Percutaneous Interventional Radiology Procedures in Patients With Epidermolysis Bullosa: Modifications and Challenges
Abstract
OBJECTIVE. The purpose of this study is to describe the role of interventional radiology in palliation and maintenance of nutritional support in children with epidermolysis bullosa, with a focus on safety and required procedural modifications.
MATERIALS AND METHODS. This was a retrospective study of all patients diagnosed with epidermolysis bullosa who underwent interventional radiology procedures between January 1991 and December 2008 at a pediatric tertiary care institution. The type of epidermolysis bullosa, the patient's age, the indication and type of interventional radiology procedure, modifications used, and complications were recorded.
RESULTS. Fifteen patients (9 boys and 6 girls) with a diagnosis of epidermolysis bullosa underwent a total of 87 procedures during 82 different patient interventional radiology visits. Twenty-seven esophageal dilatations for symptoms of dysphagia, 11 percutaneous gastrostomy tube placements resulting from failure to thrive, 30 maintenance procedures, 15 central venous access device insertions, and three other procedures (nasojejunal tube insertion, liver biopsy, and inferior vena cava filter insertion) were performed. Modifications were related to anesthetic management, access, tissue handling, and dressings. Complications were categorized according to Society of Interventional Radiology grades: Minor A (n = 6), Minor B (n = 4), Major C (n = 7), Major D (n = 2), Major E (n = 0), and Major F (n = 0).
CONCLUSION. Our experience suggests that interventional radiology procedures can be done successfully and safely when utmost care is given to skin and mucosal protection. Wound healing is adequate and should not deter intervention.
Introduction
Epidermolysis bullosa is an inherited multisystemic disorder characterized by skin and mucous membrane fragility. It is seen in approximately one to three per 100,000 live births [1]. There are three major inherited types (epidermolysis bullosa simplex, junctional epidermolysis bullosa, and dystrophic epidermolysis bullosa) and one acquired autoimmune type (epidermolysis bullosa aquisita) [2–5], each with a different clinical course and prognosis.
Skin blistering and scarring are the major problems associated with epidermolysis bullosa, and they occur anywhere over the entire body secondary to minor trauma, friction, or shearing stress. As a result, many children with epidermolysis bullosa are extensively bandaged from head to toe and suffer repeated cycles of blistering, infection, healing, and scarring, ultimately leading to contractures, digital fusions, and amputations. Similarly, repeated trauma resulting from swallowing causes blistering and scar formation in the upper esophagus and oropharynx, leading to strictures [1, 5, 6]. Chronic wounds and recurrent skin infections increase the nutritional demands of these patients. The long-standing cycle of blisters, infection, healing, high nutritional demands, and secondary comorbidities lead to a reduced quality of life for patients with epidermolysis bullosa. In addition, the many hours spent daily changing dressings over extensive areas of the body incur significant economic and emotional costs to the patient, family, and care providers.
There is no cure for epidermolysis bullosa. Treatment is supportive, and management requires a multidisciplinary approach directed toward the treatment of symptoms, prevention of infections, pain management, protection of the skin against trauma, correction of nutritional deficiencies, minimization of deformities and contractures, and psychologic support for the patient and the entire family [5, 7]. At our institution, a multidisciplinary epidermolysis bullosa team includes specialists from the departments of dermatology, gastroenterology, interventional radiology, anesthesiology, plastic surgery, cardiology, hematology, and general pediatrics. Although most of the interventional radiology procedures requested for patients with epidermolysis bullosa are not technically complex, they are challenging in this population because of the lack of intact skin, the susceptibility of the skin and mucosa to severe blistering from minimal friction or trauma, the associated comorbidities, and the added risk associated with the use of general anesthesia. A request for an interventional procedure on a child with epidermolysis bullosa may therefore be daunting for the interventional radiologist, who may have limited experience with this condition, is concerned with patient's safety, and may question the wisdom of such an intervention.
