Predicting outcomes in acute coronary syndrome using biochemical markers

Abstract

Objectives

To assess risk prediction in patients with acute coronary syndrome (ACS) during the hospital stay, at 6 weeks and at 6 months period using high sensitivity C-reactive protein (hs-CRP), serum creatinine, cardiac troponin I, creatine kinase total, and MB levels.

Methods

It was a prospective observational study. The primary outcome was taken as all-cause mortality. Patients with ACS were enrolled and followed up at 6 weeks and 6 months duration from the index event. Mortality and cause of death were recorded. The hs-CRP was estimated on admission, at 6 weeks, and at 6 months. The estimated glomerular filtration rate (eGFR) was calculated using the abbreviated modification of diet in renal disease (MDRD) formula at admission, at 6 weeks, and 6 months.

Results

There were a total of 108 cases of ACS in the duration of 6 months who completed the follow-up. The hs-CRP level of >5 mg/dl was highly significant for predicting mortality during hospital stay and at 6 weeks (p < 0.001). There was 11% of in-hospital mortality (p < 0.001). At 6 months, the overall mortality was 28% (p < 0.001). There was a statistical significance with low eGFR (median eGFR 45 ml/min/1.73 m²) levels during the admission.

Conclusion

hs-CRP levels above 5 mg/dl and the eGFR levels ≤30 ml/min/1.73 m² were significant in predicting mortality of the patients with ACS. This may provide simple assessment tools for predicting outcome in ACS in resource-poor settings if validated further.

1. Introduction

Cardiovascular disease (CVD) is the prevailing non-communicable cause of death and disability in the Indian subcontinent, and will become the prevailing overall cause of mortality among the inhabitants of South Asia in the next 20 years. The current epidemic and imminent growth are due to the huge burden of CVD risk factors, largely driven by urbanization.1, 2 Although we do not have any national data on ischemic heart disease (IHD), it was found that the prevalence of CVD is increasing and there is a fivefold increase in the incidence of coronary artery disease (CAD).³

The significance of the contribution of laboratory medicine to clinical cardiology has grown in importance over the years. This is witnessed by the recent incorporation of biomarkers into new international guidelines and in the re-definition of myocardial infarction (MI). There are mainly two classes of indicators: markers of early injury/ischemia and markers of inflammation and coronary plaque instability and disruption.⁴

There are various biomarkers associated with acute coronary syndrome (ACS). The clinical application of cardiac biomarkers in ACS is no longer limited to establishing or refuting the diagnosis of myocardial necrosis. Cardiac biomarkers provide a convenient and noninvasive means to gain insights into the underlying causes and consequences of ACS that mediate the risk of recurrent events and may be targets for specific treatment.⁴ Biochemical markers play a major role for risk assessment in patients with an ongoing non-ST segment elevation ACS. Although the cardiac troponin in particular is generally recognized as an important risk indicator, other markers of left ventricular performance (i.e. N-terminal pro-brain natriuretic peptide), inflammation (i.e. C-reactive protein), and renal function (i.e. estimated glomerular filtration rate (eGFR)] are equally important in providing strong prognostic information.5, 6

With the availability of highly specific and sensitive methods for evaluating myocardial tissue damage, such as the immunoassays for MB isoenzyme of creatine kinase (CK MB), myoglobin, and especially, cardiac specific troponin T and I (cTnT and cTnI) and their introduction in clinical practice, the definition of acute myocardial infarction (AMI) has radically changed.⁷

The information gathered from this study would help us in various ways. Firstly, it will help us predict the outcomes of the patients with ACS using the commonly used biomarkers and thus the treatment of the patient can be guided. The demographic information provided by this study shows us that ACS is increasing in Nepal, especially in the young population. This study provides us a complete profile of patients with ACS.

2. Objective

The study was undertaken to assess risk prediction in patients with ACS during the hospital stay, at 6 weeks, and at 6 months period using high sensitivity C-reactive protein (hs-CRP), serum creatinine, cTnI, and CK MB fraction.

3. Materials and methods

3.1. Study design

The study was hospital-based prospective observational study, which was more practical in our setting and ethically acceptable.

3.2. Setting

This study was conducted in the Department of Internal Medicine at B.P. Koirala Institute of Health Sciences, Dharan, Nepal for a period of 1 year.

