Article Text

Download PDFPDF

Loss to follow-up of adults with repaired congenital heart disease
  1. Christopher Wren,
  2. John J O'Sullivan
  1. Department of Congenital Cardiology, Freeman Hospital, Newcastle upon Tyne, UK
  1. Correspondence to Dr Christopher Wren, Department of Congenital Cardiology, Freeman Hospital, Newcastle upon Tyne, NE7 7DN, UK; christopher.wren{at}

Statistics from

The development of cardiopulmonary bypass 60 years ago was spurred on by the need for surgical repair of congenital heart disease.1 Early operations concentrated on common malformations that were mostly repaired after infancy. Most patients undergoing surgery were children and were cared for by paediatric cardiologists. When they became adults there was, for many years, no procedure for continuing long-term care, and only more recently has adult congenital cardiology become a well-defined subspeciality.

Surgical mortality in the early days was high but since the 1960s most patients have survived long term. It is estimated that survival to adulthood now exceeds 90%,2 and by the year 2000 the number of adults with cardiovascular malformations exceeded the number of children.3

Although repair of some malformations (such as ventricular septal defect, atrial septal defect and persistent ductus arteriosus) is mostly curative, so-called ‘complete repair’ or ‘total correction’ of many others is not. These include aortic valve stenosis, pulmonary valve stenosis, coarctation of the aorta, tetralogy of Fallot and transposition of the great arteries, as well as more complex malformations. It is now recognised that patients have many late problems that require assessment and treatment.4 Because of this one would expect the long-term outcome to be better in patients under the care of a specialist, and failure to continue follow-up to lead to a poorer outcome.5

Mackie et al5 investigated a cohort of children born in Quebec in 1983 and diagnosed under the age of 6 years who were still alive at the age of 25 years.6 Twenty-eight per cent were lost to follow up by 6–12 years, another 19% by 13–17 years and another 15% by 18–22 years, so that by adult life only 39% of the total cohort was still under specialist review. Many had relatively less severe malformations, but even among those with the most severe malformations only 79% were under review by adult life, despite the fact that they mostly maintained contact with primary care physicians. Because of the study design the authors were unable to assess the impact on outcome of loss to follow-up.

By contrast, Goossens et al7 found that only 10% of Belgian children born in 1984–8 were not receiving appropriate follow-up after leaving the paediatric cardiology clinic. Van Deyk et al8 found that 78% of Dutch adolescents knew they needed long-term review but only 46% knew why.

Wray et al9 report their investigation of the continuation of specialist follow-up in 893 survivors of repair of tetralogy of Fallot in Great Ormond Street Hospital since the first operation in 1964. They found that 24% of their patients were lost to follow-up and 48% of late deaths were in this group. Those lost to follow-up were mostly adults but included some children. Attempts were made to contact them, but because of constraints imposed by the ethics committee, a poor response from general practitioners, and a low response rate from patients, only 19% were successfully contacted. The advantage of using tetralogy of Fallot as a model includes the fact that it is relatively common, diagnosis is consistent, and it has been repaired since the early days of cardiac surgery. The potential benefits of long-term specialist follow-up include the earlier detection of arrhythmias, heart failure and pulmonary regurgitation all of which will require treatment.

In the Great Ormond Street Hospital cohort two-thirds of patients under review received a pulmonary valve replacement by the age of 50 years, whereas obviously none of those lost to follow-up did so. It cannot be proved that the excess mortality in patients not under review was due to the lack of specialist care but this is the logical conclusion, although there is known to be an increase in non-cardiac mortality in adult congenital heart patients.10

Moons et al2 investigated survival to adulthood in patients with congenital heart disease in Belgium, and showed that there has been a constant improvement in survival over time and a fall in late mortality, mainly reflecting the benefit of continued expert surveillance for the recognition and treatment of late complications. They failed to find any difference in survival for different birth eras after the age of 18 years, although this may partly reflect the history of the provision of cardiac surgery in their centre.

In a study based on the whole population of Finland, Nieminen et al10 found no real improvement in long-term outcome with approximately 75% survival in patients aged 25–30 years after repair of tetralogy of Fallot in the 1950s, 1960s and 1970s. Only 20 years follow-up was available for the 1980s cohort at the time of publication of their report but again there was no evidence of an improvement.

Pillutla et al11 reached similar conclusions in a report from the USA, which showed no change in mortality in adults with repaired tetralogy of Fallot in 1979–2005, with the leading causes of death being arrhythmias and heart failure.

Repair of coarctation was first reported almost 70 years ago so there is now a large cohort of patients who have had surgery, most of whom are adults. Because aortic coarctation appears superficially to be a relatively simple problem it is probable that more patients with repaired coarctation have been lost to follow-up. In fact, coarctation is associated with widespread vascular dysfunction, and late problems (including hypertension, aortic restenosis, aortic and mitral valve disease, left ventricular failure, coronary artery disease, stroke and sudden death) are more frequent than in patients with tetralogy of Fallot. Hager et al12 investigated 273 patents aged 16–73 years who had undergone repair of coarctation 1–27 years previously in Munich. More than half had hypertension, mostly without evidence of restenosis. Høimyr et al13 reported 229 patients operated on for coarctation in 1965–85 in Aarhus, Denmark, and followed for up to 40 years. They found a mortality rate ratio of 4.3 compared with a normal population and half had hypertension or vascular morbidity. The freedom from death, reoperation or serious complications (not including hypertension), was 60% at 30 years. Half of their patients had hypertension or vascular morbidity.

The main marker of success of intervention for congenital cardiac defects in childhood has been survival. It is now increasingly recognised that we also need to look at other long-term outcomes, including neurological sequelae and quality of life. This fundamentally important information that patients and parents seek is not available for most congenital cardiovascular malformations, and one of the main reasons for this is evident from the paper by Wray et al.9 The limited follow-up into adulthood of patients who have had surgery for a major cardiac malformation at a major institution may suggest that the problem is similar or worse at other institutions and for other important malformations. We will not achieve the goal of providing meaningful long-term outcome data until this problem with follow-up is resolved.

Transition from paediatric to adult services is a process that has received a lot of attention recently. The proportion of patients who remain under specialist follow-up will give a good indication of the adequacy of the transition process, and audit of the process could be a useful indicator of institutional performance. Even with careful counselling some patients in teenage years may reject the offer of long-term review, and the process is made more difficult by leaving home, moving address, gaining employment, etc. There is an obvious need for a database that can alert carers when patients are not under review and can accommodate information from all the specialist congenital centres in the UK.

The main point of the paper by Wray et al9 is to show that there are patients who are not under review who should be, and that there is a continuing loss of patients from the paediatric to the adult clinics. Initiatives to locate these patients rather than wait for an urgent admission14 or chronic deterioration would have resource implications. This paper suggests there are some patients who could be retrieved relatively easily, and referral of patients back from general cardiology clinics should be encouraged. The longer term issue with engaging with follow-up is not unique to congenital heart disease, and the solution to this problem will include a restructuring of the way patients are transitioned and monitored when leaving the paediatric clinic. However, the main potential benefit to the cardiology community is that we may eventually be in a position to provide data on long-term outcome for various conditions and treatment strategies used in childhood.

The information that young patients and their families seek as they transition to the adult service centres mainly on their likely future quality of life, education and employment potential. This requires follow-up of complete cohorts and the unavailability of this type of information is likely to undermine confidence in, and compliance with, long-term medical follow-up.


View Abstract


  • Funding None.

  • Competing interests None.

  • Provenance and peer review Commissioned; internally peer reviewed.

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

Linked Articles