Optimal medical therapy, clinical outcome and its predictors in patients with acute coronary syndrome after discharge with secondary prevention medications at University of Gondar Comprehensive Specialized Hospital, North West Ethiopia, 2023: A retrospective follow-up study

Abstract

Background

The life-threatening diseases known as ACS (acute coronary syndrome) continue to produce considerable rates of morbidity and mortality despite breakthroughs in therapy. The study determined clinical outcome and its predictors in patients at the University of Gondar Comprehensive and Specialized Hospital (UOGCSH), North West Ethiopia.

Design, setting, and participants

A retrospective cohort study design was employed at UOGCSH from January 31, 2018 to February 1, 2023. The hospital used a systematic random sampling procedure to select study participants from the medical records of patients in chronic cardiac follow-up clinics.

Main outcome measures

Exposures were optimal medical therapy (OMT) versus non-optimal medical therapy collected from May to August 2023. Descriptive and analytical statistics were employed to compare study groups. A binary logistic regression model was employed to identify candidate variables for further analysis. Cox's proportional hazard model and log-rank test were employed, with a P-value < 0.05 used to evaluate statistical significance. A five-year all-cause mortality after discharge estimate was displayed by using Kaplan–Meier curves.

Results

Among 422 patients with ACS [mean age, 61.56 (SD = 9.686) years; 54.7% male], of whom only 59.2% (250) received optimal medical therapy at discharge. Age ≥ 65, atrial fibrillation, chronic kidney diseases, and cardiogenic shock were negative independent predictors of optimal medical therapy. On the other hand, male sex was independently associated with the use of optimal medical therapy. All-cause mortality occurred in 16.6% (n = 70) and major adverse cardiac events occurred in 30.8% (n = 130) of patients with a 95% CI of 0.132–0.205 and 0.264–0.355, respectively. Multivariate analyses indicated that OMT was significantly associated with reduced all-cause mortality (aHR: 0.431, 95% CI: 0.222–0.835; P = 0.013).

Conclusion

This study revealed that the use of preventive OMT in patients discharged with acute coronary syndrome was associated with a reduction in all-cause mortality. However, the use of this OMT is suboptimal.

Introduction

One of the main causes of mortality globally, cardiovascular illnesses continue to have a significant impact on morbidity, mortality, and healthcare costs [1, 2]. The term "acute coronary syndrome" (ACS) explains a variety of illnesses that are indicative of an abrupt decrease in coronary blood flow and are therefore consistent with acute myocardial infarction [3, 4]. The formation of coronary atherosclerotic plaque, which causes intracoronary thrombosis and ischemia, is the most frequent cause of ACS. Prolonged cessation of coronary blood flow can result in myocyte necrosis or infarction [5, 6]. Despite advancements in therapy, ACS, which encompasses unstable angina and myocardial infarction (MI) with and without ST-segment elevation, remains a life-threatening condition with significant rates of morbidity and mortality [3]. It is estimated that there are 7 million cases of ACS worldwide, making it one of the most prevalent diseases globally [4, 5].

Among the most recent years for which information is available, cardiovascular disease continued to be the leading cause of death among the member nations of the European Society of Cardiology (ESC), accounting for slightly over 1.9 million deaths in men and slightly under 2.2 million deaths in women. These account for 45 percent of all deaths in the case of women and 39 percent in the case of males. The most frequent cause of CVD mortality, ischemic heart disease, accounts for 44% of fatalities from the condition in men and 38% in women [7]. The most prevalent illness in Africa is ACS, with a pooled estimate of 22% for all-cause mortality [6]. The situation is not different in Ethiopia; CVD prognosis is poor, and ACS is increasingly common and leading cause of cardiovascular death, ranging from between 1 and 20% [8]. According to a study conducted at Tikur Anbessa Specialized Hospital (27.4%) [9], and Mekelle (24.5%) [10] of patients died during hospitalization. However, in Ethiopia, there is no data known about optimal medical therapy among patients with the acute coronary syndrome after discharge and clinical outcomes after secondary prevention in patients with ACS.

Early prediction of outcomes is crucial for patients with acute coronary syndrome, as it can greatly improve their prognosis and help control the progression of the disease. While several cardiovascular risk markers have been identified, further research is needed to confirm their usefulness and diagnostic effectiveness [11, 12]. Available data suggests that inflammation plays a prominent role in all stages of coronary atherosclerosis [13]. Prior research has demonstrated that fibrinogen is involved in both the initiation and progression of inflammation by turning on the synthesis of pro-inflammatory proteins through the nuclear factor transcription pathway. In addition to the advancement of atherosclerosis by binding to glycoprotein receptors on platelet surfaces, promoting platelet aggregation [14, 15]. On the other hand, the negative inflammatory protein albumin has a negative correlation with erythrocyte aggregation. In patients with ACS, a decrease in serum albumin levels has been linked to higher mortality rates [16]. Furthermore, a study has revealed that individuals with ACS are more likely to develop poor outcomes if they have a high fibrinogen to albumin ratio (FAR), syntax score II, neutrophil–lymphocyte ratio (NLR), platelet-lymphocyte ratio (PLR), monocyte-to-lymphocyte ratio, high-sensitive C-reactive protein (hsCRP), and low HALP (hemoglobin, albumin, lymphocyte, and platelet) levels below 3.72 [11, 17–21]. Additionally, lifestyle education was also found to be a reliable and independent predictor of worse outcomes [19]. These biomarkers can be easily detected using non-invasive and innovative techniques. However, these biomarkers are not applicable in our setup.

