Umeå University's logo

Background

Cardiovascular disease, of which coronary heart disease constitutes the lion’s share, is the leading cause of premature morbidity and mortality worldwide. Management of the condition has evolved rapidly in recent decades, and mortality has more than halved in the western world. Because of intense research, solid evidence supports effective and inexpensive means of preventing disease progression. However, secondary prevention still yields disappointingly low success in meeting guideline-recommended risk factor targets. It is therefore vital to develop more effective risk factor management.

Aims

We aimed to assess the feasibility of a nurse-led, telephone-based, secondary preventive intervention in an unselected population with acute coronary syndrome (ACS). Furthermore, we sought to evaluate the flexibility of the intervention to adapt to a change in guidelines. We also aimed to evaluate whether the intervention was more effective than usual care at improving risk factor levels for blood pressure (BP) and low-density lipoprotein cholesterol (LDL-C) 12 months after discharge. Finally, we aimed to measure whether the intervention improved long-term adherence to statins.

Methods

All papers are based on the Nurse-based Age-independent Intervention to Limit Evolution of Disease after ACS (NAILED-ACS) trial. The NAILED trail has two arms, one after stroke/transient ischemic attack (NAILED-Stroke) and one after ACS (NAILED-ACS). All studies are based on NAILED-ACS aside from study II which includes both arms. The trial was an open, 1:1 randomized, controlled, parallel group trial that compared nurse-led telephone follow-up with medical titration (intervention) to a control group with follow-up by a general practitioner (control). All patients admitted to Östersund Hospital for ACS during 2010–2014 were eligible if available for preventive management by telephone. A baseline assessment was made at 1 month after discharge and thereafter every 12 months for at least 3 years. 

Feasibility was assessed among patients admitted until 31 January 2013, and predictors of exclusion and non-participation were identified. The performance of the intervention in implementing a guideline change was evaluated in patients with diabetes with both ACS and stroke as inclusion events after a change in LDLC target from <2.5 mmol/L to <1.8 mmol/L. LDL-C levels were compared between intervention and control patients before and after the guideline changed. Reasons for not reaching the target level were recorded. The outcomes of the intervention on BP and LDL-C were studied in patients admitted until 31 December 2013. We measured proportions reaching targets and levels of LDL-C and BP during the first 12 months of follow-up, with comparisons between the intervention and control groups. Adherence to statin treatment was measured in the entire study cohort, with at least 36 months of follow-up, with classification of reasons and analysis of predictors for both a first and a permanent discontinuation.

Results

Of 907 screened patients with ACS in the first study, 72.9% were included, and 11% declined participation. Among the 16.1% who were excluded, the predominant reasons were participation in other trial, dementia, and advanced disease. Non-included patients were significantly older, with more comorbidities, decreased functional capability, and lower level of education compared to included. Excluded and declining patients also had a reduced oneyear survival in comparison with included.

Before the guideline changed, 96% of the 101 patients in the intervention group reached LDL-C <2.5 mmol/L compared to 70% of the 100 control patients (p<0.001). One year after target reduction to <1.8 mmol/L, the same proportions were 65% and 36%, respectively (p<0.001). The predominant reason for nonattainment of target in the intervention group was full-dose treatment; for the control group, it was that no medication adjustment was made. After medical titration, at 1 month (baseline), 94.1% in the intervention group achieved target for LDL-C (<2.5 mmol/L) compared to 68.4% in the control group. Mean LDL-C was 0.38 mmol/L lower in the intervention group (p<0.05 for both). At the 12-month assessment, 77.7% of the intervention group attained the LDL-C target compared to 63.2% of the control group, and mean LDL-C was 0.3 mmol/L lower among intervention patients (p<0.05 for both). In the intervention group, 91.9% achieved targets for systolic BP and 96.2% for diastolic BP after baseline titration compared to 65.6% and 82.0%, respectively, in the control group (p<0.05 for both). At 12 months, 68.9% in the intervention group reached the target for systolic BP and 88.1% for diastolic BP, compared to 63.7% and 82.8%, respectively, in the control group (p=0.125 and <0.05). Mean systolic BP was 7 mmHg lower and mean diastolic BP 4 mmHg lower in the intervention group after 1-month titration compared to controls. At 12 months, the mean systolic BP was 1.5 mmHg lower and mean diastolic BP 2.1 mmHg lower in the intervention group.

