Author + information
- Received September 21, 2015
- Revision received March 28, 2016
- Accepted April 21, 2016
- Published online August 8, 2016.
- Truls Råmunddal, MD, PhDa,∗ (, )
- Loes P. Hoebers, MDb,
- José P.S. Henriques, MD, PhDb,
- Christian Dworeck, MDa,
- Oskar Angerås, MD, PhDa,
- Jacob Odenstedt, MD, PhDa,
- Dan Ioanes, MDa,
- Göran Olivecrona, MD, PhDc,
- Jan Harnek, MD, PhDc,
- Ulf Jensen, MD, PhDd,
- Mikael Aasa, MD, PhDd,
- Per Albertsson, MD, PhDa,
- Hans Wedel, PhDe and
- Elmir Omerovic, MD, PhDa
- aDepartment of Cardiology, Sahlgrenska University Hospital, Gothenburg, Sweden
- bDepartment of Cardiology, Academic Medical Center, Amsterdam, the Netherlands
- cDepartment of Coronary Heart Disease, Skåne University Hospital, Lund, Sweden
- dDepartment of Cardiology, Stockholm South General Hospital, Stockholm, Sweden
- eHealth Metrics, Sahlgrenska Academy at University of Gothenburg, Gothenburg, Sweden
- ↵∗Reprint requests and correspondence:
Dr. Truls Råmunddal, Sahlgrenska University Hospital, Department of Cardiology, Blå Stråket 5, 413 45 Gothenburg, Sweden.
Objectives The aim of this study was to determine the prognostic impact of chronic total occlusion (CTO) on long-term mortality in a large prospective cohort.
Background CTO is present in many patients with coronary artery disease and is difficult to treat with percutaneous coronary intervention.
Methods The study population consisted of all consecutive patients who underwent coronary angiography in Sweden between January 1, 2005 and January 1, 2012, who were registered in SCAAR (Swedish Coronary Angiography and Angioplasty Registry). The patient population was heterogeneous with regard to indication for angiography (stable angina, ST-segment elevation myocardial infarction [STEMI], unstable angina or non-STEMI, and other) and treatment options. The long-term mortality rates of patients with and without CTO were compared by using shared frailty Cox proportional hazards regression adjusted for confounders. Tests were conducted for interactions between CTO and several pre-specified characteristics: indication for angiography and percutaneous coronary intervention (stable angina, STEMI, unstable angina or non-STEMI, and other), severity of coronary artery disease (1-, 2-, and 3-vessel and/or left main coronary artery disease), age, sex, and diabetes.
Results During the study period, 14,441 patients with CTO and 75,431 patients without CTO were registered in SCAAR. CTO was associated with higher mortality (hazard ratio: 1.29; 95% confidence interval: 1.22 to 1.37; p < 0.001). In subgroup analyses, the risk attributable to CTO was lowest in patients with stable angina and highest in those with STEMI. In addition, CTO was associated with highest risk in patients under 60 years of age and with lowest risk in octogenarians. There was no interaction between CTO and either diabetes or sex, suggesting an equally adverse effect in both groups.
Conclusions In this large prospective observational study of patients with coronary artery disease, CTO was associated with increased mortality. This association was most prominent in younger patients and in those with acute coronary syndromes.
- acute coronary syndrome(s)
- chronic total occlusion
- coronary artery disease
- percutaneous coronary intervention
Chronic total occlusion (CTO) is present in about 16% of patients with significant coronary artery disease (CAD) (1). These lesions are difficult to treat with percutaneous coronary intervention (PCI) and are regarded as the remaining challenge in myocardial revascularization. PCI in patients with CTO has gained much attention in recent years with the introduction of new techniques and devices resulting in high rates of procedural success. However, the scientific evidence for this treatment is based on retrospective studies and on expert consensus. Clear recommendations regarding the management of patients with CTO are not available in European and in American guidelines (2,3). No study has investigated the prognostic impact of CTO in patients with stable angina. In studies of the effects of CTO on short- and long-term mortality, CTO was associated with increased mortality (4) in patients with ST-segment elevation myocardial infarction (STEMI). However, these reports were based on subgroup analyses of observational databases and small cohorts. Therefore, larger studies are needed to establish the true effect of CTO on mortality in patients with CAD.
