Is Noncompliance Among Adolescent Renal Transplant Recipients Inevitable?

METHODS

Medical records of 79 patients (46 boys and 33 girls) who underwent their first kidney Tx were reviewed. All patients were followed closely in our unit for at least 1 year.

Noncompliant behavior was defined as discontinuation of 1 or more immunosuppressive medications, deviation from prescribed dose or frequency, and/or “drug holidays.” Patient self report and/or cyclosporine A (CsA)/tacrolimus levels (CsA <20 ng/ml, tacrolimus <2 ng/ml) served as indicators of noncompliance.

The mean follow-up period was 4.8 ± 2.7 years (range: 1–13). All children were followed at our unit during the predialysis period, while on dialysis, and after kidney Tx by the same team of pediatric nephrologists. Kidney Txs were performed in 2 centers in Israel or abroad. Fifty-one patients were of Arab descent, and 28 were Jewish. Data on the number of Txs, the source of the organ (cadaveric [CTx], living related [LRTx], and living unrelated donors [LURTx]), age at Tx, time on dialysis, and post-Tx follow-up are presented in Table 1. All LURTx were commercial Txs and were performed abroad (mostly in Iraq) at the parents' initiative and against our advice, as this practice is illegal in Israel. Four patients underwent preemptive Tx. All but 2 patients received triple immunosuppressive therapy, including CsA/tacrolimus (41/36), mycophenolate mofetil/azathioprine, and prednisone. Nine patients included in the tacrolimus group were switched from CsA because of adverse effects.

Adolescents or young adults were included in this study when they were followed in our service for at least 1 year after their 12th birthday. The incidence of noncompliance after Tx was compared with the pre-Tx period in a subgroup of 53 patients who had been on dialysis for at least 6 months.

Routine follow-up for our post-Tx patients entailed twice-weekly visits during the first 3 months after transplantation, with a gradual decrease in frequency to a monthly visit after 1 year. Patients who missed their appointment to the clinic received a telephone call from our team. Blood tests, including CsA/tacrolimus levels, were drawn at each visit. None of our adolescent or young adult patients were transferred to an adult transplant unit. For the last 3 years, patients who received a transplant were also followed by a psychologist, who is a member of our team.

RESULTS

Noncompliance with immunosuppressive medications was documented in 13 (5 girls) of 79 patients. Age at Tx was 13.6 ± 4.8 years (SD) (range: 2.5–22.5), and the age when noncompliance was first recognized was 16.1 ± 5.1 years (range: 6–27). All but 1 patient were older than 12 years (Table 2). Two thirds of the patients demonstrated noncompliant behavior before Tx, while on dialysis. Among children with adequate post-Tx compliance, the frequency of poor drug adherence while on dialysis was significantly lower (Table 2).

The first indication of “drug holiday” in 2 of the 13 patients was an acute rejection episode (Table 3). In the remaining 11 children, it was low levels of plasma calcineurin inhibitors, with very mild (<20%) or no elevation of serum creatinine levels. In 7 of these 11 patients, repeated episodes of “drug holiday” led to acute rejection later (21.4 ± 13.2 months after the first decrease in plasma CsA/tacrolimus level). Noncompliant behavior with frequent “drug holidays” did not influence graft function in the remaining 4 children for at least 32.3 ± 16.8 months. Chronic rejection ensued in all 9 patients who developed acute rejection. Three of them progressed to end-stage renal failure 3 to 7 years after the first indication of noncompliance. Three patients resumed taking immunosuppressive medications regularly after the first episode of acute rejection. Others continued with “drug holidays” and subsequently experienced repeated rejection episodes.

Noncompliance varied with graft source: it was highest (26.2%) among CTx recipients, lower after LRTxs (12.5%), and not documented in any patient after LURTx. The time from Tx to the first indication of noncompliance was 19.1 ± 15.5 months in the CTx group and 20 and 48 months in the 2 patients from the LRTx group. In all 3 groups (CTxs, LRTxs, and LURTxs), there was a similar percentage of children above 12 years of age (Table 1). Most of these adolescents were responsible for taking their own medications. The prevalence of noncompliance among adolescents was 45.5% in CTx group, 28.6% in LRT group, and 0% in the LURTx group (Table 4). Only 1 of 17 children who were younger than 12 years and underwent CTx and none of the 15 children in the other groups was found to be noncompliant with immunosuppressive medications.