3.3. Primary outcome

All-cause mortality.

3.4. Inclusion criteria

All the patients meeting the diagnostic criteria of ACS, who were admitted under the Department of Internal Medicine at B.P. Koirala Institute of Health Sciences and who gave consent, were enrolled for this study.

3.5. Exclusion criteria

Alternate diagnosis of chest pain and/or refusal to give consent for the study.

For the purpose of the study, different components of ACS were defined as following:

• (1)
  ST elevation Myocardial Infarction (STEMI):

  • (a) Characteristic rise and fall of cardiac biomarkers (presence of troponin I, and/or rise of CK NAC and CK MB)
  • And
  • (b) Central ischemic chest pain (described as retrosternal pressure, pain, discomfort, or heaviness radiating to neck, jaw, left arm, or shoulder precipitated by exertion more than 20 min)
  • And/or
  • (c) Typical ischemic ECG changes:

    • •
      ST elevation in at least two contiguous leads, ≥0.2 mV in leads V1–V3 or ≥0.1 mV in all other leads.
    • •
      Established MI (in the absence of confounders) is indicated by any Q wave in leads V1–V3 or by Q waves of ≥1 mm for ≥30 ms in two other contiguous leads.
    • •
      Presumed new left bundle branch block.
• (2)Non-ST elevation myocardial infarction (NSTMI): Features as described for STEMI, but not meeting electrocardiographic ST-T criteria.
• (3)
  Unstable angina:

  • At least one of the following:
  • (a) Chest discomfort occurring at rest or at minimal exertion and lasting for >10 min;
  • (b) Is new onset and severe (within last 6 weeks);
  • (c) Occurs with crescendo pattern.

3.6. Variables studied

Cardiac troponin I (cTnI), hs-CRP, eGFR using four variable modification of diet in renal disease (MDRD) equation, and serum CK MB levels in patients with ACS during index event, at 6 weeks, and 6 months.

3.7. Methods

All the consecutive patients with the diagnosis of ACS and who gave informed consent for the study were enrolled and followed up at 6 weeks and 6 months duration from the index event. Mortality and the likely cause of death were recorded along with the day since admission. The hs-CRP was estimated on admission, at 6 weeks, and at 6 months. The eGFR was calculated using the abbreviated MDRD formula at admission, at 6 weeks, and 6 months. For estimating cTnI, qualitative membrane-based immunoassay was used. For estimating CK, the serum sample was analyzed using a semi-automatic analyzer using standard commercially available kits on admission. The patient socioeconomic status was assessed by modified family income groups in Nepalese Rupees of the Kuppuswamy's socioeconomic status scale modified for 2009.⁸

3.8. Statistical analysis

The sample size was calculated as per the WHO's Software for calculating sample size depending on the prevalence of ACS as per the previous epidemiological study in the same setting.⁹

Data were entered in Microsoft Excel 2000 and converted into SPSS version 10 for statistical analysis. Median was calculated for the demographic presentation of the sample. In the bivariate analysis for the numerical data in both the groups (improved or death at 6 months), mean comparison was done and p value was calculated using the non-parametric test (Mann–Whitney U test). Percentage, graphical presentation, and other descriptive statistics were calculated. Multiple variables, which were significantly affecting the outcome, were together taken into multivariate logistic regression analysis. The significance level of the data, which were taken for the multivariate analysis, was ≤0.05 at the bivariate analysis.

3.9. Ethical clearance

Ethical clearance was taken as per the guidelines of Institutional ethical review board (IERB), BPKIHS, Dharan.

4. Results

There were a total of 114 cases of ACS in duration of 6 months. As 6 patients did not complete the follow-up, they were excluded from the study. The total numbers included in the final analysis were 108, out of which 44 were STEMI, 41 were NSTEMI, and 23 were unstable angina (UA) (Fig. 1 and Table 1).

Fig. 1.

Patients with ACS enrolled in the study.

Table 1.

Socio-demographic characteristics of patients with acute coronary syndrome.