The mortality rate for individuals with ACS has decreased due to the widespread use of revascularization through percutaneous coronary intervention (PCI) and pharmaceutical therapy [22]. In patients with ST-elevation myocardial infarction (STEMI) or non-ST-elevation myocardial infarction (NSTEMI), the American College of Cardiology/American Heart Association (ACC/AHA) recommends aspirin, statins, P2Y12 inhibitors, beta-blockers, and angiotensin-converting enzyme inhibitors (ACEI) or angiotensin receptor blockers (ARB), often referred to as optimal medical therapy (OMT), in patients with STEMI or NSTEMI. The early versions of the ESC guidelines contain a Class I, Level A recommendation for aspirin, P2Y12 inhibitors, and statins in all ACS patients unless contraindicated [23, 24].

Existing evidence showed that OMT prescription at discharge was associated with significantly lower clinical outcomes in patients with ACS [25–30]. However, several reports have shown large gaps between the recommended and prescribed dose and regimen of secondary prevention medications in a patient with ACS after hospital discharge [27, 31–36]. Different studies in Ethiopia determined in-hospital mortality of ACS; however, there are no studies done to assess the use of OMT at discharge and its association with mortality and MACE. Therefore, the study aims to assess OMT, clinical outcome, and its predictors among ACS patients after discharge with secondary prevention medications.

Methods

Study setting and design

The study was conducted at the University of Gondar Comprehensive and Specialized Hospital (UOGCSH). It is found in central Gondar administrative district. The hospital is found 176 km away from the regional city of Bahir Dar and 750 km from the capital city of Ethiopia, Addis Ababa. This is a teaching hospital serving approximately 5–7 million people throughout most of Amhara State and neighboring regions. Gondar town includes 24 kebeles and has Weyina Dega climatic conditions [37]. According to the 2007 Ethiopian census report, the population of Gondar town in 2021 is estimated to be 378,000 [38].

Teaching hospitals is vital to the delivery of community services, research, and education. The facility offers a variety of fields of study, including psychiatry, internal medicine, surgery, gynecology, pediatrics, and HIV care, in addition to more than 500 inpatient beds and outpatient clinics. The study was conducted from January 2018 to February 2023, and the data was collected from May to August 2023. Retrospective cohort study design was employed among ACS patients.

Inclusion and exclusion criteria’s

Inclusion criteria

Adult patients of an age greater than 18 who were admitted to the hospital with a diagnosis of ACS based on the Third Universal Definition of Myocardial Infarction [39]. Moreover, these individuals were discharged with secondary prevention medication between January 31, 2018, and February 1, 2023.

Exclusion criteria

Patients refused to receive secondary prevention medications; missing time variables, no prescribing information, those who were transferred in, those without at least one follow-up, and those who developed re-infarction, CHF, and stroke at the start of the study were excluded from the study.

Sample size determination and sampling technique

Sample size determination

The sample size was calculated by using the single population proportion formula by considering the following: 5% for two-tailed type-I error (Zα/2 = 1.96), 95% confidence interval (CI), assumed the magnitude of all-cause mortality is 50% since none of the study was done in Ethiopia.

$$\begin{array}{cc}\text{Therefore}\text{n}\hfill & ={\left(Z,\frac{a}{2}\right)}^{2}Pq/d^2\hfill \\ \hfill & ={(1.96)}^2(0.5\ast 0.5)/{(0.05)}^2\hfill \\ \hfill & =384\hfill \end{array}$$

where n is the required sample size, Za/2 is the confidence level, p is assumed proportions of all-cause mortality, q is 1-p, and d is the margin of error, n = 384. By taking d = 0.05, p = 0.5, and Zα/2 = 1.96. Then we added 10% contingency by considering the possible missed and lost data.

$$\begin{array}{c}\text{Then},384+38.4=422\hfill \\ \text{Therefore, the final sample size was 422}\hfill \end{array}$$

Sampling technique

A retrospective cohort review of patients with a documented ACS event presenting between January 31, 2018 and February 1, 2023 was conducted, using data from UOGCSH. According to UOGCSH Health Management Information System registries, a total of 1200 ACS patients were treated from January 31, 2018 to February 1, 2023. A systematic random sampling technique was employed to select the study participants (K = 1200/422 = 2.84 ~ 3), then select every 3rd unit. The first patient medical chart was selected by lottery method. The next patient's medical chart was taken if the medical chart was not available and did not meet the inclusion criteria. Then, patient chart was retrieved from chart rooms using the medical registration numbers of each of the patients.

Outcome measurements

Clinical outcomes in this study include all-cause mortality and MACE (heart failure, stroke, and re-infarction). The event of interest was death/MACEs, and the time to death/MACEs was calculated at the time between the dates of discharge and the date of death/MACEs.

Operational definitions

Optimal medical therapy (OMT) was defined as the treatment with a minimum of four drug regimens, which consist of DAPT (aspirin and P2Y12 inhibitors), statins, and beta-blockers.

