Introduction

The Herlitz subtype of junctional epidermolysis bullosa (JEB-H) is a lethal genetic disorder characterized by recurrent and often persistent erosions of the epithelial surfaces. It is often complicated by sepsis and has a high rate of mortality in infancy. Respiratory compromise, failure to thrive and healing with exuberant granulation tissue are additional features of this rare disease. There is currently no cure for JEB-H, and mortality in the first year of life approaches 90%.1 Both medical specialists and parents of affected children struggle to find a balance between the expected benefits of medical intervention and the pain and distress the child endures. Some desire to limit interventions; others push to continue aggressive interventions to prolong life.

Appraising medical treatments for children with fatal medical conditions is a difficult and uncomfortable subject. Lethal abnormalities in neonates include a wide variety of conditions such as anencephaly, thanatophoric dysplasia, bilateral renal agenesis and Trisomy 13, as well as JEB-H. Although there is extensive coverage in the medical literature on many of these conditions, there is, at present, no comprehensive discussion on treatment decisions for infants with JEB-H. This article provides an overview of JEB-H and examines the ethical norms for treatment of neonates with end-stage diseases. These principles are applied in the context of JEB-H to help guide decision-making for patients with this genetic disorder.

An overview of JEB

Epidermolysis bullosa (EB) is a family of rare, inherited disorders characterized by fragility of the skin and mucous membranes in response to minor mechanical trauma. It is classified into three major types based on where blistering arises in the skin in relation to the dermal–epidermal junction, and each EB type can be further divided into subtypes based on inheritance patterns and clinical findings (Table 1).2

Table 1 Classification of EB types and major subtypes
Full size table

Junctional EB (JEB), the rarest type of EB, is an autosomal-recessive disorder that presents at birth. The estimated prevalence in the United States is one per 2.5 million live births.3 The Herlitz and non-Herlitz subtypes of JEB (JEB-H and JEB-nH) are characterized by generalized skin and mucosal (ocular, oral and gastrointestinal) erosions, nail dystrophy, dental enamel hypoplasia, and a propensity for erosions and scarring of the respiratory epithelium. A spectrum of disease severity is seen in patients surviving beyond the neonatal period. Severely affected individuals often develop significant anemia, failure to thrive and skin healing with ‘exuberant’ granulation tissue. Such patients are clinically classified as having JEB-H and are very likely to die in infancy or early childhood.4 Those with JEB-nH tend to have a milder phenotype and can have normal life expectancy. In the neonatal period, however, JEB-H and JEB-nH may be clinically indistinguishable, and a significant portion of infants with JEB-nH succumb to their disease as well.1 JEB-H is caused by loss of function mutations in the genes encoding the basement membrane protein laminin 5, leading to an absence of laminin 5 in the skin and mucosa.5 JEB-nH is caused by less deleterious mutations in the laminin 5 genes or by mutations in the gene encoding another protein, type 17 collagen.5, 6

JEB-H has also been called JEB-letalis, and as this moniker suggests, it is a lethal disease. According to the National Epidermolysis Bullosa Registry (NEBR), as of 1 December 1995, 9/22 (40.9%) enrolled JEB-H patients had died, with nearly 90% of deaths occurring before 1 year of age; and 51/148 (34.5%) enrolled JEB-nH patients had died.1 Outcome data regarding the surviving patients is currently not available. Registry data may actually underreport mortality as infants with rapid demise may not have been referred for inclusion. The experience of most clinicians suggests that death in infancy or early childhood is the norm for patients with JEB-H.

The reported causes of death of JEB patients enrolled in the NEBR are listed in Table 2.1 Most die of sepsis or respiratory failure. Widespread skin erosions allow for the entry of bacteria into the bloodstream, leading to overwhelming sepsis. Respiratory failure due to airway disease is preceded by signs such as hoarseness and stridor. Blisters, erosions and edema of the respiratory epithelium heal with scarring and stricture formation or exuberant granulation tissue, and acute sloughing of the mucosa or granulation tissue can lead to sudden respiratory obstruction. A tracheostomy can bypass progressive strictures of the airway, but even with a tracheostomy, distal regions of the airway can become occluded, leading to respiratory failure.7

Table 2 Causes of death among JEB patients enrolled in the NEBR1
Full size table

JEB patients likely live lives filled with pain and distress. The tendency to develop blisters and erosions is lifelong, and these defects of the skin and mucosa are painful. Even normal activities, such as feeding and defecating may be excruciating unless the infant is provided analgesia. As there is no cure for JEB, the goals of care are to optimize comfort and minimize the likelihood of complications. Such care consists of limiting trauma to the skin and mucosa, meticulously dressing the skin with nonadherent bandages, treating pain and distress with analgesics and sedatives if necessary, and close follow-up to monitor for complications such as anemia, failure to thrive or respiratory compromise. Despite optimal palliative care, it is impossible to prevent new blisters and erosions from forming. Many JEB patients live with open erosions over 10 to 50% of their body surface area.

