Medication non-adherence in heart transplant patients

1 Universidade Federal de São Paulo, Escola Paulista de Enfermagem, São Paulo, SP, Brazil. ABSTRACT Objective: To measure medication non-adherence in patients after heart transplantation using the Basel Assessment of Adherence to Immunosuppressive Medications Scale (BAASIS) and the Visual Analogue Scale (VAS); to compare the results of biopsies performed with the prevalent comorbidities and survival. Method: Quantitative historical cohort. The population consisted of patients undergoing transplantation between 2009 and 2016. Results: Participation of 60 patients. The measurement using the BAASIS was 46.7% of non-adherence and 53.3% of patient adherence. The group with greater difficulty in non-adherence reported up to 2 hours delay of medication intake in relation to the prescribed time (25%), although there was no interruption in medications. The initial diagnosis was Chagas disease (33.3%). The studied comorbidities were systemic arterial hypertension (SAH), diabetes mellitus (DM), dyslipidemia (DLP) and chronic renal failure (CRF). Conclusion: Assessment using the BAASIS showed medication non-adherence in 46.7% of heart transplant patients. The VAS according to patients’ self-report and nurse’s assessment showed high values (93.3% vs 83.3%). The BAASIS tends to address the difficulties reported by patients, when there is a change in doses, delays or anticipations of time and dose.


This complica
ion reduces survival in the late stage of transplantation (8) .

The results related to adherence to treatment guidelines are important for its success, for example, taking the medication correctly; following the prescribed diet; seeking changes in lifestyle, such as physical activity practice; and not smoking (7) .Since it is difficult to measure adherence, predictors have been studied to help monitor the failures arising in the treatment process (9) .

Among the predictors to detect rejection, such as acute worsening of the graft and its accurate diagnosis, endomyocardial biopsy (EMB) and control by immunosuppressive drugs are used with the function of modulating such episodes (5) .In the study in question, the institutional protocol adopts the recommendat ons of the International Society for Heart Transplantation (ISHLT) for proposing follow-up procedures in the first year after heart transplantation (6) .

The assessment of medication adherence in tran planted patients has been performed especially by means of biopsies and biochemical measurements, but the assessment in the light of patient information is poorly studied.In addition to the clinical procedures already adopted, the objective was to measure medication nonadherence in patients after heart transplantation using the Basel Assessment of Adherence to Immunosuppressive Medication Scale (BAASIS) (10) and the Visual Analogue Scale (VAS) (11) by comparing with the results of biopsies performed and identifying the survival of these patients and their prevalent comorbidities.
METHOD


Type of sTudy

This is a quantitative, historical cohort study.


scenario

The research was performed to measure non-adherence to immunosuppressive therapy in heart transplant patients treated at the Institu o Dante Pazzanese de Cardiologia (Portuguese acronym: IDPC), located in the city of São Paulo, Brazil.


populaTion

The study population were all heart transplant patients seen at the outpatient clinic of the IDPC from August 2014 to May 2016, totaling 60 patients.

Patients with at least one year after transplant and five years of follow-up were selected, and those with less than one year of treatment were excluded.


daTa collecTion

Data collection was performed by the nurse responsible or the study, from an invitation during visits to the heart transplant clinic of the IDPC.Patients with heart transplantation (Cardiac TX) with advanced heart failure and refractory to treatment are treated daily by a specialized team in the IDPC.

Multidisciplinary care is offered in nursing consultations focused on providing guidance and teaching selfcare in the search for better quality of life for patients and their families.

The study outcome was medication non-adherence assessed with use of instruments validated in Brazil, namely the BAASIS (10) and the VAS (11) through patients' self-report and nurses' reports of adherence.

The VAS was applied by patients themselves with the self-report of their perception of medication adherence and simultaneously by the nurse professional involved in the study, who followed the patients in nursing consultations throughout the pre-and posttransplant process (12) .

An instrument o collect information and trace the sociodemographic profile was also developed, including age, sex, marital status, employment status and education, in addition to clinical variables such as post-transplant time, immunosuppressive therapy, comorbidities and rejection episodes (12) .