In our assessment of adherence to statin treatment, 89.3% in the intervention group and 81.7% in the control group were adherent to treatment during a mean follow-up of 3.9 years (p<0.001). In the intervention group, 27.8% discontinued at least once during the period, compared to 20.8% in the control group (p<0.05). The main reason for a first discontinuation was avoidable in both groups: sideeffects without a compelling association with treatment. The main reason for permanent discontinuation was predominantly non-avoidable in the intervention group (advanced disease and dementia) but avoidable in the control group (sideeffects without a compelling association with treatment). Predictors for increased risk for discontinuation were female sex, and for a first event, inclusion in the intervention group. Predictors for reduced risk of non-adherence were ST elevation myocardial infarction as an including event, and for permanent discontinuation, inclusion in the intervention group.

Conclusion

A nurse-led telephone-based method for secondary prevention can encompass a large proportion of an ordinary ACS cohort. Compared to usual care, it is more adaptable to changes in treatment guidelines and leads to better achievement of major risk factor targets as well as improved medication adherence.

Enkel sammanfattning på svenska

Hjärt-kärlsjukdomar är idag den största orsaken till nedsättning av livskvalitet och förtidig död i världen. Detta till trots att utveckling av behandling och förbättringar i livsstilsrelaterade faktorer har halverat dödligheten under de senaste decennierna i västvärlden.

Hjärtkärlsjukdom utvecklas ur åderförkalkning, som är en inflammatorisk process i kroppens pulsådror med en inlagring av blodfetter och efterföljande förkalkning. Detta leder till plackbildning i kärlväggen som sedan helt eller delvis kan täppa till blodflödet och då ge upphov till hjärtmuskelcelldöd (hjärtinfarkt).Den akuta behandlingen av hjärtinfarkt går ut på att med mediciner minska belastningen på hjärtat samt att återställa blodflödet. Detta görs idag oftast med ballongsprängning och inläggande av ”stent” (PCI) eller kranskärls bypassoperation (CABG). För att undvika nya förträngningar är det därefter avyttersta vikt att begränsa åderförkalkningsprocessen genom att behandla de riskfaktorer som driver densamma, så kallad sekundär prevention. Detta görs genom livsstilsförändringar i form av rökstopp, kostomläggning och ökad motion tillsammans med att med mediciner behandla kända risker som högt blodtryckoch blodfettsrubbning. Tyvärr visar alla studier att en stor andel inte uppnår de målnivåer för de mest allvarliga riskfaktorerna som visats minska risken för ny hjärtinfarkt och död. I den här avhandlingen undersöker vi om en sjuksköterskeledd uppföljning via telefon kan förbättra den sekundärpreventiva behandlingen efter hjärtinfarkt.

Syfte

Vi vill specifikt undersöka om en sjuksköterskeledd telefonbaserad uppföljning efter hjärtinfarkt med individuell medicinjustering:

• Är tillämpbar i den normala patientgruppen med hjärtinfarkt.

• Är i jämförelse med dagens vård mer följsam till vetenskaplig utveckling.

• Uppnår bättre resultat under de första 12 månaderna än sedvanliguppföljning avseende behandling av blodfettsrubbning (LDL kolesterol)samt blodtryck.

• Ökar följsamheten till statiner, den vanligaste behandlingen avblodfettsrubbning.

Metod

Avhandlingen baseras på den randomiserad kontrollerade studien NAILED (Nurse-based Age independent Intervention to Limit evolution of Disease) som har två armar: en med akut koronart syndrom (hjärtinfarkt och instabil kärlkramp) samt en med stroke och TIA (transient ischemisk attack). Alla ingående delarbeten förutom studie II fokuserar på patienter med akut koronart syndrom

NAILED-ACS

Alla patienter som lades in på Östersunds sjukhus i Jämtland mellan 2010–2014 för hjärtinfarkt eller instabil kärlkramp beroende av åderförkalkning utvärderades för deltagande. Kriterier för att inte kunna deltaga var oförmåga att följa studiedesignen eller deltagande i annan klinisk studie. Efter utvärdering och godkänt deltagande lottades patienter till en interventionsgrupp eller enkontrollgrupp. Alla patienter oavsett grupptilldelning fick en månad efter utskrivning mäta blodtryck och blodfetter på sin närmaste Hälsocentral. Därefter blev de uppringda av en studiesköterska. Vid samtalet intervjuades patienterna avseende symptom från hjärtat, livsstilsrelaterade faktorer samt följsamhet till behandling och eventuella biverkning av läkemedel. En dylik bedömning med föregående mätning av blodtryck och blodfetter skedde därefter årligen tills minst 3 år efter händelsen eller studiens slut.