The purpose of this article is to describe our 17 years of cumulative experience with interventional procedures for palliation and maintenance of adequate nutritional support for this unique complex group of children, focusing on safety and procedural modifications required.
Materials and Methods
This is a descriptive retrospective study of all patients diagnosed with epidermolysis bullosa who underwent interventional procedures between January 1991 and December 2008 at a tertiary pediatric care institution. The study was approved by the institution's research ethics board, and the requirement for informed consent was waived.
Eligible patients were identified by cross-referencing data from the epidermolysis bullosa clinic records, from radiology information services using an ISYS search engine (Odyssey Development), from the interventional radiology database, and from medical records discharge codes. Records of patients with epidermolysis bullosa were then searched for interventional procedures. Paper charts, radiology reports, and electronic records were reviewed. The type of epidermolysis bullosa, the age at any interventional procedure, the indication and type of procedure, preprocedural preparation and monitoring, procedure modifications, anesthetic-related issues, postprocedure care of the patient, and complications were recorded. Procedural complications included adverse events that occurred during the performance of the procedure itself or in association with administration of anesthesia. Postprocedural complications were those that occurred up to 30 days after the procedure [8, 9]. Complications were then graded according to the Society of Interventional Radiology criteria and were determined to be either minor (grades A and B) or major (grades C, D, E, and F) depending on the management required, length of hospital stay, and outcome [10]. We erred on the side of overestimating the severity of complications and their grade. Routine tube changes were categorized as maintenance procedures and were not counted as complications.
Patient Management
All patients with epidermolysis bullosa underwent a thorough medical and nutritional assessment in the epidermolysis bullosa clinic. The majority of interventional procedures included esophageal dilatations, gastrostomy tube insertion, and vascular access device placement. Patients presenting with dysphagia, nutritional deficiency, and failure to thrive underwent an esophagogram to assess the presence and extent of esophageal strictures. The presence of a clinically symptomatic stricture on an esophagogram was an indication for esophageal dilatation. Children who were unable to sustain adequate nutrition by mouth to meet their high metabolic needs (for a variety of reasons, such as acquired microstomia, odynophagia, ankyloglossia, and esophageal stricture) were potential candidates for gastrostomy tube insertion. Children needing total parenteral nutrition, administration of fluids, and multiple transfusions or treatments were potential candidates for vascular access.
Procedural Techniques
In children with epidermolysis bullosa, we use our standard techniques as outlined in the following subsections, as well as some specific modifications required for epidermolysis bullosa (Appendix 1). Informed consent includes descriptions of the recognized complications of the intended procedure, along with the added issues of skin blistering, site infections, and dislodgement of devices (e.g., central lines or gastrostomy tubes) because of the inability to use standard securing methods (e.g., adhesive tape). In addition, the anesthetic risks are outlined in detail.
Interventional Radiology Procedures
Esophageal dilatations—The standard technique for esophageal dilatation includes insertion of a directional catheter through the nose or mouth over a floppy-tipped wire (Benston, Cook), injection of water-soluble contrast material to outline the stricture, subsequent passage of a balloon catheter over the wire through the stricture, manual inflation of the balloon for 30 seconds (one to three times), and a check for leakage of contrast material or tracheal aspiration. We usually use hand inflation of the balloon using a large 20-mL syringe (low pressure). At the end of the procedure, as much contrast material as possible is aspirated from the upper esophageal pouch. The patients are observed for 6 hours before discharge [11].
For patients with epidermolysis bullosa, modifications include protection of the mouth and skin of the face from any handling (using petroleum jelly [Vaseline, Tyco Health Care Group] or other lubricant); use of a no-touch technique as much as possible; minimizing wire or catheter handling in the mouth, hypopharynx, and esophagus; significant lubrication on all material inserted per os; and postprocedure overnight (24 hours) observation before discharge. Whenever possible, esophageal dilatations are performed through a gastrostomy stoma, with protection of the skin surrounding the stoma with a barrier agent (e.g., petroleum jelly) and avoidance of any friction or rubbing of the skin. The diagnosis of epidermolysis bullosa does not alter the choice of diameter of balloon, which is based on the size of a healthy esophagus.