We identified a large prospective cohort of patients with CTO registered in SCAAR (Swedish Coronary Angiography and Angioplasty Registry), a prospective national registry that accumulates data on all patients undergoing coronary angiography and PCI in Sweden (1). The CTO cohort in SCAAR consists of more than 14,000 patients with stable angina and acute coronary syndrome (ACS) (1).
In this study, we analyzed data from the CTO cohort in SCAAR to determine whether CTO is associated with increased mortality in patients undergoing coronary angiography and/or PCI.
Established in 1999 as part of the national SWEDEHEART (Swedish Web-System for Enhancement and Development of Evidence-Based Care in Heart Disease Evaluated According to Recommended Therapies) registry (5), SCAAR gathers data on all consecutive patients from all hospitals that perform coronary angiography and PCI in Sweden. It is sponsored solely by the Swedish Health Authorities and receives no commercial funding. The registry’s technology was developed and is administered by the Uppsala Clinical Research Center. Since 2001, SCAAR has used a Web-based case-report platform with automatic data surveillance. In total, 30 hospitals in Sweden, including 9 university hospitals, have cardiac catheterization facilities. In SCAAR, a coronary angiographic procedure is described by about 50 variables and a PCI procedure by about 200 variables. After reviewing the clinical information, the PCI physician immediately enters clinical characteristics and procedural details into the registry. SCAAR obtains data on patients’ vital status continuously from the national death registry. Because the use of personal identification numbers is mandatory, the death registry in Sweden has a high degree of completeness, but it is not reviewed or adjudicated to establish cardiac versus noncardiac causes of death.
CTO was defined as 100% luminal diameter stenosis and absence of antegrade flow of at least 3 months’ duration (known or assumed) (6). The validity of this definition in SCAAR has been confirmed previously (Online Appendix) (1). CAD was defined as luminal narrowing ≥50% on angiography. PCI was considered successful if the residual stenosis was <50%, the stenosis grade decreased by at least 20%, TIMI (Thrombolysis in Myocardial Infarction) flow grade was 3, and there were no serious complications during hospitalization. Proximal CTOs were defined those in segments 1 to 3 (right coronary artery), 5 (left main coronary artery), 6 and 7 (left anterior descending coronary artery), and 11 and 12 (left circumflex artery). CTOs in all other segments were defined as distal.
The study was based on patients who underwent diagnostic coronary angiography and were registered in SCAAR between January 1, 2005, and January 1, 2012 (Figure 1). Only patients diagnosed with significant CAD were included in the analyses. A patient with CTO was identified from available information about the percentage of luminal stenosis at the level of the coronary segments (Online Figure S1, Online Appendix). Information from a diagnostic coronary angiogram could also be used to determine whether a coronary segment was totally occluded. To differentiate between acute and chronic occlusions, we excluded patients who underwent procedures for ACS in whom the 100% occlusion was located in the same coronary artery as the culprit vessel. Patients who underwent procedures in the same vessel within the previous 3 months were also excluded. Patients who had undergone previous coronary artery bypass graft surgery were excluded from analysis, because graft patency could not be determined. The study was approved by the regional ethical board in Gothenburg according to Swedish law and regulations.
To compare the clinical characteristics of the groups, we used the chi-square test for categorical variables and the Mann-Whitney U and Kruskal-Wallis tests for continuous variables not normally distributed. The Shapiro-Wilk test was used to assess normality of distribution. A p value <0.05 was considered to indicate statistical significance. The primary outcome was all-cause mortality. Unadjusted survival was examined with Kaplan-Meier survival curves and the log-rank test. To evaluate the association between CTO and mortality, we used Cox proportional hazards regression models to calculate multivariate-adjusted hazard ratios (HRs). Because SCAAR has a hierarchical structure and patients are clustered by hospital, the assumption of independence between patients was violated. To adjust for this clustering effect, we used multilevel modeling and shared frailty Cox proportional hazards regression (7) as the primary model. We scrutinized the database for missing data and found that a number of variables had missing data. Thus, in addition to the complete case analysis, we used the multiple imputation method to estimate the missing data (8,9). We performed shared frailty Cox proportional hazards regression of the imputed dataset under the assumption that the data were missing at random. For the imputation protocol, we analyzed 20 imputed datasets with the chain-equation method (10) and a predictive-mean matching algorithm, which used the same covariates as the main analysis plus a cumulative hazard and event indicator. The secondary model was complete-case analysis on the basis of shared frailty Cox proportional hazards regression, which included only cases with available data for all variables listed in Table 1. The results from secondary models are presented in the Online Appendix. All tests were 2 sided.
