Discussion
The main, and the most important, finding is that, among those patients reaching PCI hospitals included in this study and surviving 24 hours, the vast majority were investigated and treated in line with current ESC guidelines. Previously, we have documented the dramatic expansion in health facilities equipped for PCI across Russia.7 Our findings suggest that across a range of these mainly new PCI hospitals the needs of patients with STEMI that reach them on time, are largely met.
Notwithstanding the need for caution given the purposive nature of the study design, within those meeting our inclusion criteria, the majority of patients with STEMI in Russia present in the early phase and those who reach PCI hospitals at this stage and have angiography have a very high probability of PCI. Indeed, while a number of studies in countries other than Russia have found that compared with men, women with CVD receive suboptimal treatment and have higher risks of adverse outcomes, our study finds, among patients with early phase STEMI who can be expected to benefit most, no significant difference in PCI use by sex after adjusting for comorbidity, age and hospital type. This is in line with more recent research suggesting that the well-known sex disparities could be explained by differences in case mix.14 15 Moreover, consistent with ESC guidelines, patients who did not receive PCI after angiography tended to have multivessel disease or other contraindications, while PCI use did not vary according to the admission route taken by the patient or time of admission, suggesting that hospitals really do offer a 24-hour service. We did, however, find a sex difference in PCI use among patients with evolved STEMI.
Adherence to ESC recommendations did vary; we identified differences in angiography and PCI use across different levels of hospital. Although angiography rates are high overall in patients with STEMI, they are significantly lower in City hospitals in our sample. Moreover, patients with STEMI in City hospitals who have an angiogram are less likely to have PCI (table 3). Additionally, we identified one hospital, Rostov, as an outlier in our sample: nearly half of all patients with early phase STEMI do not receive an angiogram there. While official figures indicate that a very low proportion of patients arrive there within 24 hours, the MAMIR data highlight that even among patients who do arrive early, angiogram use is low. Pre-hospital delay alone does not explain the low use of angiography, and thus subsequent PCI, in these patients with STEMI. Further investigation revealed that even though this hospital does provide PCI, it is not the main hospital for PCI in the region. Thus, while it seems most patients with AMI are taken to another hospital if time and distance permits, improvements in treatment for those patients left behind are possible.
A key finding from the study is that a number of hospitals in our study performed particularly well in specific areas of AMI management that were in line with ESC guidelines. Indeed, four hospitals offered angiography to all patients with early phase STEMI. These hospitals could potentially serve as a beacon to others, offering scope for shared learning. Additionally, AMI management of patients with early phase STEMI is particularly good, while there remains room for improvement for evolved and recent STEMI, in particular, with sex differences in PCI use existing among patients with evolved STEMI. Inter-hospital knowledge transfer will be particularly important as PCI health facilities continue to expand across the Russian Federation, in addition to strategies to improve pre-hospital delays.
The challenge now is to ensure that those facilities that are lagging behind are brought up to the standards observed in the best. Since 2010, the Federal Ministry of Health has adopted a range of measures designed to improve quality of care. These include both additional resources for healthcare and various targeted new initiatives such as revised clinical practice guidelines, drawing on international standards, pay for performance schemes, electronic medical records and various quality control programmes. These have been described in a recent review, which documented a high level of commitment to better quality care but also very limited evaluations, reflecting the lack of capacity for health services research in Russia.16 However, a further challenge is the persisting siloed and hierarchical management structure in many Russian hospitals, something that is deeply embedded in the system and will be difficult to overcome.17
This study also provides some evidence on pathways tohospital and TLT use for patients with STEMI in these hospitals. Just over a quarter of patients arrived at hospital 12 hours after symptom onset. Pre-hospital delay is a complicated issue, which involves both health seeking behaviour (patient delays) and health system factors (transportdelays). Patient delays (prolonged delays fromsymptom onset to seeking help) can reflect many factors, from symptom recognition to proximity and access to care. A proper understanding will require detailed research on health seeking behaviour using qualitative and quantitative approaches.
A key finding of this study is that there is a largeproportion of patients with STEMIwhohave long travel times to PCI hospitalsbut do not receive TLT. The STREAM trial examined the effectiveness of TLT inthose unable to receive immediate PCI. Subjects were randomised to either delayed (median 3 hours) PCI or rapid administration of TLT with rapid PCI where that failed or otherwise at 6-24 hours. Those receiving TLT achieved as good cardiac outcomes as those undergoing primary PCI, although they experienced a slightly high rate of bleeding.18 Thus, in Russia, when patients are unable to obtain immediate PCIs butcan be transferred to a hospital where it is available, the administration of TLT is a tractable measure that could improve prognosis for patients with AMI.
In a separate study, we have mapped road travel times from all Russian districts to the nearest facility providing PCI in 2015 (both in the same region and in neighbouring ones).8 This identified two strategies that could reduce travel times and thus, hopefully, delays. The first was the creation of 67 new PCI centres, in addition to the 260 that were then operating. This would increase the share of the population within 60 min travel time by almost nine percentage points, benefiting 5.7 million people. The second was to permit people living near regional borders to attend the closest facility, wherever it was. However, this would increase the number of people within 60 min travel time by only 340 000.
Ultimately, in Russia as elsewhere, there is likely to be scope for integrated planning of services, spanning the entire patient journey. This has proven successful with the acute management of stroke, as in London and, in varying forms, some other places, and a similar approach is being evaluated in seven tertiary hospitals in Germany.19–22 However, the organisational challenges are considerable as this model has been difficult to implement elsewhere.
The results we have presented should be interpreted with caution as the MAMIR Study has several important limitations. These, to some extent, reflect the challenge of conducting health services research outside major hospitals. Our sample is intentionally diverse, covering multiple levels of the health system to describe contemporary management of patients with AMI in Russia. The hospitals included cannot, however, be taken as representative of the overall situation in Russia, as we were constrained in selecting only those where it was possible to gain commitment by a local co-investigator. Within this constraint, we did sample purposively to cover the geography of Russia and facilities at all levels within the health system. However, as we have shown previously, there is considerable variation in the extent to which regions have adopted advanced management of AMI7 and, taking the PCI rate per 100 000 in 2013 as an indicator of progress, we also span the entire range of those providing a service at that time. Thus, we included two regions in the top quintile of activity, and 4, 2, 3 and 2 in each quintile of declining activity.
Despite strenuous efforts, there was a relatively high level of missing data, including lack of information on pre-hospital delays in 96 patients with STEMI and we were unable to obtain detailed data that would have allowed us to understand overall patient flows within the hospital system and, specifically, how many patients arrived at a non-PCI hospital and were transferred. Nonetheless, the data from this study provide a much improved evidence base from which to assess the current management of patients with AMI in the Russian Federation.