Gastrostomy tube insertion—Our standard percutaneous imaging-guided retrograde technique for gastrostomy tube includes the administration of a single dose of antibiotic prophylaxis (a first-generation cephalosporin, such as cefazolin [30 mg/kg]) before the procedure and the insertion of a nasogastric tube for stomach inflation during the procedure and for gastric decompression after the procedure for the first 12–24 hours. We perform a single stomach-wall puncture with an 18-gauge one-part thin-walled needle (BSDN 18-4, Cook) preloaded with a pediatric retention anchor suture (Cope Gastrointestinal Suture Anchor Set, GIAS-100-Chait, Cook), insert a 0.035 inch (0.09 cm) straight wire deploying the retention suture into the stomach, dilate the fascia over the wire, and insert a pigtail catheter (DawsonMueller MAC LOC, Cook). The retention suture thread is rolled around a piece of gauze applying some tension to appose the gastric and abdominal walls. Polysporin ointment (Pfizer Canada) is applied to the site to prevent infection for the first few days [8, 9, 11]. The tube is secured using adhesive tape (Hypafix, BSN Medical) and the Statlok device provided with the tube (Venetec International). Additional triple antibiotics (ampicillin, gentamycin, and metronidazol) are given only if there is evidence postprocedure of peritonitis.
Modifications for patients with epidermolysis bullosa include the assessment of skin integrity before the procedure to ensure a potential site for tube placement. The risk of skin blistering is explained during the consent process. It is important to avoid any friction on the skin by using a no-touch technique as much as possible. Skin preparation requires the use of gentle dabs of chlorhexidine and avoiding rubbing the skin. Extra lubrication of the nasogastric tube before insertion is required, and securing the nasogastric tube can be achieved by using a gauze ribbon around the patient's head; in the last few years, a safe self-adherent tape (Mepitac, Molnlycke HealthCare) has also become available. Great care is required to protect the skin around the gastrostomy site during fascial dilatation so as to reduce trauma to the epidermis. As with our standard practice, we use only one retention anchor suture in patients with epidermolysis bullosa, because the deeper layers of the abdominal wall are not involved with epidermolysis bullosa, and we use gentle tension only on the thread of the retention suture. There is no change to the antibiotic practice in children with epidermolysis bullosa. Nonadhesive dressing with Telfor (Tyco Health Care Group) is applied to the gastrostomy site, and an elastized tubular netting (Burnet Medical) or a self-adhesive dressing (Mepilex Lite, Molnlycke Healthcare) is applied to the abdomen.
Vascular access—Our standard procedure uses sterile preparation with chlorhexidene, a Seldinger technique for peripherally inserted central catheters (PICCs) or central venous line (CVL) insertions, with distention of the vein with either a tourniquet (PICC) or the Trendelenburg position (CVL), puncture of the vein under real-time ultrasound guidance, local anesthesia, insertion of a cuffed CVL or PICC, and application of a clear semiporous adhesive dressing (e.g., Tagaderm, 3M Health Care) [12].
For patients with epidermolysis bullosa, modifications include an assessment of potential uninvolved noninfected skin sites. During the consent process, the risks of blistering, line infection, and device dislodgement are stressed. Skin preparation is performed with gentle dabs of chlorhexidine without rubbing of the skin and avoidance of a tourniquet, if possible. If venous distention is required, the use of a padded blood pressure cuff as an inflation tourniquet may be less traumatic than a tied or tight tourniquet to achieve venous distention. Minimal or no handling of the skin and gentle manipulation of the arm (no shear forces) are important. The vascular access device cannot be secured with standard semiporous adhesive dressing; instead, sutures, gauze roll, and self-adherent gauze wrap (Coban, 3M Health Care) or self-adherent tape is used.
