Pre-hospital delay in Vietnamese patients hospitalized with a first acute myocardial infarction: A short report

Background: Administration of coronary reperfusion therapy to patients with an acute myocardial infarction (AMI) within the proper timeframe is essential in avoiding clinical complications and death. However, the extent of pre-hospital delay is unexplored in Vietnam. This report aims to describe the duration of pre-hospital delay of Hanoi residents hospitalized with a first AMI at the Vietnam National Heart Institute . Methods: A total of 103 Hanoi residents hospitalized at the largest tertiary care medical center in the city for first AMI, who have information on prehospital delay was included in this report. Results: One third of the study sample was women and mean age was 66 years. The mean and median pre-hospital delay duration were 14.9 hours and 4.8 hours, respectively. The proportion of patients who delayed <6 , 6-<12, and ≥ 12 hours were 45%, 13%, and 42%, respectively. Conclusions: Our data shows that a prolonged pre-hospital delay is often observed in patients with a first AMI in Vietnam. In order to confirm these preliminary descriptive findings, a full-scale investigation of all Hanoi residents hospitalized with first AMI is needed. Increasing public awareness about AMI treatment is vital in encouraging patients to seek medical care timely after experiencing AMI symptoms such that received treatment is most effective.


Introduction
Vietnam is a low-middle income country, and has been in the midst of an important epidemiological transition. Over the past two decades, the overall morbidity and mortality rates resulting from non-communicable diseases (NCDs) has been rising rapidly and presents itself as a major problem in Vietnam. One quarter of all deaths annually in Vietnam are now the result of cardiovascular diseases, making it the leading cause of death.
Timely administration of reperfusion therapy to patients with an evolving acute myocardial infarction (AMI) is of utmost importance in reducing clinical complications and death. Previous research studies have convincingly proved that reperfusion treatment is most effective if patients with ST-segment elevation myocardial infarction (STEMI) are treated in a timely fashion, particularly within one hour of acute symptom onset 1,2 ; the relation between extent of pre-hospital delay and outcomes after non-STEMI has not been firmly established.
Despite the importance of prolonged pre-hospital delay on the timely receipt of effective treatments and short-term outcomes, so far we have found no evidence of studies conducted to examine the extent of pre-hospital delay among adult patients hospitalized with AMI in Vietnam. This short report hence aims to describe the extent of pre-hospital delay in Hanoi residents who were non-transferred, hospitalized with a first AMI at the Vietnam National Heart Institute in 2010.

Ethical statement
This study was approved by the Institutional Review Board at the Institute of Population, Health and Development. A waiver of patient consent was approved by the Institutional Review Board. All patients' personal information was de-identified before analysis.

Study setting
This study was conducted at the Vietnam National Heart Institute (VNHI) 3,4 . This hospital is a 250 bed tertiary care medical center in Hanoi (2009 census = 6.5 million) which manages the majority of Hanoi residents hospitalized with acute coronary disease and other NCDs.

Case ascertainment
Computerized printouts of discharged patients from the VNHI in 2010 with possible AMI were collected and International Classification of Disease codes for possible AMI (I20-I25) were reviewed. Two cardiologists validated these cases utilizing predefined criteria developed by the World Health Organization. They include having a suggestive medical history, serum enzyme increasing above the hospital's reference range, and serial electrocardiographic (ECG) findings during hospitalization consistent with the presence of AMI; at least two of these three criteria needed to be present for an AMI to have occurred 5 . There were 315 Hanoi residents, who satisfied the criteria for AMI were identified. However, 13 (4%) patients with history of AMI and 199 (66%) patients transferred from other facilities were excluded from the present analysis. Patients were then further divided into groups of STEMI and non-STEMI, utilizing standard classification techniques 6 . All ECGs of potential AMI patients were reviewed by a physician under the supervision of a senior cardiologist. Due to reviewed ECGs' inadequate quality, nine patients with AMI could not be classified as STEMI or non-STEMI.