Intervention

I interventionsgruppen skedde utefter behov efter svar på mätvärdena en medicinjustering för att uppnå målvärden på blodtryck (<140/90 mmHg) samtLDL-kolesterol (<2,5mmol/L). Justeringen utvärderades efter 4 veckor med nya mätningar och om fortsatt behov upprepades denna rutin tills målvärde uppnåttseller en bedömning om att målet var ouppnåeligt. Patienterna fick även råd om livsstilsförändringar.

Kontroll

I kontrollgruppen skedde ingen justering eller rådgivning, och resultaten av mätningarna skickades till patientens ordinarie vårdcentral och eventuell vidarejustering sköttes därifrån enligt gällande rutin.

Studie I

I den första studien undersökte vi hur stor andel av patientgruppen som kunde och ville delta i studieupplägget. Vi undersökte vilka faktorer som bidrog tilldeltagande eller icke-deltagande.

Studie II

Under 2013 genomförde primärvården i länet en justering av målvärdet för LDLkolesterol hos diabetiker från <2,5 mmol/L till <1,8 mmol/L för att följanationella riktlinjer. Vi antog samma målvärde i studien, och undersökte vad som hände med behandlingen i de två grupperna under det följande året. I dennastudie inkluderade vi patienter med både akut koronart syndrom och stroke/TIA.

Studie III

Vi undersökte resultatet av interventionen i jämförelse med kontrollgruppen med avseende på LDL-kolesterol och blodtryck under det första årets uppföljning.

Studie IV

I fjärde studien undersökte vi andelen som avslutat behandling med statiner under minst 3 år av uppföljning. Vi tog reda på orsaker till avslut och vilkaindividfaktorer som ökar risken för avslut av behandling.

Resultat

Av 907 bedömda patienter kunde 72,9% inkluderas, 11% avstod deltagande och 16,1% exkluderades. Den vanligaste orsaken till exklusion var deltagande i annanstudie följt av demens eller annan svår sjuklighet. Icke-deltagande patienter (exkluderade och de som avstod) var i allmänhet äldre, med mer samsjuklighetsamt lägre utbildningsnivå jämfört med deltagande patienter. Detta avspeglades även i en ökad risk för död i gruppen av icke-inkluderade patienter.

Före förändringen av riktlinjerna för LDL-kolesterol uppnådde 96% av de 101 diabetikerna i interventionsgruppen målnivån på <2,5 mmol/L. Samma andel ikontrollgruppen var 70% av 100 patienter. Ett år efter målnivåjustering till <1,8 mmol/L uppnådde 65% av diabetikerna i interventionsgruppen och 36% ikontrollgruppen den nya målnivån. Den vanligaste orsaken till att inte uppnå den nya målnivån var i interventionsgruppen fulldos behandling och ikontrollgruppen att ingen läkemedelsjustering hade utförts.

Efter den medicinska justeringen 1 månad efter utskrivning uppnådde 94,1% iinterventionsgruppen målvärdet för LDL-kolesterol (<2,5 mmol/L), jämfört med 68,4% i kontrollgruppen. Vid 12-månaderskontrollen uppnådde 77,7% iinterventionsgruppen samma målvärde jämfört med 63,2% i kontrollgruppen. Vid 1 månadskontrollen var LDL-kolesterol i medel 0,38 mmol/L lägre i interventionsgruppen och vid 12 månaders kontrollen 0,3 mmol/L lägre jämfört med kontroller. Gällande blodtryck efter justeringen vid 1 månad uppnådde iinterventionsgruppen 91,9% målvärdet för systoliskt blodtryck och 96,2% målvärdet för diastoliskt blodtryck jämfört med kontrollgruppen där 65,6%uppnådde målet för systoliskt samt 82,0% för diastoliskt blodtryck. Vid 12-månaderskontrollen uppnådde i interventionsgruppen 68,9% målvärdet försystoliskt och 88,1% för diastoliskt blodtryck. Motsvarande uppfyllelse i kontrollgruppen var 63,7% för systoliskt och 82,8% för diastoliskt blodtryck.Medelblodtryck var efter medicinjustering vid 1 månad 7 mmHg systoliskt och 4 mmHg diastoliskt lägre i interventionsgruppen jämfört med kontrollgruppen.Vid 12 månader var medelblodtrycket i interventionsgruppen lägre jämfört med kontroller med 1,5 mmHg systoliskt och 2,1 mmHg diastoliskt.