The resu t of endomyocardial biopsies to assess the degree of graft rejection was the third variable of choice for the analysis of non-adherence measured through the instruments.These were performed according to the Biopsy Protocol of the IDPC, totaling 11 biopsies in a one-year period.In the first month, one biopsy is performed per week, totaling four.In the second and third months, biopsies are per ormed every two weeks, and monthly until the sixth month, ending with one year after the transplant (5) .Biopsies are classified as: 0R, absence of infiltrates; 1R, focal and/or diffuse infiltrate without necrosis; 2R infiltrated with cellular aggression or myocyte damage; 3R diffuse inflammatory infiltrate with necrosis, vasculitis, hemorrhage and edema (9) .

During outpatient visits, patients filled out the assessment instruments and biopsy-related data were taken from medical records.


analysis and TreaTmenT of daTa

Descriptive and analytical statistics were used for data treatment.The association between two categorical variables was verified through the chi-square test, or alternatively, in cases of small samples, with the Fisher's exact test.Student's t test was used to compare means between two groups.

One of the assumptions of the Student's t test is the normality in data distribution, which was verified using the Kolmogorov-Smirnov test.In case of data normality violation, the means were compared using the non-parametric Mann-Whitney test.

The linear association between

ion of selfrep
rted adherence and that of nurses was assessed th

ugh Spear
an's correlation (rS) given the small variation in the amplitude of the score.

In the survival analysis, the probability of a patient survival after a certain period was estimated by the adherence with use of the Kaplan-Meier model.The

og Rank tes
(Mantel-Cox) was used for comparisons.

For all statistical tests, a significance level of 5% was adopted.Statistical analyzes were performed using t e statistical software SPSS 20.0.


eThical aspecTs

The study complied with assumptions of Resolution number 466/12 of the National Health Council and wa

submitted to the
Research Ethics Committee of the Instituto Dante Pazzanese de Cardiologia and the Plataforma Brasil (CEP number 647.508), (Date of the Rapporteur: 05/14/2014).The Informed Consent form was applied and signed by study participants.


RESULTS

According to Table 1, the predominant sociodemographic profile f patients was the following: men (71.7%); married (66.1%); does not live alone (88.3%); retired (43.3%); white color (63.3%); and 11.7% had a college degree (incomplete or complete).Da a of 60 patients were analyzed, and their mean age was 47.2 years (SD=15.6years), minimum age of 14 years and maximum of 75 years.The median of age was 47.5 years.

The McNemar's test was used to assess pr valence before and after transplantation; for systemic arterial hypertension (SAH), the results remained similar, 48.3% and 30.0%(p=1,000), respectively.However, occurrences (p=0.008) of diabetes mellitus (DM) increased from 11.7% to 25%.The percentage (p=0.012) of patients with hronic renal failure (CRF) also increased from 15.0% to 30.0%.Regarding comorbidities before transplantation, 38.3% had SAH and 30.0%had dyslipidemia (DLP), and after transplantation these percentages remained similar (48.3% and 30.0%, respectively).On the other hand, the occurrences of DM increased (p=0.008)(11.7% to 25.0%), as well as those of CRF (p=0.012)(15.0% to 30.0%).Additionally, 35.0% were hospitalized, although from a cause other than rejection.The prevalent diagnosis of the transplanted group was Chagas heart disease with 33.3%, followed by dilated heart disease with 26.7% and ischemic heart disease with 23.3%.The higher incidence of chagasic etiology can be related to the care of the population coming from endemic areas such as the North and Northeast of Brazil.

Table 2 shows that 15% stopped taking their immunosuppressive medications once or twice in the four weeks prior, and five patients skipped one or two doses, while 25% of patients acknowledged having changed the prescribed schedule, which could compromise their graft quality and eviden e of rejection.

Although 46.7% of patients had difficulties with medication, that is, they presented non-adherence, no patient st

ped taking immunosuppressive me
ications completely without medical advice.At the same time, 53.3% presented medication adherence in 100% of the instrument questions with a positive response.

Most patients used prednisone -PRD (95%), mycophenolate mofetil -MMF and clyclosporin (58.0%).A small portion used two or four medication (5.0%).As shown in Table 3, there were no associations between adherence and rejection.However, rejection in the 1 st month was marginally significant (p=0.088),indicating lower rejection in patients with adherence (75.0%), compared to nonadher nt patients (92.9%).At other times, these percentages were very similar.The results of biopsies were distributed in the 1 st , 3 rd and 6 th months, and those above 1R were considered as rejection.
s results were similar in the other biopsies, it is not possible to infer that the improvement was the result of better patient adherence.This fact may be related to changes in the dosage of immunosuppressants d ring treatment.