Perioperative Care and Anesthesia
Sedation for interventional procedures at our institution depends on the child's clinical status and the procedure being contemplated. The spectrum includes full general anesthetic with paralysis and intubation, IV medications (propofol, ketamine, or fentanyl), inhalational agents, nurse-administered sedation, and local anesthetic only. An anesthesiologist is usually present for high-risk procedures or high-risk patients, even if anesthesia is only local and the anesthesiologist is “on standby.” Although preprocedure assessment of anemia secondary to malnutrition is ideal, obtaining blood work for these children may be extremely difficult because of scarring of the skin and, as a result, blood work is often omitted for minor procedures. Coagulation studies are not routinely performed, unless attempting a procedure for which it is required (e.g., liver biopsy). A minimalist approach is usually adopted for anesthetic management for patients with epidermolysis bullosa undergoing an interventional procedure. Patients are transferred as little and as carefully as possible, from one bed to the interventional table and on soft padded surface, such as foam or mattresses (e.g., egg crate mattress). All pressure points (e.g., elbows, heels, and occiput) are padded with self-adhering foam (Reston, 3M Health Care) without using the adhesive side or soft abdominal pads or are placed on soft sponges or gel padding. If mask ventilation is necessary, a padded face mask with the contact areas coated with petroleum jelly is used. Whenever possible, no adhesive tapes are applied to the patients. The adhesive parts of the ECG and cautery pads are cut off, and the pads are kept in place using self-adhesive dressing or elastized tubular netting with petroleum jelly placed under the hub. Pulse oximeters of the clip-on type are used with petroleum jelly and padding applied under the clip. If clip-on oximeters are not adequate, the adhesive type of pulse oximeter is used. Noninvasive blood pressure cuffs, if used, are placed over a layer of soft roll and are cycled manually every 15 minutes. For short-duration procedures, the noninvasive blood pressure cuff is not cycled. Airway management depends on the severity of disease and the nature of procedure and is determined on a case-by-case basis by the anesthesiologist, after discussion with the interventional radiologist. Supplemental oxygen is supplied via nasal prongs or face mask, and a protective barrier of petroleum jelly or gauze is used between the skin and prongs or mask. Intubation is avoided if possible because of the risk of blistering that can occur in the mouth, hypopharynx, or airway from the laryngoscope and endotracheal tube. For patients who definitely require tracheal intubation, fiberoptic placement is the preferred method if the patient has a difficult airway. Otherwise, gentle direct laryngoscopy is used, after applying lidocaine gel or petroleum jelly to the laryngoscope blade. High-volume low-pressure cuffed tracheal tubes are used and are secured with a soft petroleum jelly–coated cloth roll tied to the tube, rather than using adhesive tapes.
Statistics
Descriptive statistics only were required in this review.
Results
Between 1991 and 2008, 55 children were identified as having epidermolysis bullosa. Thirty-two (58%) children had dystrophic epidermolysis bullosa, 12 (22%) had junctional epidermolysis bullosa, 9 (16%) had epidermolysis bullosa simplex, and 2 (4%) had untyped epidermolysis bullosa. Of the 32 children with dystrophic epidermolysis bullosa, 23 (72%) had the recessive dystrophic subtype and 9 (28%) had the dominant dystrophic subtype. Fifteen (27%) of the 55 patients (nine boys and six girls) underwent an interventional procedure (Table 1). Ten (66%) of the 15 patients had recessive dystrophic epidermolysis bullosa, four (27%) had junctional epidermolysis bullosa, and one (7%) had epidermolysis bullosa simplex. A total of 87 interventional procedures were performed during 82 different patient visits to the interventional suites.
