Duration Analysis for Coronary Artery Disease Patients With Chronic Chest Pain: An Output From Saudi Arabia

Introduction: Coronary artery disease (CAD) is a persistent public health problem worldwide. Chest pain is one of the perceptible symptoms of the same disease. Literature has found acute chest pain as plausible risk factors for CAD. Nevertheless, none of the study has estimated duration from chronic chest pain to the diagnosis of CAD. The objective of the study was to estimate duration from chronic chest pain to CAD and to assess impact of risk factors on same duration. Methods: Data were obtained from community based study on 17,232 Saudi adults. History of patients about onset of chest pain and other risk factors were inquired. Descriptive measures were obtained by Kaplan-Meier curve. Effect of demographic and clinical factors was assessed by Cox regression models. Results: Out of 24% patients with chest pain, 21% diagnosed with CAD. The average duration was 5 years. About 12% of patients with chest pain diagnosed with CAD after one year. Advancing age, female gender, no exercise and reduced high density lipoprotein (HDL) were significantly hazardous predictors throughout duration from chest pain to diagnosis of CAD. Conclusion: The duration from chest pain to CAD was 5 years. Age, gender, exercise and HDL can be variables of concern to deteriorate hazards of CAD for patients with chest pain.

was also filled which included questions related to history and current activities of the participants. They were interviewed about their demographic profile and investigation of risk factors associated with CAD was also made. Demographic profile comprised of questions about age, gender, marital status, education status, monthly household income, occupation and housing type. Family history, exercise frequency, smoking status, presence of any other symptoms and number of times of hospital admission were the questions asked for assessing probable risk factors of CAD. Height, weight and other anthropometric measurements were taken for physical examination. Moreover, clinical diagnoses were performed by well-trained physicians. These diagnoses included lipid tests, electrocardiography (ECG), exercise tests, coronary angiography, blood pressures, glucose levels etc. The patients were asked about when they had their first chest pain in past. The number of patients who were diagnosed to have CAD was established in finding of one or more of the following criteria: either physician's clinical assessment of the chest pain as angina, previous myocardial infarction (MI), or findings of evidence of previous MI by ECG. An article published in this series reported the prevalence of CAD in KSA was approximately 6%. 13 Age, gender, body mass index (BMI) 14 , waist circumference (WC) 15 , systolic & diastolic blood pressures (SBP & DBP) 16 , fasting blood sugar (FBS) 17 , serum cholesterol (S. Chol.), serum triglycerides (S.TG) and high density lipoprotein (HDL) 18 were the prominent associated factors with CAD among Saudi population. This article is the series of earlier publication of the same research on assessing prevalence of CAD and its risk factors.

Statistical analysis
The statistical analyses were conducted using Stata v. 12.0. The data were converted from snapshot (cross sectional) to time dependent. Duration of chest pain was set as time variable. CAD presence in the patients was set as event variable. The patient who had chest pain in past but had not diagnosed with CAD were censored. The patients, who were diagnosed as CAD patients, had not chest pain in past were truncated. Descriptive duration analysis was computed in terms of mean ± standard deviation and median (inter-quartile range). Frequency and percentages were computed for categorical variable. Bivariate association of factors with CAD among patients with chest pain was assessed by Chi-square test. Kaplan-Meier duration analysis was executed to estimate duration of chest pain on different time span. The duration of chest pain was compared in study population who were adhered with different status of significant risk factors. Mann-Whitney U test and Kruskal-Wallis test followed by Tukey's type non-parametric post-hoc test were employed for the factors with two and at-least three categories respectively. Stratification with the status of CAD presence was also done and two-way non-parametric analysis of variance was run to study effect of risk factors on duration of CAD. To predict the duration of chest pain to CAD in presence of factors and covariates with p value less than 0.20 at univariable analyses, stepwise Cox regression model was run while setting p values of stay and removal as 0.05 and 0.10 respectively. The accuracy of model was checked by Receiver Operating Characteristic (ROC) analysis. Somers' D coefficient was calculated to assess magnitude of accuracy of the model.