Diagnosing JEB

JEB presents at or shortly after birth with blisters and erosions of the skin and mucosa. During the neonatal period, a diagnosis of JEB cannot be made on the basis of clinical findings alone, as all types of EB can have similar phenotypic findings in this time frame. To properly counsel the family of a neonate with suspected EB, it is important to make an accurate diagnosis of the EB type and subtype as quickly as possible. The most accurate and rapid way to do so is to carry out skin biopsies for immunomapping studies using EB-specific antibodies, with transmission electron microscopy (TEM) as a back-up study if needed.2

Skin biopsies must be performed on freshly induced blisters, as re-epithelialization of intact blisters can lead to false results. It is also imperative that biopsies are sent to a laboratory with expertise in diagnosing EB. Immunomapping for EB relies on the use of fluorescent-tagged antibodies specific for the structural proteins that make up the basement membrane zone. Strong staining with an antibody signifies the presence of the corresponding protein, the expected result in normal skin. In the case of EB, however, absent or attenuated staining for certain proteins is diagnostic for the corresponding EB subtype.2, 8, 9 Immunomapping studies for JEB owing to laminin 5 mutations show diminished to absent staining with antibodies to laminin 5.10, 11 In those cases where immunomapping is not conclusive, TEM studies may help narrow the diagnosis. TEM studies for JEB show absent anchoring filaments and epidermal-dermal separation through the lamina lucida.2

Genetic analysis for mutations in the laminin 5 genes is commercially available through the laboratory GeneDx (http://www.genedx.com). This analysis takes 5 to 13 weeks to complete and is 40 to 90% sensitive, depending of the extent of analysis done. Finding a genetic mutation confirms the clinical and pathologic diagnosis. Such information may be reassuring for families and can be used to guide counseling for future pregnancies where the risk for recurrence is 25%. Owing to the lengthy turnaround time and potential for false-negative results, however, genetic analysis should not be relied on to guide medical decision-making. If the results of mutation analysis are available, they should be considered as another piece of evidence, but should not trump the child's clinical condition in the decision-making process.

Ethical principles

From the time of Hippocrates, it has been understood that the physician's first duty is to act for the patient's well-being. With that imperative in mind, what should be the appropriate care of the infant with JEB-H? Given the marked fragility of the skin, the prospects for the patient to thrive or even survive are extremely poor. Treatments that substantially ameliorate the natural course of disease are not available. Nor is there any mechanism for limiting mortality. Most children affected with the disease will succumb to overwhelming sepsis or respiratory failure. These, it must be understood, are not comorbidities or isolated events, but anticipated symptoms or complications of the disease itself.

When a child becomes unstable with these symptoms, medical interventions can, at best, temporarily delay death. Although medical treatments might help in the management of secondary complications, they do not address the underlying disease nor can they prevent the symptoms from reappearing. Inevitably, and generally quite quickly, the infant will succumb to one of these complications.

In an insightful commentary on the limits of transplant surgery, Francis Moore condemned ‘desperate measures for desperately ill patients desperately hopeless from the outset.’12 As he put it, ‘There must be a rationale on which the desperately ill patient may be offered not merely pain, suffering and cost but also the true hope of prolonged survival [without devastating sequalae].’ These insights from surgery apply as well to perinatology where the patient is necessarily dependent on others to act on his or her behalf.

Here, Moore was echoing Paul Ramsey's insight that imposing treatment on the dying patient was not a benefit but an abuse.13 For Ramsey, the appropriate care of the dying patient is not ‘pretended remedies’ but comfort and company. Ramsey's conclusions force us to examine the distinctions in medical ethics on the appropriateness of utilizing medical intervention. Those distinctions, whose origins can be traced back to the Hippocratic corpus, serve as an antidote to the notion that the physician's duty is to do everything possible to prolong life. In the treatise entitled The Art, Hippocrates defines medicine as having three roles: ‘To do away with the sufferings of the sick, to lessen the violence of their disease, and to refuse to treat those who are overmastered by their diseases, realizing that in such cases medicine is powerless.’14

The best contemporary restatement of the standards for appropriate care of the sick is found in the President's Commission report, ‘Deciding to Forego Life-Sustaining Treatment.’15 There, the Commission notes that decision-making for infants should focus on the ‘best interests’ of the child. These it defines as ‘relief of suffering, the preservation or restoration of functioning; the quality as well as the extent of life sustained.’