In the VAS self-reported perception (11) , nurses' assessment (93.3% X 82.5%, p<0.001) was superior.Additionally, there was no s

nificant correla
ion between the two adherence values (rS=0.008,p=0.950).Patients with adherence were those who indicated no or never in all items.According to the BAASIS (10) , 53.3% had medication adherence, while patients' self-report (VAS) reached 93.3% and the lowest value in nurses' assessment using the VAS was 82.5%. (11).Survival of transplant patients was an

yzed at
, 2, 6 and 7 years after transplantation.There were no differences in patient survival in comparisons by adherence.There were six deaths (10.0%), half of which in each adherence group.According to the Kaplan Meier model, the mean survival time was 85.1 months (95%CI=[78.6 ; 91.6]) (Table 4).


DISCUSSION

In recent years, heart transplantation has proven to be an effective therapy for patients with advanced heart ailure and refractory to all available therapies (2) .This remains a great choice for patients who have undergone several clinical and surgical treatments that have shown limited activities and clinical worsening over time.Thus, transplantation improves quality of life and prolongs survival (5) .

Regarding sociodemographic variables of heart transplant patients, their profile does not differ from other Brazilian centers, where the median age of 47.5 years characterizes them as young adults.The multicenter study with  (10)   participation of Brazilian transplant centers obtained similar variables, such as an mean age of 53.7 years, predominance of men (72.7%), married (68.9%), employed (29.7%), university education (34.1%) and white (85.9%) (13) .

A recent American study assessed the impact of racial and economic disparities with data from the United Network for Organ Sharing (UNOS), including 33,893 adult patients transplanted between 1994 and 2014.Although transplant disparities have narrowed over time, especially the socioeconomic level (health insurance, education, black race and neighborhood), they are still not explained by differences in the clinic or characteristics of grafts, which can be an important differential for survival.There are few studies on this relationship between survival of the recipient and organ, especially in associations with borderline donors (14) .

The First Brazilia Registry of Heart Failure, considering clinical aspects, quality of care and hospital outcomes, included a total of 1,263 patients (64 ± 16 years old, 60% women) and data from 51 centers in different regions of Brazil.The most common comorbidities were arterial hypertension (70.8%), dyslipidemia (36.7%) an diabetes (34%) (15) .

Regarding comorbidities before transplantation, 38.3% and 30.0%had SAH and DLP, respectively.The study showed that 33.3% of initial diagnoses were due to Chagas disease.

There was also 15% of Chagas reactivatio after transplantation (16) .Even receiving preventive treatment with recommended medication (benznidazole) to inhibit parasitic reactivation, the disease can occur as a result of the use of immunosuppression (17) .

Thus, Chagas disease is a negative predictor for the quality of life of patients and graft survival, because of the disease reactivation that increases the mortality of transplanted patients (18) .

Despite the implementation of policies to control the disease transmission in Brazil, difficulties still exist.In addition to modern drugs for its control, the disease involves sanitary, educational and economic measures.Studies have shown high mortality in chagasic patients wi hout the prospect of heart transplantation, and a survival of only 1.5 months (13) .

With the expansion of Chagas disease to the United States of America and Europe, it ceased to be a disease of poor countries.Although small, this may be another cause for the development of heart disease, which brings the need for heart transplantation in these other countries and the same consequences, such as disease reactivation (18) .

In this study, there was an increase (p=0.008) in occurrences of DM that went from 11.7% to 25.0% in the post-transplant period, as well as an increase (p=0.012) in occurrences of chronic renal failure from 15.0% to 30.0%.Additionally, 35.0% were hospitalized, although from causes other than rejection.

The onset of other clinical conditions is another worrying reality that may compr

ise the gra
t.The reason for this result is that post-transplant diabetes mellitus (DMPT) is a possible complication due to the use of immunosuppressants, mainly corticosteroids and calcineurin inhibitors (cyclosporin and tacrolimus) (19)(20) .