Descriptive duration analysis
Out of 17232 patients, 24.1% (n = 4157) endured chest pain in past. Cardiac chest pain was experienced by 884 (85.3%) patients out of 1,036 CAD patients. This also reckoned that 21.2% (884/4157) of patients with chest pain experienced coronary artery disease. The shortest duration from chest pain to the diagnosis of CAD was 1 year and the longest was 40 years. The median duration was 5 years with 2 and 10 were 1 st and 3 rd quartiles of the same respectively. Kaplan-Meier estimator revealed that 88% of the patients with chest pain did not experience CAD by 1 st year (95% CI: 86.5%-89.5%). This endurance reached about half of the above proportion (43%) by 5 years of the duration. All the patients with 17 years of chest pain were diagnosed as CAD patients with duration function of 8.3% (95% CI: 6.6%-10.2%). A similar panorama is suffered by the CAD patients with duration of at least 25 years of chest pain with very low duration function ( Figure 1).

Stratified duration analysis
About 24% of the patients with chest pain endured CAD ( Figure 2). The proportion of CAD was observed significantly more in female (P=0.004), those who did not exercise (P<0.0001), hypertensive (P<0.0001), patients having low HDL (P<0.0001), less triglyceride nitrate (P<0.0001) and low fasting triglyceride (P<0.0001). Table 1 indicated effect of these demographic and clinical risk factors on duration from chest pain to CAD. The mean duration of chest pain for the patients who exercised was 0.55 ± 2.23 which is less than those who did not exercise (P<0.0001). The male patients with chest pain diagnosed with CAD 5 months earlier than female patients (P=0.003). Similarly, diagnosis of CAD for the patients with less than 2.3 triglycerides nitrates delayed more than those who had more than 2.3 values of the same (P<0.005). Comparing the duration from chest pain in those participants who either suffered CAD or not, the patients with CAD had shorter duration ( Table 2). Hypertensive patients had longer duration of encountering CAD than normotensive (P=0.002) whereas patients without CAD had less duration of being hypertensive and suffering chest pain (P=0.002). Male has less duration of chest pain in either cases (CAD or without CAD) (P=0.013). Patients with lowest and highest HDL diagnosed with CAD within half of the year once encountering chest pain while normal HDL delayed the same for about 9 months. This panorama for patients without CAD observed for about 2 years and 1 year respectively (P=0.034). When triglyceride was less than 2.3 the duration of chest pain for CAD free patients     was 2 years and for CAD patients, it was 0.54 years ( Table  2). The increased BMI shortened the duration from chest pain to CAD for both type of patients (P<0.001). On the other hand, increased FTG prolonged the event of CAD as patients having FTG ≤0.6 diagnosed with CAD prior to 2 years (P<0.0001).

Duration Model
Crude and Adjusted hazard ratio obtained by duration model are shown in Table 3 and Table 4 respectively. Table  3 indicated that age was not significantly hazardous for the patient with chest pain. Female had 14% less hazard to endured CAD than males (P=0.004). Hypertension and diabetes had also not significant hazard for patient of chest for having CAD. Others risk factors which were significant on overall status of chest pain patients were not found to be significant in univariable analysis for the hazards of bearing CAD. Multivariable Cox regression via stepwise method revealed different prospect for the hazards of CAD for the patient with chest pain. It stipulated that when effect of other factors accounted for, female had 9% times less likely to have CAD, increased age and exercise delayed the same hazard and higher values of HDL reduced the hazard for chest pain's patients for encountering CAD.
The area under the receiver operating characteristics curve was 64% (Figure 3) indicating moderate accuracy of stepwise model. Somer's D coefficient was also depicted significant magnitude of accuracy of the model (D= 0.261; 95% CI: 0.211-0.310).