Data collection
For all AMI-validated cases through the aforementioned process of independent adjudication utilizing eligibility criteria, information on social demographics, clinical, medical management and hospital discharge status were collected from medical record reviews and recorded onto a standardized case report form by trained study physicians.
Pre-hospital delay is the primary outcome investigated by this study. The delay time was defined as the time interval between the onset of symptoms suggestive of AMI and the patient's arrival at medical centers 7,8 . Our trained physicians reviewed any information they could seek in medical records, which described the duration of pre-hospital delay from emergency personnel, nurses, and physicians notes. Information on prehospital delay was collected in minutes (as a continuous variable) then was further categorized according to cut-points of 6 and 12 hours, which were commonly applied in prior studies, and based on the data distribution Patients, in whom no exact time of symptom onset was reported, and only had a delay time interval recorded in their medical record (e.g. <6, 6-<12, and ≥12 hours), were incorporated into our analysis of delay time when it was

Amendments from Version 2
In the version 3 of this paper, we have attempted to address each of the concerns raised by the third reviewer about our submission as follows: We have included several additional limitations in the discussion section. First, since our data were collected from only one hospital in Hanoi, some of observed patient characteristics may not reflect the general population of patients hospitalized with AMI in Hanoi; the distribution of Non-STEMI appeared to be low in this study sample, though the frequency of various manifestations of AMI in Vietnamese adults is unknown. Second, the subgroup analysis according to type of AMI, namely by STEMI and Non-STEMI, was subject to potential bias in interpretation due to the small study sample size. Third, due to the design of this retrospective study, we were unable to obtain systematically collected and/or recorded information about the time of onset of symptoms suggestive of AMI in hospital medical records. This information was abstracted from notes written by different physicians, nurses, and emergency personnel which may have been collected and recorded in a nonstandardized manner.

REVISED
constructed as a categorical outcome. We restricted our patient population to those with information available on prehospital delay (either exact time noted or time interval) in their hospital medical records.

Data analysis
The overall mean (standard deviation-SD) and median (inter quartile range-IQR) duration of pre-hospital delay was calculated according to standard methods. Data were shown as percentages for categorical variables and compared between patients who delayed ≥ 6 hours, and patients who sought medical care earlier, using chisquare or Fisher exact tests; medians (inter quartile range-IQR) for continuous variables were calculated and compared using the Wilcoxon sum rank tests. Since the sample size in this study was small, no regression analysis was performed. All analyses were performed using STATA 11.0 (StataCorp. TX).

Results
A total of 103 non-transferred Hanoi residents hospitalized with a first AMI at VNHI in 2010, and had information available on prehospital delay recorded in their hospital charts were included in the report. The study sample had an average age of 66 years with a standard deviation of 13 years. One third of the study sample was women, and 69% were classified as STEMI.
Extent of pre-hospital delay Information on pre-hospital delay was collected in minutes (as a continuous variable) for 63% (n=65) of patients. In 37% of patients (n=38), an exact time of symptom onset was not reported, and only a delay time interval was recorded in patient's medical record (e.g. <6, 6-<12, and ≥12 hours); these patients were included in the analysis of delay time when it was constructed as a categorical outcome.
Analysis of prehospital delay as a continuous outcome. Within the group of 65 patients reporting an exact time of symptom onset, the overall mean and median durations of pre-hospital delay were 14.9 hours (SD: 23 hours) and 4.8 hours (IQR: 3-10 hours), respectively ( Figure 1). Among patients with STEMI (n=42), mean and median delay times were 16.4 hours (SD: 26 hours) and 4.8 hours (IQR: 3.0-6.2 hours), respectively. Among patients with non-STEMI (n=19), mean and median delay times were 14.0 hours (SD: 19.2 hours) and 8.3 hours (IQR: 2.5 -12.3 hours), respectively.

Figure 1. Mean and Median of Pre-hospital Delay in Patients
Hospitalized with an initial Acute Myocardial Infarction.

Figure 2. Distribution of Duration of Pre-hospital Delay in Patients
Hospitalized with an initial Acute Myocardial Infarction.
Patient characteristics associated with prolonged pre-Hospital delay Patients who had a delay time over 6 hours were more likely to be female compared to their counterparts, who had a delay time of less than 6 hours (Table 1) For all AMI-validated cases, information on social demographics, clinical, medical management and hospital discharge status were collected from medical record reviews and recorded onto a standardized case report form by trained study physicians. GFR: Glomerular filtration rate; STEMI: ST elevated myocardial infarction; LDL: Low density lipoprotein.