If comorbidities like hypertension and diabetes are present, OMT is defined as the use of five medications, including ACEIs or ARBs, in addition to the above medications according to ACC/AHA recommendations without contraindication [24, 40, 41], and was defined as reporting taking ≥ 75% of the indicated medications at discharge, which was used as a cut-off point to say OMT [36].

The definition of statin intensity will be based on the ACC/AHA guideline on the management of blood cholesterol [33]. Atorvastatin at 40 or 80 mg daily and rosuvastatin at 20 or 40 mg daily were defined as high-intensity statins, and the other statins were classified as a low-to-moderate intensity statins [23, 27, 42, 43].

Clinical outcome: In this study, defined as:

• ➢ All-cause mortality and MACEs reported by caregivers and/or health care professionals.

Major adverse cardiovascular events (MACEs): MACE in this study includes re-infarction, stroke, and heart failure [27, 44].

Event: Death from ACS was an event of interest.

Time to death: The time to death was calculated at the time between the dates of discharge and the date of death.

Data collection procedures and data quality control technique

A questionnaire form was created using Kobo Toolbox to collect data on the medical records of patients and, if necessary, by phone calls on ACS patients admitted from January 31, 2018, to February 1, 2023, that was adapted from a tool developed by previous studies [25–27, 29, 45, 46], which contains four sections. The first section is regarding socio-demographic characteristics; the second section is for the assessment of ACS; the third section is regarding patient-related characteristics; and the fourth section is regarding treatment-related characteristics.

Patients were selected from the Health Management Information System (HMIS) registration book for accessing the card numbers of ACS patients admitted from January 31, 2018, to February 1, 2023. Then the medical registration numbers were listed, and the list was reentered and checked for duplications by using Microsoft Office Excel 2013 in order to get an independent observation. Their medical charts were searched using this number from the card room. Then, patients’ charts were retrieved and collected from the record and documentation office. A pre-test was done on 5% of the total sample size (21 medical charts) to assess the completeness of the patient charts before the actual data collection time. Data collectors were trained for one day about the objective of the study, the technique of data extraction, and ethical considerations. The principal investigator carried out the supervision. Regular check-ups for the completeness and consistency of the data were done on a daily basis.

Data processing and analysis

The data were exported to Excel and Statistical Package for Social Science (SPSS) version 26 statistical analysis from Kobo toolbox. Then the repeated data were cleaned in Excel before being exported to SPSS. Descriptive analysis (frequency, percentage, mean, and standard deviation) was calculated and presented in tables. A chi-square test was performed to assess the differences in proportions. A binary logistic regression model was employed to identify candidate variables for further analysis. Variables with P-value ≤ 0.25 in the binary logistic regression model were further entered into multivariable logistic regression analysis to identify factors associated with OMT. Assumption of multi-collinearity was met as variance inflation factor is < 10 and tolerance > 0.1 for all variables. The outcomes of logistic regression stated that the selected model was a good logistic regression model fit, since the Hosemer-Lemeshow goodness of fit p-value is greater than 0.05 (OMT; P = 0.13), then fails to reject the null hypothesis, and it is stated that the logistic model is a good fit for the data set. Assumptions like outliers were checked before running the analysis, and the data values had no outliers (cooks distance test 0.071, which was between 0 and 1).

An all-cause mortality estimate was displayed by using Kaplan–Meier curves. Multivariate Cox proportional hazard models and log-rank tests were conducted to estimate adjusted hazard ratios (aHR) if there are significant changes in clinical outcome between different sets of predictor variables, and 95% confidence intervals were used to evaluate statistical significance [47]. All candidate variables would be from the patients’ baseline characteristics. Variables with a p-value less than 0.05 were declared to be associated with clinical outcomes. During the retrospective follow-up period, any recordings that were lost or where an individual did not experience the event of interest by the end of the follow-up were considered censored. For time-to-all-cause mortality, the proportional hazards assumption was checked using the Schoenfeld residuals test (global test, P-value = 0.8297), complementary log–log plots, graphical parallel assumption test, and Cox-Snell residual was used to test model goodness of fit. Age, dose of medications, and Killip class were among the potential confounding factors that were taken into account by multivariate analysis and stratification in our primary analyses.

Ethical consideration and consent to participate

An official permission letter was obtained from the University of Gondar's specialized referral hospital administrative office, and ethical clearance was obtained from the School of Pharmacy's Ethical Review Committee and the University of Gondar Institutional Review Board (IRB) with the approval number of SOPS/276/2023. Permission to access the medical charts of patients was obtained from the UOGCSH clinical directorate. Participants provided informed consent, both written and verbal, stating their desire to participate in the study after being informed of its aim. Confidentiality was ensured through no patient’s name and identifier was taken; rather, codes were used as identifiers. Information regarding patients was used for the study purpose only.

Result

Socio-demographic characteristics of acute coronary syndrome patients

A total of 422 ACS patient charts were included in the study. The mean ages of patients were 61.56 ± 9.686 years, with a range of 34–84 years, and 54.7% of the patients were men. Of the total study subjects, 57.6% were getting health insurance coverage, and 44.3% were urban residents (Table 1).

Table 1.