Discussion
The population of KSA includes inhabitants from various countries. 19,20 These individuals endure noteworthy    prevalence of CAD. 13 Several studies have been done for studying risk factors of CAD. [21][22][23][24] Wilson et al developed predicted score for the risk of coronary heart disease 25 which is considered to be generalized score for any cardiac patients. 11 The score also predicted first coronary events 26 and determined highest risks of coronary events in elderly age of both men and women. Later, prediction scores for coronary disease via logistic model were built for primary care patient with acute chest pain. 9,10,12 Duration of symptom was considered to be the significant predictors in the model. 12 Furthermore, duration of mortality due to CAD was predicted in Australian population. 27,28 Mingala and Estolano made an attempt to estimate duration of hospital stay for patients suffering MI. 29 As far as authors' knowledge is concerned, the time duration from the first onset of chronic chest pain to the true diagnosis of CAD has never been studied yet. For the analysis of duration data, two study designs have been recommended cross sectional or cohort study designs. Though, cohort study requires time to observe complete duration while cross sectional study needs knowledge of duration of disease for the currently enrolled patients. 30 Current duration survival analysis was introduced by Keiding et al 31 and Ali et al 32 . These gaps incited the idea that we exerted an attempt to analyze the duration of CAD from chest pain from a cross sectional sample.
The study estimated average duration from chest pain to the diagnosis of CAD as 5 years. Mingala and Estolano reported the average duration of hospitalization for CAD patients was 0.5 years. 29 The average duration of CAD mortality was varied from 10 years 33 to 25 years. 34 Combining these outcomes, we can say now that duration from chronic chest pain to the CAD mortality varied from 15.5 years to 30.5 years. Hermingway et al studied prevalence of angina across 31 countries by systematic review. The study reported average weighted prevalence in female was 6.7% while in male, it was 5.7%. In population of Iran, the same prevalence was found to be 4.3%. 3 The prevalence of chest pain in Saudi adults is found to be 24% by this study. This prevalence is also relatively high in male and female residing in KSA. Sarraf-Zadegan et al reported that in Iran cardiac chest pain leading to CAD endured by 19.4% of population. 4 By our study, it was found that 85.3% Saudi population encountered cardiac chest pain leading to CAD. Though, in primary care 6 , study identified 8.2% of cases as CAD in a cohort with chest pain. A study from United States indicated 83.3% prevalence of CAD among patients with chest pain. 35 Similarly, a 300 weeks follow up study 5 reported that CAD was prevalent in 12% of the patients with chest pain. This study also advised that great considerations should be made on patients with chest pain as in this study out of 24620 cohort of patients with chest pain, 672 (2.7%) diagnosed with any one of the thoracic ailments.
Amongst the known risk factors for the development of coronary artery disease, this study indicated certain predictors which were hazardous for the chest pain patients concerning the same disease. Increasing age is significantly hazardous during the time span from chest pain to CAD. Though, this effect is slight similar as reported for overall population at risk of CAD in the earlier article. 13 Advancing age and gender had also considered as significant risk factors for CAD among Iranian population. 36 Exercise was not a significant predictor by regression model in the same study. However, in our study increasing exercise showed significant hazards for CAD for the patients who encountered chest pain. This indicated that patient who encountered chest pain should be vigilant for their exercise habits. These findings can be comparable with another study by Bosner et al. in which age/gender with exercise were the leading predictors for the development of CAD among primary care patients with acute chest pain. 10 HDL was the only metabolic syndromes affected the hazards of CAD to the chest pain patients. The accuracy of the model is moderately good and magnitude of accuracy is acceptable. This indicated that stepwise Cox model for studying duration of chest pain is good enough to apply.

Conclusion
The duration from chest pain to CAD was computed to be 5 years. Age, gender, exercise, and HDL were the significant predictors that best describe the distribution of duration from chest pain to CAD. Physicians should prioritize these factors as variable of concern for the patient with chest pain to avoid risk of having CAD in long haul.

Ethical issues
Approval of study was obtained from Board of Advanced Study and Research of the institute.