Vid bedömning av följsamhet till statinbehandling var 89,3% i interventionsgruppen och 81,7% i kontrollgruppen följsamma efter i medel 3,9års uppföljning. I interventionsgruppen avslutade 27,8% sin behandling någon gång under studietiden och motsvarande siffra i kontrollgruppen 20,8%. Denvanligaste anledningen till ett första avslut var i bägge grupperna undvikbar (biverkan utan tydlig relation till medicinering). Den vanligaste anledningen tillett permanent avslut var i interventionsgruppen oundvikbar (avancerad sjukdom inklusive demens) och i kontrollgruppen fortsatt undvikbar (biverkan utanrelation till medicinering). Kvinnor löpte en ökad risk för behandlingsavslut och s.k. ST-höjningsinfarkt vid vårdtillfället minskade risken. Att vara deltagare iinterventionsgruppen ökade risken för ett första behandlingsavslut men minskade risken för att permanent avslut.

Slutsatser

En sköterskeledd telefonbaserad metod för sekundär prevention efter akut koronart syndrom kan innefatta en stor andel av en normalhjärtinfarktspopulation. Jämfört med nuvarande rutin leder metoden till ökad följsamhet vid förändringar i rutiner samt bättre måluppfyllelse av de viktigasteriskfaktorerna högt blodtryck och blodfettsrubbning. Metoden leder även till ökad följsamhet till statinbehandling och minskad risk för undvikbara avslut av densamma.

Background: One of the leading causes of death and disability worldwide is cardiovascular disease (CVD), including acute myocardial infarction (AMI). Despite improvements in medical treatment, management, and care over the years and the halving of mortality in recent decades, there is considerable room for improvement. Following myocardial infarction (MI), a patient is at great risk for subsequent infarctions or other related complications. In addition, the risk of ischemic stroke is increased following MI. Secondary prevention after MI is paramount for reducing further complications and consists of lifestyle changes, optimised medical treatment, and risk factor control of blood pressure (BP) and blood lipid levels. Although secondary preventive measures are effective, the proportion of patients reaching set treatment target levels is disappointingly low.

Most patients are prescribed dual antiplatelet therapy (DAPT) following MI as part of their secondary preventive treatment. Several articles have been published on treatment efficacy based on comparisons with different kinds of antiplatelet drugs and in different combinations. However, little data specifically address the incidence of ischemic stroke after MI in real-world populations. In addition to antiplatelet treatment, secondary prevention comprises risk factor control of hypertension and hyperlipidaemia. Given the low proportion of patients reaching set target levels for BP and blood lipids, new strategies are needed.

Aims: The aim of this dissertation is partly to elucidate if the rapid change in preferred DAPT in Sweden, from clopidogrel to ticagrelor in addition to aspirin, affected the incidence of ischemic stroke in patients suffering AMI (paper I) and in patients suffering AMI who have a history of ischemic stroke (paper II).

The second part of the dissertation aims to investigate the feasibility and implementation of a randomised controlled trial of a nurse-led telephone-based secondary preventive program, and to assess the proportion of patients who can be included in an unselected acute coronary syndrome (ACS) population (paper III). Furthermore, the aim of the trial was to assess the long term results regarding systolic BP (SBP), diastolic BP (DBP), and low-density lipoprotein cholesterol (LDL-C) after 36 months of intervention and follow-up compared to a control group receiving usual care (paper IV).