The prevalence of noncompliance was similar between Jewish and Arab children (21.4% vs 13.7%), a consistent finding even when patients after CTx were evaluated separately (29.4% vs 30.0%). Gender, type of immunosuppression (CsA or tacrolimus), and duration of dialysis before transplantation did not affect the rate of low drug adherence (data not shown).

Noncompliance with nonimmunosuppressive medications (including antihypertensives, magnesium and phosphorus supplements, and erythropoietin injections) was much more common than with immunosuppressive medications: it was documented in all patients who were noncompliant with immunosuppressive medications and in 10 additional children.

DISCUSSION

Noncompliant behavior in post-Tx patients may be a problem that is unique to adolescents who become independent enough to be made responsible for taking their own medications. During this period, the risk for noncompliance and the consequent deterioration of graft function in kidney Tx recipients rises dramatically. The problem of poor drug adherence in adolescent patients has been described in other chronic diseases. For example, in asthma, the most common chronic illness of childhood,⁹ noncompliant behavior is particularly high in adolescents^10,11 and may contribute to the high level of morbidity and mortality in this age group.¹⁰ Low compliance with prednisone and antibiotic treatment was also found in half of adolescent outpatients with lymphoblastic lymphoma.¹²

For various psychological reasons, being an adolescent makes one more vulnerable and less approachable. Teenagers with chronic illnesses realize that they remain different from their peers despite their continual effort to lead a normal life. These existential issues bring about anxiety, and denial becomes one's worst enemy in a situation in which compliance is essential. These issues should be discussed with each adolescent and his or her family individually, based on their cognitive and emotional level. Among the teenagers who were responsible for their own medications and followed in our service, the overall incidence of noncompliance was 26.2% compared with only 3% in children who were younger than 12 years.

There is wide variation in post-Tx noncompliance rate as recorded in previously published studies.^1,13–16 This may be attributable to the use of different definitions and methods of assessment of noncompliance. Meyers et al¹⁵ found an overall noncompliance rate of 22% with no difference in frequency between teenagers (11–15 years) and all other age groups. Blowly et al,¹⁶ using an electronic-monitoring device that detected access to medication containers, identified noncompliance in 21% of post-Tx adolescents. Watson¹⁷ reported a very high noncompliance rate with consequent graft loss in 40% of adolescents and young adults after they had been transferred from pediatric to adult transplant units. A high rate of noncompliance was noted in our patient population despite that none of them was transferred to an adult unit.

The rate of noncompliance in patients who underwent CTx was a remarkable 45.5%. This figure may still underestimate the scope of the problem because our means of identifying noncompliance are limited. Noncompliance was suspected in cases of poor clinic attendance, absence of signs of steroid toxicity despite high prescribed doses, suboptimal control of hypertension, or unexplained late graft dysfunction. Low plasma CsA/tacrolimus levels were usually confirmatory of “drug holiday.” Adequate levels, however, do not exclude irregular use and may have merely reflected adherence to medications during the few days preceding testing. Patients with either low drug levels or declining renal function were questioned to elicit a report of compliance.

In most patients, an episodic decrease of plasma CsA/tacrolimus levels with stable renal function was the first indication of noncompliance. Acute rejection episodes occurred significantly later with a mean lag time of 21 months after the first sign of noncompliance. These episodes in turn led to gradual deterioration of graft function.