| Patient No. | Age | Sex | Epidermolysis Bullosa Type | Types (No.) of Procedures | Complications (No. of Episodes) |
|---|---|---|---|---|---|
| 1 | 3 mos. | Male | Junctional | Gastrostomy tube (n = 1) | Pain |
| Gastrostomy to gastrojejunostomy tube (n = 1) and tube check and change (n = 3) | Gastrostomy tube fell out (n = 2) and gastrostomy tube blockage | ||||
| Peripherally inserted central catheter lines (n = 2) | Dislodgement | ||||
| 2 | 7 y | Male | Recessive dystrophic | Esophageal dilatation (n = 13) | Fever, vomiting, and aspiration (n = 2) |
| Gastrostomy tube (n = 1) | None | ||||
| Tube check and change (n = 1) | None | ||||
| 3 | 12 y | Female | Recessive dystrophic | Esophageal dilatation (n = 4) | None |
| Peripherally inserted central catheter lines (n = 2) | Phlebitis at insertion site | ||||
| 4 | 8 y | Male | Recessive dystrophic | Esophageal dilatation (n = 2) | Self-limiting hemorrhage and aspiration |
| Gastrostomy tube (n = 1) | Pain | ||||
| Tube check and change (n = 4) | Gastrostomy tube blockage (n = 2) and external leakage (n = 2) | ||||
| Peripherally inserted central catheter lines (n = 4) | Arrhythmia, blistering at insertion site, partial migration, and dislodgement (n = 2) | ||||
| Liver biopsy (n = 1) | None | ||||
| Nasojejunal tube | None | ||||
| 5 | 13 y | Male | Recessive dystrophic | Esophageal dilatation (n = 3) | None |
| Gastrostomy tube (n = 1) | None | ||||
| 6 | 10 y | Female | Recessive dystrophic | Gastrostomy tube (n = 1) | Pain |
| Tube check and change (n = 8) | Gastrostomy tube blockage (n = 2) and external leakage | ||||
| 7 | 5 y | Male | Recessive dystrophic | Esophageal dilatation (n = 3) | Self-limiting hemorrhage, aspiration, and severe esophageal stenosis 13 days after procedure |
| Gastrostomy tube (n = 1) | Pain | ||||
| Tube check and change (n = 4) | None | ||||
| 8 | 5 mos | Male | Junctional | Gastrostomy tube (n = 1) | None |
| Tube check and change (n = 1) | Gastrostomy tube external leakage | ||||
| Peripherally inserted central catheter line (n = 1) | None | ||||
| Nasojejunal tube (n = 1) | None | ||||
| 9 | 11 y | Female | Recessive dystrophic | Esophageal dilatation (n = 1) | Submucosal leak |
| Gastrostomy tube (n = 1) | Skin blistering | ||||
| Tube check and change (n = 4) | Gastrostomy tube external leakage (n = 2) and granulation tissue | ||||
| 10 | 11 y | Male | Recessive dystrophic | Esophageal dilatation (n = 1) | None |
| Gastrostomy tube (n = 2) | Gastrostomy tube fell out | ||||
| Tube check and change (n = 2) | Gastrostomy tube blockage | ||||
| 11 | 2 mos | Male | Junctional | Peripherally inserted central catheter line (n = 1) | Dislodgement |
| Central venous line (n = 1) | None | ||||
| 12 | 17 y | Female | Recessive dystrophic | Central venous line (n = 1) | None |
| Inferior vena cava filter (n = 1) | Vena cava thrombosis | ||||
| 13 | 1 mo | Male | Recessive dystrophic | Central venous line (n = 1) | None |
| 14 | 4 y | Female | Simplex | Gastrostomy tube (n = 1) | None |
| Tube check and change (n = 2) | External crack in gastrostomy tube | ||||
| 15 | 2 mos | Female | Junctional | Peripherally inserted central catheter (n = 2) | Skin blistering, dislodgement (n = 2), and swelling of the upper extremity |
Of the 87 interventional procedures, 27 (31%) were esophageal dilatations performed for seven patients (four boys and three girls) for symptoms of dysphagia with proven esophageal strictures on previous esophagograms (Fig. 1A, 1B, 1C). The children ranged in age from 6 to 18 years (mean [± SD] age, 10.7 ± 4 years) at their first dilatation. The strictures ranged in severity from 70% to 90% of the lumen and were at the cervical esophagus (C5–C6) in one patient, at the midesophagus (T4–T5) in two patients, and at multiple levels (cervical, upper, middle, and distal esophagus) in three patients. The strictures seen on barium swallows studies were focal web in two patients, long and smooth in three patients, and a combination of weblike strictures and long smooth strictures at different levels in one patient; for one patient, the images were not available (pre-PACS era). Five of these seven patients required multiple esophageal dilatation procedures (two to 13 dilatations; mean, 3.9 ± 4.2 dilatations). The maximum diameter balloon used at each dilatation procedure ranged from 8 to 20 mm (mean, 13 mm; median, 15 mm). The intervals between dilatations ranged from 3 to 57 months (mean, 16.5 ± 14.5 months). For three patients with a gastrostomy tube in place, the six esophageal dilatations were performed through the gastrostomy stoma.
