Allowing Adolescents to Weigh Benefits and Burdens of High-stakes Therapies

Dr Kriti Puri, Dr Caridad M. de la Uz, and Dr Antonio G. Cabrera

Long QT syndrome is a potentially fatal condition characterized by predisposition to ventricular tachycardia (VT), which may cause syncope and death.^1–3 Long QT syndrome may be caused by any of the 17 mutations described so far (15 autosomal dominant and 2 autosomal recessive). However, the 3 most common types of long QT syndrome are types 1 (KCNQ1), 2 (KCNH2), and 3 (SCN5A), each of which is characterized by a specific mutation that can be detected on gene testing.^1–3 There is no cure for this disease. Risk of death can be reduced by β-blockers, exercise restriction, and the insertion of an ICD to avoid syncope and/or death if the patient is at risk for VT.

Emily had a family history of long QT syndrome, electrocardiogram with features of long QT syndrome, and an episode of syncope with exercise. These clinical findings are concerning for VT during exercise leading to low cardiac output and syncope. She then also had electrical findings in the electrophysiology laboratory at a reputable and experienced institution, as well as subsequent gene testing confirming long QT syndrome type 3. Patients with long QT syndrome type 3 have been reported to have up to a 10% to 20% risk of cardiac events, including an ∼6% risk of sudden cardiac death, and up to a 5% risk of aborted cardiac arrest in various registry studies. Women have more frequent cardiac events (syncope, cardiac arrest) than do men. Cardiac events are rare before age 10 years, with the incidence of new cardiac events increasing during adolescence, and new events are rare after 40 years of age. In addition, patients who have already experienced 1 syncopal or other cardiac event are at higher risk for further cardiac events.³ Hence, our patient met the clinical indications for ICD implantation.^1–3

ICDs have been shown to be safe and effective in treating potentially lethal arrhythmia syndromes.⁴ However, ICD implantation in children and adults is also known to be associated with a higher rate of anxiety and/or depressive disorders, as well as lower physical quality of life scores.^5–7 In the pediatric literature, there is no association of these psychological changes with appropriate or inappropriate shocks or with frequent shocks. However, adults with ICDs will decrease their activity level after a shock even if the shock is determined to have been inappropriate and unnecessary. This decrease in activity persists for months after the shock.⁷

There is also a known significant reintervention rate (including lead adjustments or replacements) in patients undergoing ICD implantation, with reported freedom from reintervention of only 74% at 3 years and 55% at 5 years after the implant.⁸ Some studies suggest that anxiety disorders and depression are more frequent in patients who are younger and female. Such disorders also may be more common in people with underlying anxiety problems or an anxious and/or overprotective family.⁹

Inappropriate shocks are common, especially in children with ICDs.^10,11 Costa et al⁶ reported that 11 of 29 patients in their study received 152 inappropriate shocks in a mean follow-up of 2.6 years (±1.2 years). Goldstein et al¹² reported an overall frequency of inappropriate shocks of ∼15% in a study of 208 devices, whereas in a study of 23 patients in the Netherlands, 7 had received an inappropriate shock in the first 7 months after the device was implanted.¹¹ Patients are counseled extensively before an ICD implantation about the risk of inappropriate shocks and potential psychological sequelae. In addition, new-generation ICDs have built-in algorithms used to decrease inappropriate discharges with a lower incidence of inappropriate shocks of ∼0.4 per 100 patient ICD years.¹³ Also, progressive programming, with higher rate limits, delayed response, and T-wave sensing optimization, as well as modifications like ensuring no fractured leads and optimal lead positioning, have been shown to decrease the overall number of shocks as well as the number of inappropriate shocks, with no change or a reduction in all-cause mortality.^14,15 In our patient’s case as well, these device setting adjustments were made several times, and lead fracture was ruled out.

Considering all of these factors, ICD implantation is always a complex decision. The decision is only made after an intense process of counseling and discussion with the patients and family members. In Emily’s case, her anxiety disorder weighed against continuing ICD therapy, whereas her history of a known episode of syncope, attributable to the arrhythmia, weighed heavily in favor of continuing the use of the device.

Emily was mature and appropriate during her interactions in clinic. Rejecting her request outright purely on the basis of her age (2 months from the age of majority) seemed arbitrary. In addition, as her electrophysiology team, it is our duty to treat her with therapies that have the best chance of efficacy and compliance. She displays psychological symptoms that are known to occur in patients with ICDs, especially those that receive shocks, including clinical depression and self-harm tendencies. She is also at a higher risk for recurring inappropriate shocks on the basis of previous literature. We tried to address all the mechanical problems of the lead and also adjusted the device to more progressive settings without relief.