The exact dose required for adequate immunosuppression varies between individual patients and is difficult to assess. Therefore, the recommended regimen is based on standard protocols. In 4 of 13 patients, graft function remained stable despite noncompliant behavior. This may suggest that the standard regimen may be excessive for individuals with some degree of tolerance. Preservation of renal function after discontinuation of immunosuppressants has previously been described.^18,19

The Hartford Transplant Center group of investigators did not find in adults an association between the source of the organ (CTx or LRTx) and the tendency to be noncompliant.^4,5 In contrast, the majority of our noncompliant patients were from the CTx group, and only 2 were from the LRTx group. All patients who underwent LURTx (commercial) were found to be adherent to their medications. This may suggest that financial considerations motivate patients and their families to adhere to treatment. Alternatively, better compliance in the LURTx group may be attributable to better preparedness of the patient and his or her family for Tx, which is preceded by a lengthy and arduous process including raising the necessary funds, planning a long trip abroad (which was the first of its type for all of the participants), and establishing connections with liaisons in the health care chain, all in a clandestine atmosphere because of the illegality of commercial Tx. This requires initiative, persistence, and ample resources. It seems that parents who were capable of making a difficult decision to undergo LURTx and successfully pursued their initiative demonstrate similar abilities when strict adherence to therapeutic regimens is required and transmit this imperative to their children.

Individuals who undergo CTx, although cognizant of the procedure and its potential benefits, invest little of themselves as they await an unexpected telephone call and then are suddenly confronted with the availability of an organ for transplantation. Even the simple but potentially anxiety-provoking questions of, “Who is the donor?” will be addressed only after surgery. The recipient will subsequently undergo a long process of adjustment to the new lifestyle, a vulnerable situation for the patients and their accompanying relatives. Suboptimal psychological readiness may lead to noncompliant behavior. Educating adolescents who are awaiting CTx regarding the process that lies before them with emphasis on the virtue of medication compliance may have an impact on their post-Tx outcome. Participation of dialysis (pre-Tx) patients in a support group with post-Tx adolescents may help to crystallize the benefits of renal Tx and enhance the importance of drug adherence.

Wolff et al²⁰ suggested that noncompliance is multifactorial and should not be perceived solely as a fault of the patient. They claim that interaction between patients and their care providers plays a major role in adherence to treatment plans. Our observation that all of the children who underwent LURTx showed compliant behavior and had strong family support underscores the pivotal role of the cohesiveness of the family in determining compliance. Drug adherence in young recipients similarly emphasizes the important role of the family. It should be noted that the families of children who underwent LURTx received less support before Tx from the medical team, as they chose an illegal practice and therefore had to make the arrangements for Tx as well as face the moral dilemmas involved on their own. Living related kidney donors demonstrate altruistic behavior, which seems to be appreciated by the recipients as manifested in low rate of noncompliance. This is in contrast to families of children who undergo CTx, who display a passive attitude. The correlation between family structure and noncompliant behavior should be addressed in a more systematic study. Although we object to the practice of LURTx, there are lessons that may be learned from this group of patients and applied to others to achieve better drug adherence. Our experience substantiates the notion that although noncompliance is prevalent, it is not inevitable.

How should the experience with renal Tx recipients be applied to adolescents with chronic ailments in general? A number of suggestions have already been proposed in the literature.^8,13,21–24 These include increasing the frequency of clinic visits in suspected patients, reducing the number of nonessential drugs, using long-acting medications, inviting the truly responsible adolescent for follow-up without his or her parents, and being careful to provide relevant information based on the patient's cognitive abilities. Reducing the dose of medications with potential cosmetic side effects, such as steroids, is imperative. Our results suggest that children with even a mild cognitive deficit require continuous supervision by a family member or caregiver. Transfer of medical care of noncompliant adolescents to adult units may result in deterioration of their health and should be conducted only once the patient is mature and psychologically prepared. The establishment of support groups for adolescents may have beneficial effects, especially if these are guided by a psychologist. This will enable them to share without shame their unique medical experiences with their peers, as well as foster learning from one another of the health benefits of adhering to medications. Because individual adolescents often seek to conform with their peers, support groups may be particularly effective in promoting compliance in this age group. Finally, we must utilize multidisciplinary resources and work creatively to better understand and improve communication between staff and families, as well as between patients and their parents. This underscores the role of the general pediatrician, who should be aware of the extent of noncompliance among adolescents with chronic illnesses and be able to detect early signs of this practice.