Potential confounders (Table 1) were all entered into the model. Six subgroup analyses were pre-specified for the following patient categories: indication for angiography and PCI (stable angina, unstable angina [UA] or non-STEMI, STEMI, and other), severity of CAD (1-, 2-, and 3-vessel and/or left main coronary artery disease), age, sex, diabetes, and calendar year. The possible effect modification of CTO on risk for dying in the subgroups was analyzed by interaction test. To assess its interaction with CTO, age was examined as both a continuous variable and a factorial variable consisting of 4 age groups (<59, 60 to 69, 70 to 79, and >80 years). Localization of CTO (i.e., left anterior descending coronary artery, left circumflex coronary artery, right coronary artery, and multiple CTOs) and the presence of CTOs in proximal versus distal coronary segments were evaluated by entering these variables separately into the regression. The assumption of proportional hazards for each covariate was reviewed separately by log-minus-log survival plots and by a formal test based on scaled Schoenfeld residuals. Possible multicollinearity between the variables in the model was assessed by calculating the variance inflation factor.
Cumulative hazard was estimated using the Nelson-Aalen test. All analyses were performed with Stata version 13.1 (StataCorp LP, College Station, Texas). Rubin’s (8) protocol was used for the imputation procedure and subsequent Cox proportional hazards regression estimation.
Patients and procedures
CTO and long-term mortality
The mean follow-up time was 3.2 years for the CTO group and 3.1 years for the non-CTO group. No patients included in the study were lost to follow-up. The mortality rate was higher in patients with CTO (unadjusted HR: 1.41, 95% confidence interval [CI]: 1.35 to 1.48; p < 0.001) (Figure 2, Online Figure S2, Online Appendix). After adjustment, CTO remained an independent predictor of long-term mortality (primary model: HR: 1.29; 95% CI: 1.22 to 1.37; p < 0.001; complete-case analysis: HR: 1.27; 95% CI: 1.20 to 1.35; p < 0.001) (Table 2, Figure 2). The risk (per unit time) decreased rapidly in both groups during the first 12 months after angiography; it was lowest between 2 and 3 years after the index procedure but remained constant and substantially higher in the CTO group until the third year (Figure 3). In the non-CTO group, the risk gradually increased from the end of the second year to the fifth year. The interaction between CTO and calendar year was significant (p < 0.001). As shown by trend tests, risk in the non-CTO group gradually decreased throughout the study period. In contrast, risk in the CTO group increased on average by 6.6% each year (HR: 1.06; 95% CI: 1.06 to 1.07; p < 0.001) (Figure 4). Patients with CTO with histories of myocardial infarction had higher mortality risk than those without previous myocardial infarction (HR: 1.39; 95% CI: 1.28 to 1.52; p < 0.001).
Successful versus unsuccessful revascularization of CTO and mortality
During the study period, 6,442 patients underwent PCI for CTO. Successful revascularization of CTOs was achieved in 54.2% of cases and was associated with lower risk for death compared with unsuccessful PCI of CTO (HR: 0.85; 95% CI: 0.73 to 0.98; p < 0.034) (Figure 5).
CTO in patient subgroups
The indications for coronary angiography in SCAAR during the study period were categorized as stable angina, UA or non-STEMI, STEMI, and other. The last group consisted of patients who underwent angiography before valve surgery or for unexplained chest pain, heart failure or cardiomyopathy, pre-transplantation diagnosis and post-transplantation follow-up, arrhythmias, aortic aneurysm or dissection, and cardiac arrest. There was a significant interaction between the presence of a CTO and indication (p < 0.001). As shown by a forest plot, risk increased gradually as the indication for angiography shifted from stable angina to ACS, culminating with the highest risk in patients with STEMI (Figure 6). The primary model with imputation of missing data and complete case analysis showed similar point estimates for HR (other: HR: 1.19; 95% CI: 1.06 to 1.33; p = 0.003; stable angina: HR: 1.18; 95% CI: 1.05 to 1.32; p = 0.005; UA or non-STEMI: HR: 1.49; 95% CI: 1.36 to 1.62; p < 0.001; STEMI: HR: 1.66; 95% CI: 1.51 to 1.84; p < 0.001).