Characteristics	Categories	No. of patients	Percentage	p-Value
Age groups (in years)	<55	27	25.00	0.01
55–75	56	51.90
>75	25	23.10
Median age in years (IQR)		65.00 (54.50–73.75)
Sex	Male	58	53.70	0.53
Female	50	46.30
Education	Illiterate	40	37.00	0.049
Literate	41	38.00
Formal education	27	25.00
Kuppuswamy's scale	Upper	3	2.80	0.968
Middle	34	31.70
Upper lower	66	61.10
Lower	5	4.60
Body mass index (kg/m2)	<18.5	9	8.30	0.144
18.5–23	35	32.40
23–27.5	45	41.70
>27.5	19	17.60
Median body mass index in kg/m2 (IQR)		24.12 (21.25–26.87)
Abdominal circumference (males) in cm	<90	20	34.44	0.971
≥90	38	65.56
Median abdominal circumference of males in cm (IQR)		88.00 (93.50–98.00)
Waist hip ratio males	<0.9	4	6.00	0.817
≥0.9	54	94.00
Median waist hip ratio of males (IQR)		0.97 (0.93–1.02)
Abdominal circumference (females) in cm	<80	5	10.00	0.748
≥80	45	90.00
Median abdominal circumference of females in cm (IQR)		88.00 (84.00–92.00)
Waist hip ratio female	<0.8	1	2.00	0.926
≥0.8	49	98.00
Median waist hip ratio of females (IQR)		0.94 (0.91–0.98)

Based on the age distribution with the median age of 65 years, majority of the patients (56/108, 52%) were 55–75 years of age while almost equal proportions were less than 75 years of age. The sex distribution showed a nearly equal male 58 (54%) and female 50 (46%) ratio.

The patients’ socioeconomic status was calculated using the Kuppuswamy's modification for Nepal.⁸ There were 66 (61%) patients in the upper lower strata whereas 34 (32%) belonged to middle group, 5 (5%) were in lower group, and 3 (3%) were in the upper group. The median body mass index (BMI) was 24.12 kg/m². The waist to hip ratio for the males was 0.97 and that of the females was 0.94.

Out of the presenting complaints, the most common was chest pain 71 (66%) and then shortness of breath 60 (56%), epigastric discomfort 38 (35%) followed by palpitations 32 (30%), whereas 22 (20%) patients had syncope. Other symptoms include cough (1%), fever (6%), headache (1%), altered sensorium (1%), loose stools (3%), nausea/vomiting (10%), and diaphoresis (30%).

It was observed that 74 (69%) patients had hypertension and 32 (30%) patients had a history of diabetes, 24 (22%) had CAD in past, and 11 (10%) patients had renal disease previously. Fifty-seven (53%) patients were smokers, and among the smokers, 50 (46%) had more than 10 pack years of smoking. Regarding physical activity, 52 (48%) had a sedentary lifestyle.

On examination, the median pulse rate was 84/min (68–92). The median systolic blood pressure was 140 mmHg and the diastolic being 90 mmHg (Table 2).

Table 2.

Laboratory findings of patients with acute coronary syndrome on admission.

Investigations	Categories	No. of patients	Percentage	p-Value
Total leukocyte count/mm3	<4000	1	0.90	0.088
4000–11,000	45	41.70
>11,000	62	57.40
Median TLC (IQR)		12,300.00 (9050.00–16,425.00)
Differential count neutrophil %	<50	4	3.70	0.244
50–70	41	38.00
>70	63	58.30
Median DLC neutrophil count (IQR)		73.50 (64.25–82.00)
Differential count lymphocyte %	<20	34	31.48	0.303
20–40	60	55.55
>40	14	12.97
Median DLC lymphocyte count (IQR)		26.00 (17.00–34.75)
Hemoglobin (gm/dl)	<12	47	43.51	0.146
12–13	9	8.33
>13	52	48.16
Median hemoglobin (IQR)		12.45 (10.65–14.770)
Urea (mg/dl)	<10	0	0.00	0.007
10–40	74	68.50
>40	34	31.50
Median urea (IQR)		29.00 (23.00–51.50)
Sodium (mmol/l)	<135	9	8.30	0.028
135–145	76	70.40
>145	23	21.30
Median serum sodium (IQR)		141.00 (138.00–145.00)
Triglyceride (mg/dl)	<80	11	10.19	0.160
80–200	83	76.88
>200	14	12.93
Median triglyceride (IQR)		132.50 (98.00–177.00)
Total cholesterol (mg/dl)	<200	96	88.88	0.810
200–239	11	10.19
>240	1	0.93
Median total cholesterol (IQR)		138.50 (107.50–178.00)
HDL cholesterol (mg/dl)	<40	66	61.11	0.994
40–60	42	38.89
>60	0	0.00
Median HDL cholesterol (IQR)		38.00 (37.00–40.00)
LDL cholesterol (mg/dl)	<70	59	54.63	0.850
70–<100	19	17.60
100–129	19	17.60
130–159	9	8.34
>160	2	1.83
Median LDL cholesterol (IQR)		67.20 (45.05–110.90)
Blood sugar level random (mg/dl)	<60	5	4.62	0.058
60–200	22	20.38
>200	81	75.00
Median blood sugar level (IQR)		119.00 (90.00–178.00)