Socio-demographic characteristics of acute coronary syndrome patients in UOGCSH from January 31, 2018 to February 1, 2023 (n = 422)

Variables	Category	Frequency n (%)
Sex	Male	231 (54.7)
	Female	191 (45.3)
Age (years)	≤ 55	134 (31.8)
	56–64	139 (32.9)
	≥ 65	149 (35.3)
Residency	Rural	235 (55.7)
	Urban	187(44.3)
Health insurance coverage	Yes	243 (57.6)
	No	179 (42.4)

Symptoms on admission, clinical characteristics, killip class, and comorbidities of acute coronary syndrome patients

The frequently recorded patients’ symptoms on admission were chest pain alone (50.5%), followed by chest pain and nausea and vomiting (24.6%). Out of 422 patients with evidence of Killip class, 32.2% of them had Killip class II and 28.2% had Killip class III. Hypertension was the most prominent comorbidity in ACS patients (43.1%), followed by diabetes (28.4%). The mean length of hospital stay was 9.07 ± 3.86 days, with a minimum of two days and a maximum of 20 days.

Of the total study subjects, 15.9% had elevated systolic blood pressure in the range of 130–139 mmHg, and 24.6% of patients had elevated diastolic blood pressure in the range of 80–89 mmHg (Table 2).

Table 2.

Symptoms on admission, clinical characteristics, Killip class and risk factors of acute coronary syndrome patients in UOGCSH from January 31, 2018 to February 1, 2023 (n = 422)

Variables	Category	Frequency n(%)
Symptom on admission	Chest pain	213 (50.5)
	Chest pain and NV	104(24.6)
	Chest pain and SOB	90 (21.3)
	SOB	2 (0.5)
	othersa	13 (3.1)
Length of hospitalization (days)	< 7	178 (42.2)
	7–14	200 (47.4)
	≥ 15	44 (10.4)
SBP (mmHg)	< 90	2 (0.5)
	90–119	281 (66.6)
	120–129	38 (9)
	130–139	67 (15.9)
	≥ 140	34 (8.1)
DBP (mmHg)	< 60	181 (42.9)
	60–79	127 (30.1)
	80–89	104 (24.6)
	≥ 90	10 (2.4)
HR (beat/min)	< 60	10 (2.4)
	60–100	397 (94.1)
	> 100	15 (3.6)
Smoking status	Yes	26 (6.2)
	No	396 (93.8)
Killip class	Class I	50 (11.85)
	Class II	136 (32.2)
	Class III	119 (28.2)
	Class IV	117 (27.73)
Cardiac catheterization	Yes	12 (2.8)
	No	410 (97.2)
Comorbidities	Hypertension	182 (43.1)
	Diabetes mellitus	120 (28.4)
	Dyslipidaemia	93 (22.0)
	Cardiogenic shock	73 (17.2)
	Chronic kidney disease	49 (11.6)
	Obesity	13 (3.1)
	Atrial fibrillation	36 (8.5)
	Family history of CAD	44 (10.5)
	Othersb	22 (5.2)

CAD Coronary artery disease, SOB Shortness of breath, NV Nausea and vomiting, SBP Systolic blood pressure, DBP Diastolic blood pressure, mmHg Millimeters of mercury, HR Heart rate

Others^a: easy fatigability, palpitation, abdominal pain and syncope

Others^b: Anemia, Pneumonia, Thyroid disease, HIV, and Seizure

Profile of laboratory findings and diagnostic investigations of acute coronary syndrome patients

In the case of laboratory and diagnostic investigations, as can be seen from the data in Table 3, from recorded investigations, about one-fifth (20%) had elevated low-density lipoprotein. Total cholesterol was higher than normal value in 11.9% of the patients. About 35.6% had lower-than-normal values of high-density lipoprotein and 12.5% had elevated triglyceride levels.

Table 3.

Profile of laboratory findings and diagnostic investigations of acute coronary syndrome patients in UOGCSH from January 31, 2018 to February 1, 2023(n = 422)

Variable	Category (mg/dl)	Frequency n (%)
Total cholesterol	< 200	141 (88.1)
	≥ 200	19 (11.9)
Triglyceride	< 150	140 (87.5)
	≥ 150	20 (12.5)
HDL	< 40	26 (35.6)
	≥ 40	47 (64.4)
LDL	< 100	124 (80)
	≥ 100	31 (20)
Laboratory value	Mean ± SD	Reference range
Serum creatinine	0.95 ± 0.17	0.65–1.2 mg/dl
Troponin level	0.48 ± 0.13	0–0.04 ng/dl

HDL High density lipo-proteins, LDL Low density lipo-proteins

Regarding the class of diagnosis, 256 (60.7%), 105 (24.9%), and 61 (14.5%) were classified under NSTEMI, STEMI, and UA, respectively (Fig. 1).

Fig. 1.

Class diagnosis for acute coronary syndrome patients in UOGCSH from January 31, 2018 to February 1, 2023(n = 422)

Medication given at discharge

Regarding discharge medications, most of the patients (413, 97.9%) had been prescribed aspirin and statins. About 345 (81.8%) patients were discharged with beta-blockers, and clopidogrel was prescribed on 311 (73.7%). ACEIs/ARBs and calcium channel blockers were documented in 213 (50.5%) and 67 (15.9%) patients, respectively (Fig. 2).

Fig. 2.