Severity of CAD
The majority (83%) of patients had multivessel disease. Multivessel disease was associated with a higher mortality risk than 1-vessel disease (p < 0.001) (Table 2). We found no interaction between CTO and the severity of CAD (Figure 6).
Patients with multiple CTOs had highest risk for death (Figure 7). We found no difference between CTOs in the left anterior descending coronary artery and those in the left circumflex coronary artery (HR: 0.97; 95% CI: 0.85 to 1.10; p = 0.61) or right coronary artery (HR: 1.10; 95% CI: 0.99 to 1.23; p = 0.08). CTOs in proximal segments were associated with higher risk than CTOs in distal segments (HR: 1.13; 95% CI: 1.03 to 1.23; p = 0.01). Patients with CTOs in distal segments had higher mortality risk than patients without CTOs (HR: 1.18; 95% CI: 1.09 to 1.28; p < 0.001).
One-fifth of all patients with CTO (21.7%) were younger than 59 years of age, and 14.6% were octogenarians. When age was entered in the Cox proportional hazards regression as a continuous variable in interaction with CTO, the HR decreased by about 2% per year of age (HR: 0.98; 95% CI: 0.98 to 0.99; p < 0.001). The interaction between CTO and the 4 different age categories is shown in Figure 6. HR was highest in patients with CTO younger than 59 years and was lowest in octogenarians.
The majority of patients in the study were men (Table 1). However, there were more than 20,000 women, of whom 3,140 had CTO. CTO was associated with an equally negative prognosis in men and women. The interaction between CTO and sex was not significant in the primary model Figure 6.
Diabetes was present in approximately one-fifth of all patients. Women with CTO were more likely to have diabetes (28% vs. 22%, p < 0.001). CTO did not interact with diabetes, suggesting that CTO has a similarly adverse prognosis in patients with and those without diabetes Figure 6.
We investigated the effect of CTO on mortality in 14,441 patients with CTO in the prospective SCAAR database. We found that CTO is associated with moderately increased risk for long-term mortality. Although mortality decreased between 2005 and 2012 in patients without CTO, it increased on average by 6.6% annually in patients with CTO. In patients with CTO, the mortality was highest in younger patients and in those with ACS.
The large number of prospectively followed patients with CTO in SCAAR provided a unique opportunity to evaluate the association between CTO and survival rate both in the whole SCAAR cohort and in several important subgroups. The divergence in the mortality trends of patients with and without CTO has not been previously reported. Because the mortality rate in patients with CAD has fallen substantially over the past 20 years in Sweden (11), this divergence was unexpected and puzzling. At the same time, it emphasizes the clinical importance of CTO. This unanticipated divergence in mortality could reflect an upsurge in the use of coronary angiography in patients with increasingly complex disease in whom CTO is frequently diagnosed. In Sweden, the use of PCI in patients with stable angina and ACS increased steadily from about 18,000 in 2005 to 21,000 in 2012, but the number of PCIs for CTO remained low (∼1,100 per year) during the same period (1). It is also possible that modern pharmacological treatment is less effective in patients with CTO. These 2 factors could at least partly explain the divergent trends in mortality.
Another novel and important observation was the association of CTO with moderately increased mortality risk. Specifically, risk increased gradually, being lowest in patients with stable angina, somewhat higher in those with UA/non-STEMI, and highest in patients with STEMI. This risk gradient strengthens the case for a causal relationship between CTO and mortality. The mechanism for this observation is not established, but our finding is both biologically plausible and intuitively appealing. Most patients (>90%) with symptomatic CTO who have well-developed collateral circulation still have ischemia during exercise (12). An increased ischemic burden due to a CTO, combined with ongoing ischemic cellular damage, could lead to greater contractile dysfunction, larger infarct size, and electrophysiological instability. The importance of a patients with CTO in with myocardial infarction and the effect of hemodynamic instability has been noted in smaller studies (13). The most prominent effect of a CTO on mortality was in patients with STEMI. This finding may reflect more pronounced and rapid development of pathologic post-infarction remodeling, leading to left ventricular dysfunction and heart failure, both of which predict increased risk for sudden death (14,15).