The median leukocyte count was 12,300 cells/mm³. Other baseline investigations were in the normal range. The lipid profile analysis of the patients showed a normal range. The median triglyceride was 132.50 mg/dl (98–177), cholesterol was 138.50 mg/dl (107.50–178.00), HDL was 38.00 mg/dl (37–40), and the median calculated LDL was 67.20 mg/dl (45.05–110.90). It was seen that 5 (5%) of the patients had hypoglycemia on presentation, whereas 81 (75%) had a random blood sugar level above 200 mg/dl. These patients were then subjected to blood sugar fasting and/or HbA1c estimation to confirm the diagnosis of diabetes mellitus if it was not established (Table 3).

Table 3.

Cardiac markers of patients with acute coronary syndrome on admission.

Cardiac markers	Categories	No. of patients	Percentage	p-Value
CK NAC (mg/dl)	<90	22	20.30	0.950
≥90	86	79.70
Median CK NAC (IQR)		301.50 (106.00–811.50)
CK MB (mg/dl)	<25	32	29.70	0.612
>25	76	70.30
Median CK MB (IQR)		40.00 (20.00–90.25)
Cardiac troponin I	Negative	64	59.30	0.593
Positive	44	40.70
eGFR at admission (ml/min/1.73 m2)	<30	34	31.50	0.012
30–59	53	49.10
>60	21	19.40
Median eGFR at admission (IQR)		45.59 (26.11–57.28)
Highly sensitive CRP at admission (mg/dl)	<2	0	0.00	0.001
2–5	21	19.40
>5	87	80.60
Median hs-CRP at admission (IQR)		24.50 (8.25–74.75)
Urine albumin	Nil	61	56.50	0.002
Trace/1+	32	29.60
2+/3+	15	13.90

CK, creatine kinase; hs-CRP, high sensitivity C-reactive protein; eGFR, estimated glomerular filtration rate.

The CK NAC showed a median of 301.50 mg/dl (106–811). Among the group, it was seen that 86 (80%) patients had a raised CK NAC levels. The median CK MB level was 40 mg/dl (20–90.25) and 76 (70%) of the patients had raised CK MB levels. The qualitative troponin I estimation showed positive results only in 44 (41%) of the patients. The median eGFR on admission was 45.59 ml/min/1.73 m². The hs-CRP levels on admission showed a median value of 24.50 mg/dl (8.25–74.75). The urine routine examination done using multistix on admission showed albuminuria of range trace/1+ (30 mg/dl) in 32 (30%) patients and 2+/3+ (100/300 mg/dl) in 15 (14%) patients (Table 4).

Table 4.

Laboratory findings of patients with acute coronary syndrome at 6 weeks and 6 months.

Investigations	Categories	6 weeks		p-Value	6 months		p-Value
		No. of patients	Percentage		No. of patients	Percentage
eGFR (ml/min/1.73 m2)	<30	11	12.80	0.298	7	8.97	0.384
30–59	44	51.16	34	43.60
>60	31	36.04	37	47.43
Median eGFR (IQR)		55.27 (40.11–66.18)		59.02 (45.35–68.58)
Highly sensitive CRP (mg/dl)	<2	0	0.00	0.642	0	0.00
2–5	35	32.40	41	38.00
>5	73	67.60	67	62.00
Median hs-CRP (IQR)		8.0 (4.00–12.50)		5.00 (3.00–11.00)

hs-CRP, high sensitivity C-reactive protein; eGFR, estimated glomerular filtration rate.