In addition, Emily is moving to college. She will need to transition her care to other providers. Like all young adults with complex health problems, she is at risk for being lost to follow-up and of not getting skilled and appropriate cardiac care. We wanted to preserve a trusting therapeutic relationship so that she would feel free to call us with concerns or questions in the future. Under those circumstances, she would have an additional risk factor for noncompliance with her medical care and follow-up if she felt that her doctors did not hear her concerns and only followed her mother’s desires. In light of these multiple factors, we felt that the potential risks of refusing her request outweighed those of respecting it.

Dr Janet I. Malek and Dr Ernest Frugé

This case takes place at the intersection of 2 individually complex ethical debates: 1 is about adolescent decision-making and another is about the moral relevance of the distinction between withholding and withdrawing an intervention.

Emily is 17 years old, only months away from reaching the age of majority in the United States. As a result, her mother is legally authorized to make medical decisions on her behalf. However, the American Academy of Pediatrics recommends that pediatricians seek true informed consent from adolescent patients,¹⁶ including respecting their refusals of treatment. This is therefore a case in which there is tension between legal and ethical guidelines.

There are good reasons to put substantial emphasis on Emily’s perspective in this case. First, she is nearing 18, the age at which people are presumed to have sufficient decision-making ability to make acceptable choices. Second, Emily has been managing her condition for 7 years. She has demonstrated a good understanding of the disease, its prognosis, and the burdens of treatment. It is suggested in limited available data that children with chronic conditions are able to reason in a more nuanced way about difficult medical decisions than those without chronic conditions,¹⁷ supporting the claim that her request may be autonomous. A final indication of Emily’s capacity to make this decision is that she explored available alternatives, including adjusting the device and seeking strategies for managing her reaction to the unnecessary shocks. Only when these safer approaches failed to address her concerns did she request more drastic measures. On the basis of all these factors, the ethics consultants concluded that Emily’s autonomy should weigh heavily in the development of a management plan.

Clinicians may be reluctant to follow ethical guidelines when they conflict with legal ones. That could be especially true in this case given that there are no clear precedents with respect to a physician’s legal liability for following the wishes of a minor over the objections of her legally authorized decision-maker.^18,19 This concern could be mitigated by obtaining court approval before removing the device, but this would be a lengthy and expensive process.

Emily’s case, like many cases involving ICDs, raises the question of whether there is a morally relevant difference between withholding and withdrawing medical intervention.^20–22 Generally, bioethicists and lawyers agree that there is no difference. By this argument, we could conclude that if Emily could refuse to have an ICD implanted in the first place, she should be free to discontinue the use of an ICD that has already been implanted. This line of reasoning has been used to ethically justify stopping mechanical ventilation, renal dialysis, or vasopressors. The argument for this position is built on the premise that a patient should have the right to autonomously refuse medical care regardless of whether that refusal concerns the introduction of a new intervention or the continuation of an existing intervention.

The clinicians in this case had doubts about whether to respect Emily’s request to turn off her ICD. The ethics consultants suggested that they think about whether they would have gone to court to force Emily to have a device implanted if she had not had an ICD before and did not want one. If the team would respect Emily’s wish to withhold placement of a new ICD (which seems likely), they should also respect her desire to withdraw that treatment on the same ethical grounds. This casuistic analysis therefore offers additional evidence that Emily’s request should be granted.

Some commentators have explored the possibility that implanted devices belong in a different moral category and therefore ethically cannot be withdrawn because they become part of the person’s body or a piece of the person’s “self” after implantation.^21,22 With this view in mind, discontinuing an ICD would be equivalent to removing a functioning transplanted lung or artificial heart valve and would therefore be killing a patient rather than allowing the patient to die of an underlying condition. The ethics consultants noted that the fact that Emily is not dependent on her ICD for sustaining her heart rate suggests that it is not part of the patient’s body in this robust sense. As a result, removing it would be more analogous to discontinuing ventilator support or a feeding tube and so falls within the patient’s purview to decide.

When these considerations are put together with the relative risks and benefits associated with this intervention, a strong ethical case emerges in support of turning off Emily’s ICD in accordance with her wishes over her mother’s objections. There is a possibility that Emily will suffer serious injury or death after deactivating her ICD, but the desire to prevent this tragic outcome should not blind the team to the fact that they have limited evidence that such an injury or death will actually come to pass or blind them to the harm that she is currently experiencing because of unnecessary shocks. Emily’s level of decision-making capacity appears sufficient for making this decision.

Despite initial uncertainty about how to respond, close ethical analysis offers substantial justification for respecting Emily’s decision to turn off the ICD in this particular case. On the other hand, if Emily were younger, if she had demonstrated less sophisticated reasoning about her situation, and if the team had more evidence that turning off the device would likely result in serious injury or death, the ethical analysis might shift to offer stronger support for declining her request for deactivation. Nuanced understanding of the facts of the case is vital for making such challenging judgments.