Discharge medications of acute coronary syndrome patients admitted from January 31, 2018 to February 1, 2023(n = 422)

Optimal medical therapy (OMT) and clinical outcomes

A total of 250 (59.2%) patients were discharged with all indicated drugs (OMT), 95% CI: (54.4–64.0%). Concerning clinical outcomes, from the total of 422 patients who were admitted during the five years, 70 (16.6%) patients died with a 95% CI of 13.2–20.5%. Moreover, 130 (30.8%) patients had experienced MACEs with a 95% CI of 26.4–35.5%; from this, 14.2%, 11.4%, and 6.8% were stroke, CHF, and myocardial re-infarction, respectively (Table 4).

Table 4.

Optimal medical therapy and clinical outcomes in patient with acute coronary syndrome admitted from January 31, 2018 to February 1, 2023(n=422)

Outcomes			On OMT		Frequency n (%)	95% CI (%)
			Yes	No
Death		Yes	24	46	70 (16.6)	13.2-20.5
		No	226	126
MACE		Yes No	17 233	113 59	130(30.8)	26.4-35.5
	Stroke	Yes No	7 243	53 119	60 (14.2)	2.90-14.7
	CHF	Yes No	5 245	45 127	50 (11.4)	3.60-18.5
	Re-infarction	Yes No	5 245	17 155	22 (5.2)	4.90-35.5

MACEs Major adverse cardiovascular events, CHF Congestive heart failure; and OMT Optimal medical therapy

Factors associated with optimal medical therapy

Among variables tested for association in bivariable analysis, age, sex, residence, health insurance coverage, smoking status, diabetes, cardiogenic shock, family history of coronary artery disease (CAD), atrial fibrillations (AF), and chronic kidney disease (CKD) were variables with a P-value ≤ 0.25 and candidates for multivariable analysis. Patients with age greater than or equal to sixty-five were 0.488 times less likely to obtain OMT as compared to patients with age less than or equal to fifty-five [AOR = 0.488, 95% CI (0.261–0.912), P value = 0.025]. Male patients with acute coronary syndrome were 1.919 times more likely to be discharged with OMT as compared to female patients [AOR = 1.919, 95% CI (1.125–3.273), P value = 0.017].

Regarding comorbidities, acute coronary syndrome (ACS) patients with cardiogenic shock were 0.552 times less likely to be discharged with OMT as compared to patients without cardiogenic shock [AOR = 0.552, 95% CI (0.306–0.996), P value = 0.049]. ACS patients with AF were 0.187 times less likely to be discharged with OMT as compared to patients without AF [AOR = 0.187, 95% CI (0.081–0.433), P value = 0.001]. Moreover, ACS patients with CKD were 0.257 times less likely to be discharged with OMT as compared to patients without CKD [AOR = 0.257, 95% CI (0.123–0.537), P value = 0.001] (Table 5).

Table 5.

Multivariable logistic regression analysis of factors associated with optimal medical therapy at discharge in patient with acute coronary syndrome admitted from January 31, 2018 to February 1, 2023(n = 422)

Variables	Category	OMT (N)		COR (95% CI)	AOR (95% CI)	P-value
		Yes	No
Age	≤ 55	89	34	1	1
	56–64	91	42	0.828(0.483–1.418)	1.00(0.554–1.806)	0.999
	≥ 65	70	96	0.279(0.169–0.46)	0.488(0.261–0.912)	0.025*
Sex	Male	164	67	2.989(1.998–4.470)	1.919(1.125–3.273)	0.017*
	Female	86	105	1	1
Residence	Rural	149	86	1	1
	Urban	101	86	0.678(0.458–1.008)	1.294(0.801–2.090)	0.292
Health insurance coverage	Yes	162	81	2.068(1.392–4.718)	1.456(0.918–2.307)	0.110
	No	88	91	1	1
Smoking status	Yes	19	7	1.939(0.797–4.718)	0.906(0.344–2.385)	0.841
	No	231	165	1	1
Comorbidities
Diabetes	Yes	64	56	0.713(0.465–1.092)	0.782(0.482–1.267)	0.318
	No	186	116	1	1
Cardiogenic shock	Yes	27	46	0.332(0.197–0.559)	0.552(0.306–0.996)	0.049*
	No	223	126	1	1
Family history of CAD	Yes	21	23	0.597(0.319–1.116)	0.933(0.45–1.937)	0.853
	No	228	149	1	1
AF	Yes	10	26	0.234(0.110–0.499)	0.187(0.081–0.433)	0.001*
	No	240	146	1	1
CKD	Yes	13	36	0.207(0.106–0.404)	0.257(0123–0.537)	0.001*
	No	237	136	1	1

CKD Chronic kidney disease, CAD Coronary artery disease, AF Atrial fibrillation, COR Crude odds ratio, AOR Adjusted odds ratio

Predictors of all-cause mortality in patients with ACS

Multivariate Cox's proportional hazard model was applied to determine the relationship between different covariates at discharge on 5-year all-cause mortality. In the analysis, only one variable was found to be an independent predictor of all-cause mortality in patients with ACS. Overall, OMT prescription at discharge was associated with a reduction in 5-year mortality (AHR = 0.431; 95%CI: 0.222–0.835; P = 0.013) and no significant difference in MACE (AHR = 0.812; 95%CI: 0.403–1.637) (Table 6).

Table 6.