On follow-up at the 6 weeks and 6 months, the median eGFRs were 55.27 ml/min/1.73 m² (40.11–66.18) and 59.02 ml/min/1.73 m² (45.35–68.58) and the median hs-CRP values were 8 mg/dl (4–12.50) and 5 mg/dl (3–11), respectively. Among the patients with ACS, 23 (21%) were diagnosed to have UA, 41 (38%) had NSTEMI and 44 (41%) had STEMI.

The most dominant wall involvement was anterior wall in 68 (63%) patients, while 23 (21%) had inferior wall. Whereas 15 (14%) had lateral wall and 2 (2%) had septal wall involvement. In the patients with ACS, 56 (52%) had a Killip class of I.

This study showed that 10.28 (26%) of the patients had arrhythmias. The most common arrhythmia was atrial fibrillation followed by ventricular arrhythmias and then the bundle branch blocks whereas 24 (22%) of the total patients had cardiogenic shock during the index event, 2 (2%) patients had ischemic events, and 9 (8%) had uncontrolled hypertension. The other complications included coagulopathy (1%), hemoptysis (1%), and intracranial hematoma (1%).

The mean duration of hospital stay was 6 days. There were 12 patients who died during the hospital stay whose cause of death is known and for 2 patients the cause of death is unknown, and 16 died during the follow-up period. The in-hospital mortality rate was 11% whereas overall mortality at 6 months follow-up was 28% including all the patients from the index event until final outcome at 6 months.

Cardiogenic shock was seen in six (6%), sudden cardiac death in four (4%), and ventricular tachycardia in two (2%) of the patients who died during the index event.

The bivariate analysis was performed and the data were compared in two groups. G1 represented improved outcome whereas G2 included mortality at the end of 6 months follow-up.

The distribution of age was comparable in both the groups. Age >75 years showed statistical significance with all-cause mortality (p = 0.011). There was no significant association with mortality in sex, occupation of the patient, and Kuppuswamy's scale for socioeconomic status or BMI. Those who were not formally educated had increased all-cause death (p = 0.049). The abdominal circumference and waist to hip ratio were not statistically significant.

The systolic blood pressure <90 mmHg was associated with increased chances of mortality but was not statistically significant (p = 0.058).

There were no significant differences in the mean total leukocyte count and hemoglobin level in between mortality and improved group. The mean serum urea mg/dl was 36.81 ± 29.36 in the improved group but it was significantly higher (50.03 ± 30.92) in the death group (p = 0.007). Similarly, the serum sodium level >145 mmol/l was significant for death (p = 0.028). Although blood sugar levels >200 mg/dl and <60 mg/dl were associated with overall mortality, it was statistically not significant (p = 0.058). The mean eGFR in the mortality group was 36.68 ± 19.74 mg/dl in the death group (p = 0.012). The hs-CRP level at admission was highly significant (p = 0.001) with a mean value of 54.37 mg/dl in the mortality. Similarly, albuminuria of 30 mg/dl and above was significant for mortality (p = 0.002).

The type of ACS was not statistically significant. The Killip class IV was associated with increased mortality (p = 0.007).

In-hospital mortality during the index event was significant (p = <0.001). The mean duration of stay in the death group was 6.89 ± 7.53 days. In the causes of mortality, ventricular tachycardia was the most significant (p = <0.001) followed by sudden cardiac death and cardiogenic shock (Table 5, Table 6).

Table 5.

Outcome of patients with acute coronary syndrome at 6 weeks.

Outcome	Categories	No. of patients	Percentage	p-Value
Death at 6 weeks	Yes	9	8.30	<0.001
No	85	78.70
Acute coronary syndrome at 6 weeks	Yes	3	2.80	0.230
No	82	75.90
New York Heart Association at 6 weeks in class	I	31	28.70	0.097
II	42	38.90
III	12	11.10
IV	0	0.00
PCI/CABG at 6 weeks	Yes	7	6.50	0.508
No	79	73.10

Table 6.

Outcome of patients with acute coronary syndrome at 6 months.