There is unequivocal evidence that increasing age is associated with higher mortality in ischemic heart disease. Some studies suggest that men have a higher risk for dying after ACS, whereas others suggest that younger women may be at highest risk. In line with this reasoning, we modeled the possible interactions between CTO and age and between CTO and sex and found evidence for yet another risk gradient. In this analysis, CTO was associated with highest risk in younger patients, and the risk gradually decreased with advancing age. This novel finding may reflect the shorter life expectancy and the greater frequency of other strong risk factors in older patients. The possible interaction between sex and CTO was not addressed previously. Our analysis, which included more than 3,100 women with CTO, showed that CTO is associated with an equally negative prognosis in both men and women.
Another important observation was that CTO did not modify the adverse prognostic effect of the severity of CAD. In other words, a patient with a CTO has higher risk for death than a patient without CTO who has an equivalent degree of CAD. Previous reports proposed that CTO mainly explains the difference in mortality between patients with single-vessel and multivessel disease. Our findings do not support this view. Although CTO is often seen as the final stage of CAD, our study suggests that CTO can be viewed as a distinct biological property of a patient, because both CAD severity and CTO were independent predictors of mortality in the absence of an effect modification of one variable on the other. Diabetes had an equally adverse prognostic impact in patients with and those without CTO. By contrast, in earlier reports of patients with STEMI, CTO was associated with a worse prognosis in patients with diabetes. The discrepancies between this study and previous reports may reflect differences in statistical modeling for subgroup analysis, statistical power, and selection bias.
We identified a large number of patients (n = 1,920) with CTO in more than 1 vessel. These patients had a worse prognosis than those with CTO in a single vessel. We found no significant difference in the risk between CTO in different coronary vessels. However, within the same vessel, CTOs in proximal segments had worse prognosis than CTOs in distal segments. Even CTOs in distal segments were associated with worse prognosis. The fact that both multiple CTOs and CTOs in both proximal and distal segments were associated with higher risk supports the hypothesis that the negative impact of CTO on survival is mediated by the severity and extent of the ischemic burden. This risk gradient strengthens the causative relationship between CTO and mortality.
Our study provides new and stronger evidence for a possible causal relationship between CTO and increased mortality. At the same time, it poses the question of whether successful revascularization can reduce or eliminate the increase in risk for mortality. It is still a matter of debate whether revascularization in patients with stable angina improves survival. There is some evidence that revascularization is protective in patients with severe CAD, diabetes, and large areas of inducible ischemia (16–20). In contrast, the evidence is much stronger that early revascularization reduces the frequency of adverse cardiovascular events in patients with ACS (21–25). We hypothesize that the clinical benefit and cost-effectiveness are greater after successful revascularization of patients with CTO, particularly in younger patients and in those with ACS. Currently, no data are available from randomized clinical trials to support this hypothesis. In observational studies of patients with CTO, successful revascularization alleviated angina, increased electric stability, reduced the need for coronary artery bypass graft surgery, improved left ventricular function, and increased survival (14,15,26). Our study confirms the results from previous smaller observational studies that successful revascularization of CTO is associated with improved survival. However, the value of these studies, including our own data, for clinical decision making is considerably limited, because they compared successful and failed intervention without including patients on optimal medical treatment as a control group. Furthermore, all observational studies were conducted in patients with stable angina who have lower risk for adverse cardiovascular events and for procedure-related complications than patients with ACS. The OAT (Occluded Artery Trial) study did not demonstrate improvement in clinical outcomes after revascularization of the infarct-related artery in patients with STEMI in stable condition with subacute occlusions (27). The recently completed EXPLORE (Evaluating XIENCE V and LV Function in PCI on Occlusions After ST-Elevation Myocardial Infarction) trial, based on 304 patients with CTO, did not show a benefit of routine revascularization of CTO in nonculprit coronary arteries on left ventricular function and morphology in the subacute phase of STEMI (28). Consequently, the totality of evidence from randomized clinical trials to date speaks against routine revascularization of patients with ACS in stable condition with CTOs in nonculprit coronary arteries.