Multivariable cox-regression analysis of covariates associated with the predicted all-cause mortality of ACS patients in UOGCSH admitted from January 31, 2018 to February 1, 2023(n = 422)

Variables	Category	Death		CHR (95%CI)	AHR (95%CI)	P-value
		Yes	No
Killip class	Class I	3	47	1	1
	Class II	11	125	0.763(0.204–2.848)	0.836(0.149–4.687)	0.838
	Class III	11	108	1.055(0.293–3.796)	0.537(0.09–3.207)	0.495
	Class IV	45	72	1.197(0.370–3.872)	1.067(0.222–5.140)	0.936
Diagnosis	STEMI	47	63	1.564(0.842–2.004)	2.075(0.743–5.793)	0.164
	NSTEMI	10	231	1.667(0.727–3.823)	1.664(0.541–5.122)	0.375
	UA	13	58	1	1
Hypertension	Yes	42	140	0.717(0.438–1.173)	0.707(0.397–1.259)	0.239
	No	28	212	1	1
Cardiogenic shock	Yes	55	18	1.236(0.696–2.195)	0.978(0.357–2.682)	0.966
	No	15	334	1	1
Cardiac catheterization	Yes	8	6	1.996(0.936–4.256)	1.354(0.583–3.144)	0.481
	No	62	346	1	1
Dose of statin	40–80	20	312	0.555(0.326–0.944)	0.667(0.324–1.371)	0.270
	< 40	50	40	1	1
OMT initiated	Yes	24	226	0.362(0.205–0.638)	0.431(0.222–0.835)	0.013*
	No	46	126	1	1
Length of hospitalization (day)	< 7	9	138	1	1
	7–14	12	183	1.212(0.510–2.882)	2.564(0.748–8.794)	0.134
	≥ 15	49	32	1.528(0.746–3.131)	1.321(0.605–2.887)	0.485

CHR Crude hazard ratio, CI Confidence interval, AHR Adjusted hazard ratio

^*Significant

Survival estimate for time to death by life table

Totally, 422 adult ACS patients were followed for a five-year period, with the cumulative probability of survival at year one, at year two, and at year three being 0.8429 ± 0.0435 (95% CI; 0.7342, 0.9097), 0.5571 ± 0.0594 (95% CI; 0.4335, 0.6641), and 0.3286 ± 0.0561 (95% CI; 0.2223, 0.4388), respectively (Table 7).

Table 7.

life table showing the probability of survival among ACS patients in UOGCSH admitted from January 31, 2018 to February 1, 2023(n = 422)

Years	Patient at risk	Number of deaths	Censored	Cumulative survival	SD. error	[95% CI]
1	422	17	55	0.8429	0.0435	0.7342 0.9097
2	350	19	89	0.5571	0.0594	0.4335 0.6641
3	242	17	64	0.3286	0.0561	0.2223 0.4388
4	161	11	146	0.1429	0.0418	0.0733 0.2348
5	4	6	4	0.0429	0.0242	0.0114 0.1090

SD Standard deviation, CI Confidence interval

Survival estimate for time to death by Kaplan–Meier graph

In this study, a total of 70 (16.6%) ACS patients died during a 5-year period, and most patients died in the second and third years of life. The mean survival time was 3 ± 0.302 years with CI (2.407–3.593) in the OMT group and 1.861 ± 0.147 years with CI (1.572–2.150) in the non-OMT group. The median survival time of the OMT group was 3.5 ± 0.395 with CI (2.725–4.275) and 1.5 ± 0.179 with CI (1.149–1.851) in the non-OMT group. The time to all-cause mortality and time to MACE were also determined using the Kaplan–Meier survival estimate, and the fulfillment of the hazard assumptions was confirmed using the log–log plot (Fig. 3).

Fig. 3.

Kaplan Meier curve of (A) 5-year survival, and (B) major adverse cardiac events (MACE) between OMT and non-OMT group

Lastly, a Schoenfeld residual test (also known as the global test) was used to verify the hazard proportional assumption. As a result, the assumption was satisfied, and the global test's p-value, or Schoenfeld residual test, was 0.8297.

Discussion

To our knowledge, this is the first study to determine optimal medical therapy (OMT) prescription, clinical outcome, and associated factors among acute coronary syndrome (ACS) patients in Ethiopia. Based on this, we found that only 59.2% of patients received OMT at discharge. Age ≥ 65, atrial fibrillation, chronic kidney diseases, and cardiogenic shock were negative independent predictors of optimal medical therapy. On the other hand, male sex was independently associated with the use of optimal medical therapy.

In this study, the rate of OMT prescription was found to be 59.2% (95% CI = 54.4%, 64.0%). This finding was consistent with previous studies done in France (57.4%) [26, 29, 48]. But it was higher than those studies done in Sudan (45.2%), Thailand (42.6%), the Netherlands (43.2%, 43.7%), Iraq (53.5%), China (15.1%, 36%, 36.2%, 39.8%), and India (46%), respectively [25, 27, 30, 35, 46, 49–53]. This might be due to different reasons; first, in these studies, the definition of OMT is the use of five medications: aspirin, clopidogrel/P2Y12 inhibitors, beta-blockers, statins, and ACEIs/ARBs. Second, investigators from other low- and middle-income countries have also evaluated the use of optimal medical therapy in their registries. For example, the use of optimal medical therapy at discharge, defined by the concurrent use of aspirin, β-blockers, statin, and ACE-I/ARBs, was evaluated in the Clinical Pathways for ACS in China registries [54]. However, use of clopidogrel given its class I recommendation for all patients [33], which is not the case with ACE-I/ARBs, which requires concomitant heart failure, left ventricular systolic dysfunction, diabetes mellitus, or hypertension for a class I indication [24, 33]. The rate of OMT prescription in this study was lower than those studies done in Korea (63.2%, 75.71%), the Netherlands (69.1%), and Australia (65%), respectively [55–58]. The possible reason might be in OMT prescription at discharge in ACS patients among different studies: is local context, study time point, the severity of diagnosis, health coverage, and medication availability [27, 49, 53]. This suboptimal use of OMT leads to mortality and morbidity, as well as increased healthcare costs regarding hospital admission.