Thus, in addition to the need for regular intake of medications for graft preservation, other medications can be added for controlling the comorbidities, thereby leading to an a herence process that involves several factors.

According to BAASIS a four-item questionnaire, conducted through patient interviews assessing self-reported adherence, only if 100% of responses, all positive, is considered adherence.Regarding the questions, one can assess medication adherence or non-adherence of transplant patients (10) .

In this sense, the study showed that 46.7% of heart transplant patients treated in a Cardiology Center in the city of São Paulo had time-related medication non-adherence; 2 hours before or after the prescribed time and dose change.The highest percentage of medication non-adherence was by intake with more or less 2 hours' difference in relation to the prescribed time (25.0%).No patient stopped taking medications completely without medical advice and 53.3% had medication adherence.

A similar study was conducted in Israel using the BAASIS (10) to assess medication adherence.It was found that in the past four weeks, 64% of patients had problems with the implementation of medication.For example, they missed a dose or skipped two or more doses, took me ication 2 hours before or after the recommended time, or changed the prescribed amount.The highest score was for the item "non-adherence over time", 56.9% of patients.Three patients (3%) had discontinued the medication (21) .

A study that analyzed secondary data of 36 transplant centers in 11 countries, in which Brazil participated, on four continen s, called the Research Initiatives Group: Chronic Disease Management and Adherence to Transplant Study (BRIGHT) (13) totaled a sample of 1,397 and 83.3% patients responded o the study.It demonstrated that non-adherence to the implementation of immunosuppressants was observed in 37.4% of participants.More specifically, the prevalence of immunosuppressive non-adherence was 17.3% by non-adherence, 1.9% by medication not taken on holiday, 28.7% by the time of taking medication, and 1.6% by dose change.For discontinuation of medication, a prevalence of 0.5% was found (15) .

In a Brazilian study with kidney transplant patien s, general BAASIS scores (10) were used, calculated from the arithmetic sum of the score attributed to questions related to time, dose change, holiday dose and time change; 58.6% of recipients reported total adherence to the immunosuppressive medication, 41.4% did not adhere and 18.8% did not adhere to at least one or more of the four situations assessed (related o time, dose change, holiday dose, and time change) in the previous four weeks (22) .

The concept of adherence to drug therapy has been increasingly studied to assess the effectiveness of treatment (10)(11)(12) .Thus, it is emphasized the need to reflect on the process of measuring adherence, scales used, and factors related to individuals evaluated.The appropriateness of the measurement of adherence must be discussed, as well as all perspectives of factors leading to non-adherence, such as forgetting to take the medication, the lack of perception of the disease/ health status, and social characteristics.It is also important RESUMO Objetiv : Mensurar a não adesão medicamentosa nos pacientes pós-transplante cardíaco mediante o uso da Escala Basel para Avaliação de Aderência a Medicamentos Imunossupressores e Escala Analógica Visual; comparar os resultados das biópsias realizadas, com comorbidades prevalentes e sobrevida.Método: Coorte histórica de abordagem quanti ativa.A população foi composta de pacientes transplantados no período de 2009 a 2016.Resultados: Participação de 60 pacientes.A mensuração da não adesão por meio do instrumento Escala Basel para Avaliação de Aderência a Medic mentos Imunossupressores foi de 46,7% e adesão de 53,3% dos pacientes.O grupo com maior dificuldade de não adesão foi aquele com relato de atraso de até 2 horas do prescrito (25%), porém, sem interrupção nas medicações.O diagnóstico inicial foi Doença de Chagas (33,3%).As comorbidades estudad s foram hipertensão arterial sistêmica, diabetes mellitus (DM), dislipidemias e insuficiência renal crônica.Conclusão: A avaliação por meio da Escala Basel para Avaliação de Aderência a Medicamentos Imunossupressores verificou não adesão medicamentosa de 46,7% dos pacientes transplantados cardíacos.A Escala Analógica Visual pelo autorrelato do paciente e avaliação do enfermeiro apresentaram valores elevados (93,3% vs 83,3%).A Escala Basel para Avaliação de Aderência a Medicamentos I unossupressores tende a se aproximar das dificuldades informadas pelos pacientes, quanto há alteração de doses, atrasos ou antecipações de horário e dose.