Methods: Papers I and II examined the impact of a change in the antiplatelet regimen following MI in regard to ischemic stroke occurrence. Data were obtained from the Swedish Register of Information and Knowledge about Swedish Heart Intensive Care Admissions (i.e., RIKS-HIA). The register was combined with the National Patient Register (NPR) and the Cause of Death Register (CDR) in order to obtain data on stroke occurrence. Patients with AMI and treated with either clopidogrel or ticagrelor were assigned to one of two cohorts, each covering a 2- year time period, with the initial prescription of ticagrelor (20 Dec 2011) used as a cutoff point. Patients in the early cohort (n=23,447) were treated exclusively with clopidogrel, whereas those in the later cohort (n=24,227) were treated with either clopidogrel (47.9%) or ticagrelor (52.1%). In paper II, the same methodology was used, but with a study sample restricted to AMI patients with a history of ischemic stroke. In paper II, there were 1633 patients in the early cohort and 1642 in the late cohort. In the late cohort, 66.3% patients were treated with clopidogrel and 33.7% with ticagrelor. Kaplan–Meier analysis was used to assess the risk of ischemic stroke over time, with multivariable Cox regression analysis used to identify predictors of ischemic stroke. Nurse-based Age independent Intervention to Limit Evolution of Disease (Papers III and IV were based on the NAILED)-ACS trial. The NAILED-ACS trial was an open randomised controlled trial of whether a nurse-led telephone-based follow-up and medical titration after MI or unstable angina achieved lower levels of BP and LDL-C than usual care. In paper III, patients admitted for ACS during January 2010 and December 2013 were evaluated for participation. Factors predicting participation and non- participation were assessed using logistic regression. Mortality rates after one year among included and excluded patients and patients declining participation were assessed using Kaplan–Meier analysis. For paper IV, all patients admitted with ACS at Östersund Hospital between January 2010 and December 2014 were screened for inclusion based on their ability to participate in a telephone- based follow-up. Participants were randomised into two parallel groups, an intervention group and a control group receiving usual care. BP and LDL-C were measured at 1, 12, 24, and 36 months. The baseline consisted of randomised patients who completed the one-month follow-up. The intervention group  received counselling and medical titration to attain treatment targets (BP <140/<90 mmHg and LDL-C <2.5/<1.8 mmol/L). Adjusted means stratified by sex and type of ACS were calculated for SBP and DBP and LDL-C. The proportion of patients who achieved treatment target levels at the end of the study was also assessed.

Results: Among the general AMI population treated with either clopidogrel or ticagrelor, the incidence of ischemic stroke after one year was 2.8% in the early cohort vs. 2.4% in the late cohort (p=0.001) (paper I). The study population in paper II, in which all patients had a history of previous ischemic stroke, was overall older and had a higher prevalence of comorbidities than the population in paper I. In paper II, incidence of ischemic stroke in the early cohort was 12.1% vs. 8.6% in the late cohort (p<0.01). Corresponding incidence of intracranial bleeding for the population in paper II was a non-significant 1.2% vs 1.5%.

In the feasibility study of the NAILED-ACS trial (paper III), 907 patients were assessed for inclusion. Among these, 72.9% could be included (n=661), 146 patients (16.1%) were excluded, and 100 patients declined participation (11 %). Reasons for exclusion were mainly participation in another trial, dementia, inability to use a telephone, and advanced disease. Examples of predictors of both exclusion and declining participation were older age, lower functional status, and lower education. Non-participating patients had significantly higher mortality rates at one year compared to participating patients.

Paper IV presents the final results of the NAILED-ACS risk factor trial in which a total of 962 patients were randomised and completed the one-month follow- up. Of this group, 797 were available for analysis after 36 months. Compared to the control group, in the intervention group, mean SBP was 4.1 mmHg lower, mean DBP was 2.9 mmHg lower, and mean LDL-C was 0.28 mmol/L lower (p<0.001 for all). The proportions of patients reaching treatment target goals for SBP, DBP, and LDL-C were significantly higher in the intervention group. In regard to SBP, 77.6% of intervention patients achieved treatment target levels, compared to 62.9% in the control group. Corresponding numbers for DBP were 90.9% vs. 80.8% and for LDL-C, they were 65.6% vs. 53.1%

Conclusion: The incidence of ischemic stroke was significantly lower in a cohort of AMI patients following a change in preferred treatment from clopidogrel to ticagrelor (paper I). In AMI patients with a history of ischemic stroke (paper II), the incidence rate of ischemic stroke was significantly lower in the late cohort compared to the early cohort, and overall incidence rates were markedly higher than in paper I.

The NAILED-ACS trial was shown to be both feasible (paper III) and successful, with a higher proportion of patients reaching treatment target levels in the intervention group, and significantly lower mean values for SBP, DBP, and LDL- C (paper IV).