The current findings showed that elderly patients were less likely discharged with OMT than younger patients. This finding was in accord with studies done in Canada [29, 59], Korea [48], and the southern US [60]. This might be due to elderly patients exhibiting higher rates of comorbidities and having higher rates of adverse events [55, 59, 60]. Statins were found to be equally effective in lowering the risk of cardiovascular death, ACS, stroke, or coronary revascularization in patients 75 years of age or older, according to a recent meta-analysis [61]. The findings were also consistent across statin trials. Even with very low low-density lipoprotein levels, statins have no effect on patient-reported cognition [62], according to the most recent data, despite post-market studies raising concerns about cognitive damage, particularly in the aging population. However, recent emerging evidence showed no difference in efficacy and safety outcomes in relation to age [63–65]. Moreover, the existing evidence recommends OMT for all patients without considering age differences. Therefore, OMT should be considered for elderly patients with ACS with the exception of contraindications [23, 33, 40, 42, 66].

Regarding sex, our study indicated that male patients were more likely to be discharged with OMT than female patients. This outcome is consistent with research conducted in Australia [67–69], West China (17), Italy (54), Thailand [27], the Netherlands [67, 70], and Malaysia [71, 72]. Since females with ACS were less likely to present with chest pain and diaphoresis but more likely to present with shoulder pain or gastrointestinal symptoms. This could be explained by the fact that female patients are frequently older and more complex with multiple baseline adverse comorbidities [67–70, 73, 74]. Furthermore, females had higher rates of baseline renal dysfunction, which contributed to the lower rates of ACEI/ARB prescriptions. As a result, the advantages of OMT may be underestimated [70–72]. Despite the long-term benefits of ACEI/ARBs being well established, many physicians are still hesitant to prescribe them to patients with renal dysfunction. In addition, the lower rate of dual antiplatelet therapy (DAPT) prescriptions among females in this study is due to a fear of bleeding complications. However, studies have demonstrated there is no sex difference in major bleeding risk at 12 months with DAPT [67, 75].

This study found that CKD patients had a lower likelihood of receiving an OMT during discharge. The study findings concurred with research from Thailand and Canada [29]. This could be because using ACEI/ARBs increases the risk of side effects and may require precautions against hyperkalemia or acute renal injury [27]. Patients with chronic kidney disease (CKD) should typically have a renal function test in order to alter their dosage, as most secondary preventive drugs have different effects on the kidney. Nonetheless, this study brought to light concerning issues with ACEI/ARB usage in ACS patients. Only 50.5% of the study participants were discharged with an ACEI or ARBs; most of them had hypertension or diabetes. Moreover, even though statin use has been shown to lower mortality in patients with or at risk for ACS, patients with CKD who have ACS are less likely to receive this treatment [76–78]. Data from the National Registry of Myocardial Infarction 4 indicate that statin administration within the first 24 h of hospitalization for acute myocardial infarction significantly lowers the rate of early complications and in-hospital mortality, in addition to the long-term benefits of statin therapy in ACS patients [79]. In fact, statins seem to mitigate the effects of inflammation, endothelial dysfunction, and coagulation problems, which are all linked to acute myocardial infarction [80]. There is uncertainty regarding the causes of this under-use of OMT in ACS patients with CKD. Potential explanations include worry about harmful side effects and additional deterioration of renal function. Moreover, there is more co-morbidity in CKD patients, which increases their contraindications to these medications.

Our study revealed that patients with atrial fibrillation (AF) were less likely to be discharged with OMT as compared to patients without AF. This result is consistent with studies done in Thailand [27] and Serbia [81]. Compared to patients without AF, those with AF had a lower likelihood of receiving beta-blockers, according to the GISSI-3 data [82]. In this study, digoxin and antiarrhythmic medications were used more frequently to treat AF patients. Another investigation revealed that patients with AF who had never been diagnosed before had a lower chance of receiving clopidogrel [83]. AHA/ACC and ESC guidelines recommend the use of triple therapy for patients with atrial fibrillation, consisting of DAPT and warfarin or a non-vitamin K anticoagulant (NOAC) or dual therapy (P2Y12 inhibitors plus warfarin or NOACs) to prevent stroke in patients who are more likely to experience thromboembolism-related problems [41, 81, 84, 85]. However, double therapy may be used in place of DAPT, which has an impact on the percentage of DAPT prescriptions in this study.