Discussion
The results of this study indicated that Vietnamese patients experienced a considerable pre-hospital delay after the onset of AMI symptoms (median: 4.8 hours), with more than half experiencing a delay time of 6 hours or longer. Women delayed seeking care to a greater extent than men.
This report is, to our knowledge, the first in Vietnam to describe the duration of pre-hospital delays in patients hospitalized with a first AMI. In this study, patients experienced considerably longer pre-hospital delay compared with patients from both low-middle income countries and high-income countries 7-16 .  18 . We found that comparatively, female patients tended to seek treatment much later than male patients after the onset of suggestive AMI symptoms. This result is consistent with studies conducted previously 9,10,19-22 , but contradictory to others, which reported that there was no association between sex and pre-hospital delay 12-14 .
Our findings emphasize the need to implement intervention programs to increase awareness of the general population with regards to the importance of seeking treatment timely after experiencing AMI suggestive symptoms. Prolonged delay may be associated with an individual's risk of cardiac death, serious clinical complications, and delays in the receipt of effective treatments, primarily coronary reperfusion therapy. Moreover, to understand why these patients fail to react promptly to their AMI symptoms, focus group discussions and/or in-depth interviews should be carried out in patients hospitalized with AMI focusing on their levels of cognition, knowledge, and attitudes toward hospitals and health care. Additional studies are warranted in exploring the effects of educational attainment, insurance coverage, neighborhood characteristics, psychosocial aspects, and other factors that may serve as either facilitators or obstacles for patients to the timelier seeking of medical care, particularly of factors that may be amenable to modify.

Study limitations
There are several imitations in this study. First, our data were collected from only one hospital, the VNHI in Hanoi, which limits the generalizability of our results to patients hospitalized in other medical centers across Hanoi. Some of observed patient characteristics may not reflect the general population of patients hospitalized with AMI in Hanoi; the distribution of non-STEMI appeared to be low in this study sample, though the frequency of various manifestations of AMI in Vietnamese adults is unknown. Second, due to our small sample size, we failed to identify potential factors that might relate to the prolonged delay time, information which is important for designing and implementing various educational intervention approaches to improve patient's care seeking behavior. The subgroup analysis according to type of AMI, namely by STEMI and non-STEMI, was subject to potential bias in interpretation due to the small study sample size. Third, we were only able to calculate total duration of pre-hospital delay and we were unable to separately calculate patient delay time and extent of transportation delay since the time when patients made a treatmentseeking decision was not recorded in hospital medical records. In Vietnam, patients with signs and symptoms suggestive of AMI normally call for an ambulance or reach the hospital by themselves, and extremely few patients have a primary care doctor to consult with. In terms of transportation time, this study included only patients who reside in Hanoi. Forth, due to the retrospective design of the study, we were unable to obtain systematically collected and/or recorded information about the time of onset of symptoms suggestive of AMI in hospital medical records. This information was abstracted from notes written by different physicians, nurses, and emergency personnel which may have been collected and recorded in a non-standardized manner. Data on duration of prehospital delay were missing in approximately two thirds of Hanoi residents hospitalized with independently validated AMI, which was mainly due to missing data on the time of onset of acute symptoms suggestive of AMI, and not emergency department arrival time. Finally, we were unable to obtain information on additional patient-associated characteristics (e.g., socioeconomic status [eg., Education, occupation] and psychological factors [eg., reluctance to call family members for help or having no one around when experiencing symptoms to ask for help]) which may have impacted on patient's medical care seeking behavior.
Our data show that patients admitted for initial AMI in Hanoi, Vietnam experienced a significant delay time in seeking medical attention after the onset of AMI suggestive symptoms. Additional full-scale data of hospitalized patients with AMI across Hanoi will be needed in order to confirm these preliminary descriptive findings. Public educational programs for the general population, specifically targeting towards women, is greatly needed to incentivize patients to seek medical assistance immediately to ensure maximum effectiveness of treatments. Further research is required to understand potential factors associated with the prolonged delay in Vietnamese patients hospitalized with AMI and the most effective ways to encourage patients to seek medical care in a timely manner.

Data availability
F1000Research: Dataset 1. Raw dataset for Nguyen et al., 2015 'Pre-hospital delay in Vietnamese patients hospitalized with 1.