DESCRITORES

Adesão à Medicamento; Transplante de Coração; Imunossupressores; Enfermagem Cardiovascular.


RESUMEN

Objetivo: Medir la falta de adherencia a la medicación en pacientes después de un trasplante de corazón utilizando la Basel Assessment of Adherence to Immunosuppressive Medications Scale (BAASIS) y la Escala Visual Analógica (EVA ; comparar los resultados de las biopsias realizadas con las comorbilidades prevalentes y la supervivencia.Método: Cohorte histórica con un enfoque cuantitativo.La población consistió en pacientes trasplantados de 2009 a 2016.Resultados: Participación de 60 pacientes.La medición utilizando la BAASIS fue del 46,7% para no adherencia y del 53,3% para la adherencia de los pacientes.El grupo con la mayor dificultad en la no adherencia fue to review the educational process of preparing the individual before receiving the organ and the discharge guidance provided after the transplant to achieve the proposed goals.Several strategies to assist patients in maintaining drug adherence must be implemented, such as active search by telephone and mobile digital technologies (23)(24) .

According to Table 3, there were no associations between adherence and rejection.The rejection in the 1 st month was marginally significant (p=0.088),indicating lower rejection in patients with adherence (75.0%), compared to patients without adherence (92.9%).At other times, these percentages were very similar.The results of biopsies were distributed in the 1 st , 3 rd and 6 th months, and results of rejection evaluation above 1R were considered as rejection.Results were similar in the other biopsies.

It is estimated that 15% to 60% of late rejections and 5% to 36% of graft losses are related to non-adherence to the correct treatment of solid organs (23)(24) .

The self-report assessment depends on the respondent's sincerity, despite its low cost and the possibility in the clinical follow-up.In the study, the self-reported perception (VAS) (11) was superior to the nurse's assessment (93.3% X 82.5%, p<0.001).Additionally, no significant correlation was found between the two forms of assessing adherence (rS=0.008,p=0.950).

In survival analyzes of transplanted patients at 1, 2, 6 and 7 years after transplantation using the Kaplan Meier model, the mean survival time was of 85.1 months (95%CI=[78.6;91.6]).There was no difference in patient survival in comparisons by adherence, although six deaths (10.0%) occ rred, of which half in each adherence group.

In a study conducted eight years ago in the same service, survival after orthotopic transplantation was of one year in 72.7%, five years in 61.5% and seven years in 56.4%.Survival after transplantation was correlated with the variables age, cause of death and donor sex, and if the transplant was the patient's first heart surgery or not (25) .

One can consider as a limitation the instruments used to assess adherence.For example, the BAASIS (10) assesses patient dherence to drug therapy only with 100% of positive responses for non-adherence.Visual analog scales involve a subjectivity of assessment, in which the profession l's perception can be influenced by other characteristics of patients and their history, the same way that patients' self-report can also be influenced by personal concepts and values inherent to treatment.


CONCLUSION

This study allowed the conclusion that 46.7% of patients had difficulties with medication, i.e., presented non-adherence.

The highest percentage of on-adherence was when taking immunosuppressive medications with more than 2 hours' difference from the prescribed time (25.0%).No patient completely stopped taking immunosuppressive medications without medical advice, while 53.3% had medication adherence measured by 100% of items of the instrument with a positive response.

The init al prevalent diagnosis of the transplanted group was Chagas heart disease with 33.3%, followed by dilated heart disease with 26.7% and ischemic heart disease with 23.3%.Most patients used a triple regimen (prednisone, mycophenolate mofetil and cyclosporin).In addition, the self-reported perception -VAS was higher than the nurse's assessment (93 3% X 82.5%, p<0.001).This scale is criticized for its low sensitivity in assessing the real situation of patients' non-adherence, and after this study, it is no longer indicated to compose this type of assessment.There was no significant association between adherence and rejection using biopsy values (p=0.088).Systemic Arterial Hypertension and DLP had similar results; 38.3% and 30% in the pre-transplant and 48.3% and 30% in the post-transplant period.The mean 7-year survival was 85.1 mon

t's t test fo
paired samples -p<0.001.


Table 2 -
2
Distribution of patients by (BAASIS)


Do you recall of for