$54.3 mg/dL and Lp(a) 54.3 mg/L) have statistically significant links with CAC (beta per log unit increase in Lp(a) = 0.11, 95% CI 0.04–0.18, p = 0.002; Lp(a) $\ge$54.3 mg/dL vs Lp(a) 54.3 mg/L: 0.23, 95% CI 0.005–0.45, p = 0.05). Prior studies had found that Lp(a) $>$30 mg/dL could result in 5.51-fold increased risk for CAC progression $>$100, after adjustment for other potential covariates (p = 0.02) [29]. Patients with Lp(a) level $\ge$50 mg/dL were found to be more prone to having CAC progression than those with Lp(a) 50 mg/dL (odds ratio 1.333) [30]. Furthermore, except for the calcification, Lp(a) had an association with ASCVD. In particular, the mortality of patients with a high level of the Lp(a) increased than those with low-Lp(a) after PCI [16, 31]. However, we didn’t observe this association between Lp(a) and the prognosis of patients who underwent RA. Previous studies have also shown that patients with lower LDL-C will have lower rates of major coronary events [15]. Wada et al. [15] report that after an average of 6.5 years of follow-up, hs-CRP has a great association with the MACE [HR 1.1 (1.04–1.16), p 0.001] and all-cause mortality [HR 1.14 (1.06–1.22), p = 0.001]. However, we find no statistical significance in the relationship between the baseline level of LDL-C or hs-CRP with the prognosis of patients who underwent RA in our study.

There are several limitations to this study. First, the rotation operation strategies were decided by the cardiovascular intervention specialist based on experience, which might induce selection bias compared with randomized controlled trials. Although we adjust the comparison between groups for possible confounders, the selection bias might still exist because of unknown confounders. Second, the sample size of our study was small. For Lp(a), large scale prospective randomized controlled trials are needed to determine the role of Lp(a) in the prognosis of patients who undergo RA. Finally, our results are tempered by the fact that all of the data we collected came from a single medical center.

In summary, we provide evidence that patients who undergo RA with a single burr may have better outcomes than patients who receive more than one burr. We find no significant association between the occurrence of MACE with the other issues of RA procedural indices including the size of burrs, approach site and guiding catheter size. Moreover, we find that baseline levels of Lp(a), LDL-C and hs-CRP have no significant predictive value for the prognosis of RA patients.

RA, rotational atherectomy; MACE, major adverse cardiac events; Lp(a), lipoprotein(a); LDL-C, low density lipoprotein cholesterol; hs-CRP, high sensitivity C-reactive protein; CHD, coronary heart disease; CAC, coronary artery calcification; PCI, percutaneous coronary intervention; ASCVD, atherosclerotic cardiovascular disease; SBP, systolic blood pressure; DBP, diastolic blood pressure; BMI, body mass index; TC, total cholesterol; TG, triglyceride; HDL-C, high-density lipoprotein cholesterol; MI, myocardial infarction; TLR, target lesion revascularization; CABG, coronary artery bypass graft; HRs, hazard ratios; TFA, transfemoral approach; TRA, transradial approach.

NQW designed the study; SZ and WJZ collected the medical records, followed up the patients and performed statistical analysis, and wrote this manuscript; HWS, ZFL, YGS participated in collecting the data and data management; JQ was the consultant of this study. All authors contributed to editorial changes in the manuscript. All authors read and approved the final manuscript.

All subjects gave their informed consent for inclusion before they participated in the study. The study was conducted in accordance with the Declaration of Helsinki, and the protocol was approved by the Ethics Committee of Fuwai Hospital (Approval Number 2018-1086).

We would like to express our gratitude to all those who helped us during the writing of this manuscript. Thanks to all the peer reviewers for their opinions and suggestions.

This research received no external funding.

The authors declare no conflict of interest.