Outcome	Categories	No. of patients	Percentage	p-Value
Death at 6 months	Yes	7	6.50	<0.001
No	78	72.20
New York Heart Association at 6 months in class	I	40	37.00	0.740
II	34	31.50
III	4	3.70
IV	0	0.00
PCI/CABG at 6 months	Yes	4	3.70	0.780
No	74	68.50
Final status of patients at 6 months	Improved	78	72.20	<0.001
Death	30	27.80

The primary outcome was highly significant at 6 weeks and 6 months period (Table 7).

Table 7.

Outcome in patients with acute coronary syndrome by mortality or no mortality.

Outcome	Categories	Improved	Mortality	p-Value
Death at 6 weeks	Yes	0	9	<0.001
No	78	7
Death at 6 months	Yes	0	7	<0.001
No	78	0
Outcome final	Improved	78	0	<0.001
Death	0	30

For the multivariate analysis, the variables at a significance level of p = ≤0.05 in the bivariate analysis were taken. The multivariate analysis showed that age, hs-CRP, albuminuria, and cardiogenic shock were strongest predictors of mortality when used in relation to each other. In the study, the cause of death was not known in two patients who had died during the index event. Autopsy was not done in both the cases, as the facility was available in our setup. The samples of the patients were collected in the emergency, but as the cause of death was not known, they were not included in the in-hospital mortality data analysis. As their discharge status was considered death, they were included in the final analysis. The final outcome of the study was all-cause mortality.

5. Discussion

In this series, 114 consecutive patients were studied. The age of the patients ranged from age 27 to 92 years with the median age of 65 years, which is similar with the studies done in the same setting.9, 10 The mean age at diagnosis was 64 years in a study done at western part of Nepal.11, 12 Thus, it was seen that the prevalence is higher in older age group and with higher mortality rate. It was statistically significant with age >75 years. Age is a powerful predictor of adverse events after ACS. In a study on ACS in Nepal, it is stated that death claims mostly the elderly population; in US, 83% of IHD deaths were in patients of more than 65 years of age as per the GRACE registry.12, 13, 14 In the study, 27 patients were less than 55 years of age, which is also similar to the study done previously in the same setting.¹⁰ There is an increasing trend of young people having ACS. The male to female ratio is 1.16:1 whereas in previous study, it was 1.6:1, but in other studies from Nepal, it was 1.14:1,⁹ which varied with sample size and study settings.

The median leukocyte count was 12,300 cells/mm³. The leukocytosis was in response to the inflammatory nature of ACS. In a study, it was seen that there was an epidemiological association indicating a worse angiographic appearance of the culprit lesion with higher white blood count at presentation with ACS.¹⁵ The level of blood urea above 40 mg/dl was significantly associated with all-cause mortality during the index event and also during the 6 months period. There was a statistical significance with low eGFR (median eGFR 45.59 ml/min/1.73 m²) levels during the admission. It was also associated with high serum sodium levels. It is arguable on the basis of these observations that the patients are mainly dehydrated (high urea, low eGFR, and high sodium levels) in the mortality group. These all were statistically significant at the final outcome. The most important was the creatinine levels at admission. The eGFR in VALIANT study, where the four-component MDRD equation was used, found that the distribution of estimated GFR was wide and normally shaped, with a mean (±SD) value of 70 ± 21 ml/min/1.73 m² of body-surface area, the probable reason of this difference being primarily that we used median in our analysis and secondly ethnicity being one of the variables.¹⁶ A more recent analysis from the GUSTO (Global Utilization of Strategies To Open occluded arteries) IV study found that creatinine clearance and troponin elevation provided the greatest relative contribution to risk (1-year mortality or 30-day mortality/MI) beyond traditional risk factors.¹⁷

There have been different studies on the risk stratification of CK MB, but in one study done by Alexander et al., it showed a statistical difference in 30-day mortality even if there was increase in the level of CK MB levels 1–2 times the upper normal. In this study, the median CK MB level was 40 mg/dl, but there was no statistical significance for risk stratification. The reasons may be due to smaller sample size when compared with the other study where it is very large.¹⁸ Secondly, the patients presented with a median duration of 2 days after the chest pain and as the levels of CK MB come to normal level within 72 h, this delayed presentation gives us a different picture. It was seen that there is a low triglyceride levels (<200 mg/dl) in the Caucasian patients with ACS if taken within 24 h of admission. The study showed low triglyceride level to be an independent predictor of mortality at 3 years from the index event. In this study, the mean triglyceride levels was 161.53 ± 76.55 mg/dl, which is in lower range with the given reference limit. This is also similar to the studies done on ACS in the same setting.¹⁹