As for the JEB-H patient there is no realistic possibility of reversing the disease process or substantially improving the quality of the patient's life, we are forced to confront the question Richard McCormick asked about dying infants, ‘Granted we can extend the life, what kind of a life are we prolonging?’.16 Is a life suffused in pain, prone to repeated complications and destined for an early demise one that must be endured for as long as possible?

The consensus in the ethical literature is unanimous: the patient – competent or incompetent – has no obligation to undergo what the President's Commission labeled ‘disproportionately burdensome treatments.’15 If in the judgment of the patient, or in the case of an infant of the parents, a proposed treatment is overly burdensome or not of sufficient benefit, it may be forgone or withdrawn.

Caring for the child with a lethal condition: JEB-H as a paradigm

With this background on treatment for patients diagnosed with lethal conditions, we turn to a practical discussion of care for infants with JEB-H. Although many of the points below apply specifically to EB, the broader principles discussed are relevant to the care of any neonate with a lethal condition, such as anencephaly, thanatophoric dysplasia, bilateral renal agenesis and Trisomy 13.

Broadly defined, care can be divided into two types: ‘comfort care’ and ‘medical treatments.’ Comfort care includes oral feedings, dressing changes, and analgesics and/or sedatives to control pain. Medical treatments are second-line interventions and are ordered in a hierarchy from least to most invasive: oral antibiotics, intravenous antibiotics, feeding tubes, cardiac pressors, central line monitoring and intubation. All infants with EB should be provided with comfort care irrespective of their stage in life or form of EB. The use of medical interventions should be reserved for patients who may be stabilized and whose disease process may improve or be ameliorated. With JEB-H, there is at present no prospect for survival or reversal of the disease.

In the normal course of caring for children with JEB-H, at least two situations arise where medical interventions may be considered. First, newborns with suspected but unconfirmed JEB-H need to be stabilized until a diagnosis is made. If the patient is found to have a form of EB other than JEB-H, the possibility for survival makes continued medical treatment appropriate. Patients diagnosed with JEB-H, however, are likely not to survive, even with aggressive therapies. Parents should be informed of the diagnosis and its status as a lethal disease. Given the absence of any treatment for JEB-H, the appropriate response to the infant's condition is comfort care through the dying process. If the child was initially intubated, a discussion should be had with the parents about discontinuing mechanical ventilation. Many parents find the process of withdrawing supportive therapies troublesome, believing it will cause their child's death. They should be assured that the ultimate cause of death is the underlying disease rather than the removal of a ventilator.

In another scenario, some parents may not yet be convinced that their child is actually dying. JEB-H patients can have limited skin involvement and appropriate growth in early life. They may not have signs of precipitous decline until they are several months old. Under these circumstances, stepwise interventions may be used to treat complications, starting with the least invasive intervention. Before treatment is escalated, each intervention in the hierarchy must be evaluated based on the child's condition and accepted or rejected as appropriate. Regardless of the child's age, once the diagnosis of JEB-H has been confirmed, aggressive interventions ought not be proposed as a therapy.17

With appropriate guidance from the medical team, it is hoped that parents will come to accept the devastating reality of their infant's condition. Given the fact that medicine, at present, is powerless to reverse or ameliorate the disease, the most that can be done for an infant with JEB-H is to keep the child comfortable and support the parents through the infant's dying process. To respond with increasing technological complexity to symptoms, complications and other manifestations of the disease would impose additional pain and distress out of proportion to any anticipated benefit to the patient. Comfort and company, including analgesics and possibly sedatives, should be the mainstay of care for JEB-H.

Recommendations for treating infants and children with JEB-H

A proposed algorithm for withholding or withdrawing care from children with JEB-H is outlined in Figure 1. The use of this algorithm is intended to distinguish the care of JEB-H as a lethal disorder from that of nonlethal forms of EB.

Figure 1
figure 1

An algorithm to guide decision-making in withholding or withdrawing care from children with JEB-H. *The child's condition should be factored into recommendations to limit medical treatments; recommendations need to be re-evaluated and modified with changes in the child's condition.

Full size image