Currently, 2 randomized controlled trials of PCI in patients with CTO are ongoing: EUROCTO (A Randomized Multicentre Trial to Evaluate the Utilization of Revascularization or Optimal Medical Therapy for the Treatment of Chronic Total Coronary Occlusions) and DECISION-CTO (Drug-Eluting Stent Implantation Versus Optimal Medical Treatment in Patients With Chronic Total Occlusion). These trials will provide additional valuable information about whether PCI is superior to medical therapy in reducing symptoms and improving quality of life, exercise tolerance, left ventricular function and morphology, and safety. However, the trials are not designed to produce decisive evidence about how revascularization affects survival. Under these circumstances, the results of large observational studies such as ours may be useful for clinical decision making and to support treatment recommendations.
First, it was an observational study and as such it provides only associative evidence, not causative. We cannot rule out the possibility of selection bias, residual confounding and survival bias, as only surviving hospitalized patients are included in the registry. In contrast, the observational nature of our study provides real-world data on the largest cohort studied to date.
Second, information on angiographic findings at the segmental level was missing in 42.2% of patients, and this could have introduced bias in risk estimates.
Third, SCAAR does not contain data on pharmacological treatment, the presence of ischemia or viability, and patient frailty, and we were not able to adjust for the possible differences known to affect clinical outcome.
Fourth, we did not have data on cause-specific mortality. Last, we did not have data on the outcomes of revascularization of CTOs with coronary artery bypass graft surgery.
We found that CTO is associated with increased risk for mortality in patients with CAD, especially in younger patients and in those with ACS.
WHAT IS KNOWN? Previous studies based on smaller patient populations have reported that presence of CTOs in coronary arteries is associated with higher mortality.
WHAT IS NEW? Our study, based on 14,441 patients with CTO, confirms previous reports and shows that the risk attributable to CTO is lowest in patients with stable angina and highest in younger patients with STEMI.
WHAT IS NEXT? It is necessary to establish, by means of randomized clinical trials, whether routine revascularization of CTO in patients with stable angina and ACS leads to improved survival.
For a list of participating centers, a description of the SCAAR and SWEDEHEART registries, additional methods and results, and supplemental figures, please see the online version of this article.
Drs. Råmunddal and Jensen have received proctoring honoraria from Boston Scientific. Dr. Henriques has received an unrestricted research grant from Abbott Vascular. Dr. Omerovic had served as an advisory board member for AstraZeneca; has received lecturing fees from Medtronic and AstraZeneca; and has received research grants from Abbott and AstraZeneca. Dr. Harnek has received consulting and proctoring honoraria from Boston Scientific. All other authors have reported that they have no relationships relevant to the contents of this paper to disclose. Drs. Råmunddal and Hoebers contributed equally to this work.
- Abbreviations and Acronyms
- acute coronary syndrome(s)
- coronary artery disease
- confidence interval
- chronic total occlusion
- hazard ratio
- percutaneous coronary intervention
- ST-segment elevation myocardial infarction
- unstable angina
- Received September 21, 2015.
- Revision received March 28, 2016.
- Accepted April 21, 2016.
- American College of Cardiology Foundation
- Task Force on Myocardial Revascularization of the European Society of Cardiology,
- the European Association for Cardio-Thoracic Surgery,
- European Association for Percutaneous Cardiovascular Interventions,
- et al.
- Blankenship J.C.,
- Levine G.N.,
- Bates E.R.,
- et al.
- Claessen B.E.,
- Dangas G.D.,
- Weisz G.,
- et al.
- Mehran R.,
- Claessen B.E.,
- Godino C.,
- et al.
- Rubin D.B.
- Rubin D.B.
- van Buuren S.
- Werner G.S.,
- Surber R.,
- Ferrari M.,
- Fritzenwanger M.,
- Figulla H.R.
- Hoebers L.P.,
- Vis M.M.,
- Claessen B.E.,
- et al.
- Nombela-Franco L.,
- Mitroi C.D.,
- Fernandez-Lozano I.,
- et al.
- Hachamovitch R.,
- Rozanski A.,
- Shaw L.J.,
- et al.
- Pursnani S.,
- Korley F.,
- Gopaul R.,
- et al.
- Shaw L.J.,
- Berman D.S.,
- Maron D.J.,
- et al.
- Fox K.A.,
- Poole-Wilson P.A.,
- Henderson R.A.,
- et al.
- Khan M.F.,
- Wendel C.S.,
- Thai H.M.,
- Movahed M.R.
- Henriques J.P.,
- Hoebers L.,
- Råmunddal T.,
- et al.