In this study, the hs-CRP levels of >5 mg/dl were highly significant for predicting mortality during hospital stay and at 6 weeks. In the GRACE registry, 12% of patients with STEMI, 13% with NSTEMI, and 8% with UA were expected to die in 6 months within onset of symptoms.²⁰ There have been a number of studies done which found the prognostic significance of hs-CRP. Most of the studies were done in stable CAD. The JUPITER trial, which was restricted to participants with CRP levels >2 mg/l, found that treating the patients with rosuvastatin decreased the hs-CRP and LDL levels and the cardiovascular outcomes.²¹

In this study, microalbuminuria was not taken as a variable, as it was studied earlier but urine sample was sent for routine analysis, which was done from multistix/uristix kit on every patient with ACS. It was found that albuminuria was a predictor of mortality. The more the level of albumin on admission, there was increased mortality. Acuna et al.²² found microalbuminuria to be an independent risk factor of heart failure (OR = 1.75; 95% CI = 1.02–3.01; p = 0.04) and of mortality (OR = 2.6; 95% CI = 1.05–6.41).

In a study done in western Nepal, involvement of ACS was most common in anterior wall and inferior wall than any others wall.¹¹ In this study, it was found there was a very similar involvement. This was also similar to the study on ACS at BPKIHS done previously. In a study, it was seen that arrhythmias were in 27%, similar to the findings of this study, where 26% of patients had arrhythmias and it was found statistically significant for predicting mortality. Cardiogenic shock complicating STEMI resulted in high mortality.¹¹ It was observed that there was 11% in-hospital mortality in this study. From the study, if the total in-hospital mortality irrespective of the cause of death was taken, then it was found to be 13% (2 cases were excluded as the cause of death was not known). The high in-hospital mortality (14%) was observed in other study.¹⁰ This can be ascribed to markedly delayed presentation, lack of resources for the best possible available care, and major logistic difficulties to organize timely PCI. During the study period, there were no facilities for PCI at this setting. Similarly, in other study from Nepal, the in-hospital mortality was 12%.¹¹ Outcomes at 6 weeks period showed a significant mortality. The duration of hospital course was almost similar when compared with other contemporary international surveys like GRACE registry. At 6 months, the overall mortality was 28%, which is highly significant. No other study was done in this setting to compare results at 6 months duration.

We also searched to look for other studies in Nepal which did a follow-up of the patients. We found a registry of patients in western Nepal, but data for 6 months period were lacking in that study.

Logistic regression analysis of the variables studied was done. Only those variables, which were significant at p ≤ 0.05 in the bivariate analysis, were taken for multivariate analysis to increase the statistical significance. It was found that age >55 years; albuminuria and cardiogenic shock at the index event remained as a strong predictor of all-cause mortality in patients with ACS in relation to each other. Similarly, hs-CRP levels >5 mg/dl and CK levels at the index event are also a strong predictor of mortality.

6. Summary and conclusion

The levels of serum hs-CRP, serum blood urea, serum creatinine, and albuminuria (done using a simple uristix/multistix test) at admission are statistically significant independent predictors of death in patients of ACS at the index event at 6 weeks and at 6 months.

When the significant biomarkers and the significant established risk factors were studied in relation to each other, hs-CRP levels were significant along with age in predicting outcomes. Fluid balance is a very important aspect and we need to do a bigger study with larger sample size to generalize this observation.

7. Limitations

In our setting, we did not have facilities for estimation of quantitative levels of troponin and novel biomarkers like NTproBNP, which are more specific and better tools for predicting outcomes in patients with ACS. During the study, there were no facilities for cardiac catheterization at our setting.

Contributors

PK performed substantial contribution to conception and design, revised the content of the draft critically for important intellectual content, and approved the final version to be published. KKA, LM and NRS performed substantial contribution to conception and design, drafted the article, and approved the final version to be published.

Patient consent

Obtained.

Ethics approval

Institutional Ethical Review Board (ERB) of B.P. Koirala Institute of Health Sciences, Dharan, Nepal.

Conflicts of interest

The authors have none to declare.