Our study also suggested that patients who experienced cardiogenic shock had a lower likelihood of being discharged with OMT. The finding was similar to the study done in six Middle Eastern countries [32]. The possible reason might be due to the presence of patients who are comparatively contraindicated. These patients may have started off hemodynamically unstable but later improved and became candidates for starting B-blockers. Therefore, the commencement of beta-blockers, which may have begun at a low dose and increased based on the patient's state, was overlooked for these individuals when their hemodynamic stability and progress were seen. Another possible cause is a disruption in the medicine delivery [10]. Additionally, because this study was retrospective, there was a risk that the number of real contraindications may not have been recorded in the patient files, lowering the actual secondary prevention.

In terms of clinical outcomes, over a 5-year period, approximately 16.6%, 95% CI (13.2%, 20.5%), of ACS patients died, and 30.8%, 95% CI (26.4–35.5%), had MACE, which included 11.4% of CHF cases, 14.2% of stroke cases, and 5.2% of re-infarction cases. Overall, OMT prescription at discharge was associated with a reduction in 5-year mortality (AHR = 0.431; 95%CI: 0.222–0.835; P = 0.013).

Regarding to all-cause mortality, our finding is in accord with studies done in Israel (15.8%) [86], Netherlands (16.7%) [87], and Australia (16.8%) [88]. The results were higher than studies done in China (6.8%) [35], and Japan (7.1%, 6.3%) [89, 90]. On one hand, it might be due to sample size variation; in this study, only 422 patients were included, whereas the sample size in China, Australia, and Japan was larger. The other possible reasons might be due to the non-availability of thrombolytic drugs, non-adherence to the recommended guidelines, lifestyle issues, non-availability of CABG (coronary artery bypass graft), PCI (percutaneous coronary interventions) procedures, and P2Y12 inhibitors (ticagrelor, prasugrel) in the study area, which could all be the possible reasons.

Overall, our study revealed that patients who were discharged from the hospital with an optimal medical therapy (OMT) regimen consisting of four medications—aspirin, clopidogrel/P2y12 inhibitors, beta-blockers, statins, or the use of ACEI/ARBs on the top of the above medications if the patient had comorbidities like diabetes or hypertension—had a lower all-cause mortality rate. The present findings align with previous studies that have demonstrated improved outcomes for patients receiving OMT at the time of discharge [25, 26, 29, 35, 36, 48, 91, 92]. Optimal medical therapy (OMT) had been associated with lower MACE, according to a study conducted in Italy [92]. However, in this study, there was no statistically significant difference in MACE between the OMT and non-OMT groups. This study supports the use of OMT with all four drugs in combinations, as OMT was associated with a 56.9% reduced risk of all-cause mortality in patients with ACS.

Study strength and limitations

It is important to emphasize this study's advantages. This is the first study that assessed secondary prevention medication after ACS for a five-year period in Ethiopia. Because the research is based on actual data, it offers a reliable representation of the therapeutic advantages of OMT regimens. The assessment of the causal link between OMT prescriptions and therapeutic benefit on mortality yielded reliable findings with strong statistical power.

This study has some limitations. Due to the retrospective nature of the study design, it was not possible to capture important data from the patient charts, such as contraindications and pertinent laboratory values. Second, the study was conducted at a single center, which may limit its external validity and scientific rigor in justifying broad practice modifications. It is unclear if the patients took the prescribed drugs, as many had multiple comorbidities and were taking various medications that are associated with poor adherence. However, a retrospective analysis found no significant difference in medication adherence rates at 3 months and 1 year after discharge for patients with acute myocardial infarction [93].

Conclusion

This study found that the use of optimal medical therapy (OMT) in patients discharged with acute coronary syndrome (ACS) was associated with a decrease in overall mortality. However, the utilization of OMT remains suboptimal, with only 59.2% of patients being discharged with this treatment. The research also revealed that patients aged 65 years and older, those with chronic kidney disease (CKD), atrial fibrillation (AF), and cardiogenic shock were less likely to receive OMT at discharge. On the other hand, male ACS patients were more likely to receive OMT. Therefore, it is crucial for healthcare providers to actively work towards optimizing the use of OMT at discharge, especially in the early stages of the illness.

Authors’ contributions

GWG assisted with project management and conceptualization of the study. TTA, AFB, YAW, GTG, and SSA contributed to the study designs, supervision and clinical concepts. Finally, MEM, SF, and ADG contributed to the document writing. All authors participated in the data analysis and interpretation, as well as the final document production, review, and editing. GWG is the guarantor.

Funding

This research received no specific grant from any funding agency in the public commercial or not-for-profit sectors.

Availability of data and materials

Information is accessible upon justifiable request. The corresponding author can provide all pertinent information upon reasonable request, including the data and materials supporting the study's findings.

Data availability

No datasets were generated or analysed during the current study.

Declarations

Ethics approval and consent to participate

An official permission letter was obtained from the University of Gondar's specialized referral hospital administrative office, and ethical clearance was obtained from the School of Pharmacy's Ethical Review Committee and the University of Gondar Institutional Review Board (IRB) with the approval number of SOPS/276/2023. Permission to access the medical charts of patients were obtained from UOGCSH clinical directorate. Participants provided informed consent, both written and verbal, stating their desire to participate in the study after being informed of its aim. Confidentiality was ensured through no patient’s name and identifier was taken; rather, codes were used as identifiers. Information regarding patients was used for the study purpose only.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.