2.
3. This report addressed an important gap in the literature about the care-seeking delay of Vietnamese in the context of acute myocardial infarction. The major limitations of the study were small sample size and retrospective nature of the study. Nevertheless, this is a good start to provide quick preliminary findings to support a more formal and large scale study to examine the phenomenon. I have several additional comments:

Open Peer Review
69% of the sample suffered from STEMI, which is not consistent with the proportion of STEMI/NSTEMI in AMI population. This reflects that the sample is not representative and need to address in the discussion and limitation sections.
Retrospective medical record review is not a reliable data collection method for collecting data on pre-hospital delay. The inconsistency for healthcare professionals to determine the symptom onset time creates a significant bias to the results.
The very small number of NSTEMI patients made the subgroup analysis to be less appropriate.

If applicable, is the statistical analysis and its interpretation appropriate? Partly
Are all the source data underlying the results available to ensure full reproducibility? Partly

Are the conclusions drawn adequately supported by the results? Yes
No competing interests were disclosed.

Competing Interests:
I have read this submission. I believe that I have an appropriate level of expertise to confirm that it is of an acceptable scientific standard, however I have significant reservations, as outlined above. We would like to thank the reviewer of our article for the helpful comments. We have attempted to address each of the concerns raised about our submission and have indicated in the revised text where changes have been made. We believe that the manuscript is considerably stronger than before.
1. 69% of the sample suffered from STEMI, which is not consistent with the proportion of STEMI/NSTEMI in AMI population. This reflects that the sample is not representative and need to address in the discussion and limitation sections. 2. Retrospective medical record review is not a reliable data collection method for collecting data on pre-hospital delay. The inconsistency for healthcare professionals to determine the symptom onset time creates a significant bias to the results.

Response
Response: Due to the retrospective nature of this observational study, there was no systematic method established to record the time of onset of symptoms suggestive of AMI (e.g., chest pain) in method established to record the time of onset of symptoms suggestive of AMI (e.g., chest pain) in a standardized manner by trained health professionals. While we would have preferred to have directly interviewed hospitalized patients with AMI in a standardized manner, as noted in response to a similar query by the first and second reviewers, this information was abstracted from notes written by different physicians and nurses and emergency personnel which may have been collected and recorded in a non-standardized manner. We have noted this limitation in the discussion section of the revised manuscript (Study limitations section).
3. The very small number of NSTEMI patients made the subgroup analysis to be less appropriate.
Response: We agree with the reviewer and have acknowledged this limitation in the revised discussion (Study limitations section). However, in this brief report, we would prefer to describe the extent of prehospital delay for patients with a STEMI and NSTEMI separately given the extremely limited data presently available on this topic in Vietnam and due to differences in the natural history of these conditions. In our conclusions, we have emphasized the need for full-scale surveillance of Hanoi residents hospitalized with AMI at all Hanoi hospitals to confirm these preliminary descriptive findings. I have read the Summary of revisions presented by the author of the paper "Pre-hospital delay in Vietnamese patients hospitalized with a first acute myocardial infarction: A short report". Generally, the author has improved the weak points indicated by reviewers by adding information, discussing the limitation issues, or re-structuring within each section of the paper.

None
The main finding of this paper supplies evidence of pre-hospital delay among AMI patients in Vietnam that would probably result in further comprehensive study of this issue in the near future in Vietnam.
I agree with the revisions of the author, and approve this paper.
No competing interests were disclosed.

Competing Interests:
I have read this submission. I believe that I have an appropriate level of expertise to confirm that it is of an acceptable scientific standard. There is a very small population to study and to obtain valid conclusions from.

Version
It is uncertain how the time of chest pain was recorded. It appears there is no system that allows for a correct determination of the precise moment the patient arrives at the ER, due to the fact that in many patients this particular data is missing.
The report is based on more than a hundred patients but in fact the author only presents data from 65 patients.
There is no explanation about the findings nor factors.
Because only a few patients were enrolled, it is not possible to determine factors related to early or delayed arrival at the hospital.
Finally some data, which are in fact results, are written in the methods.
No competing interests were disclosed.

Competing Interests:
I have read this submission. I believe that I have an appropriate level of expertise to state that I do not consider it to be of an acceptable scientific standard, for reasons outlined above.

Competing Interests:
The benefits of publishing with F1000Research: Your article is published within days, with no editorial bias You can publish traditional articles, null/negative results, case reports, data notes and more The peer review process is transparent and collaborative Your article is indexed in PubMed after passing peer review Dedicated customer support at every stage For pre-submission enquiries, contact research@f1000.com