With respect to complications, tracheal aspiration during the procedure was seen on fluoroscopy in four of 27 procedures, three cases of tracheal aspiration were clinically silent or aymptomatic, and one was clinically significant and required emergent endotracheal intubation (Fig. 2). One patient had a small contained submucosal tear after dilatation that was managed conservatively (Fig. 3). Two of the 27 patients had small self-limited bleeding for 24 hours that did not require transfusion. One patient had transient fever and vomiting after the procedure that did not require treatment; for that patient, a contrast swallow study was negative for a tear. One patient developed acute dysphagia 13 days after an esophageal dilatation for an upper thoracic esophageal lesion performed through his gastrostomy stoma. A contrast swallow study performed while the patient was acutely dysphagic showed an acute hold-up at the level of the cervical esophagus, above the level of the originally treated stricture. This was thought to be most likely related to an acute blister resulting from the presence of the wire, which would have traversed this level. The patient required admission for the acute dysphagia, which resolved spontaneously without intervention but necessitated a prolonged hospital stay (Society of Interventional Radiology grade D).
Eleven percutaneous gastrostomy tube placements were performed because of failure to thrive in 10 patients (six boys and four girls), who were between 3 months and 15.5 years old (mean age, 7.6 ± 5.5 years) at time of gastrostomy insertion. The gastrostomy tube was converted to a gastrojejunostomy tube for one patient. The patients had follow-up ranging from 10 days to 5.5 years (mean, 2.8 ± 2.1 years). Early complications related to gastrostomy tube insertions were immediate skin blistering at the site (one patient) and external leak of fluid or contrast material around the gastrostomy tube through the tract (one patient). The latter patient needed a nasojejunal feeding tube and had a prolonged hospital stay as a result of this leakage, until the tract matured. Early postprocedure pain prompted gastrostomy tube checks in four children (2, 3, 4, and 7 days, respectively, after the procedure) to assess tract integrity. No peritoneal leak or peritonitis was detected. Over the years, 25 maintenance procedures were performed for external tube break ages, dislodgement, external tube leakage, and blockage. Exchange of the catheter or reinsertion through the same tract was successful in all instances, except for one patient for whom a decision was made not to reinsert the tube. Two years later, however, a new gastrostomy tube was requested and inserted in this patient. Minor granulation tissue around the stoma developed in one patient, which was treated using silver nitrate.
Fifteen central venous access devices (12 PICCs [3- and 4-French] and three tunneled CVLs [3-, 4-, and 7-French]) were inserted in eight children (six boys and two girls) ranging in age from 1 month to 17 years (mean, 5.7 ± 6.4 years). The basilic vein was the most commonly used vein in either arm during PICC placement (10 insertions). Thirteen lines were secured using sutures through the full thickness of the skin, and two lines were secured using a nonadhesive skin dressing only. Procedural complications included a prolonged wire-induced supraventricular tachycardia (250 beats per minute), which could only be reversed with adenosine once access was obtained. This patient had severe endstage cardiomyopathy secondary to chronic anemia associated with the underlying epidermolysis bullosa. Immediate skin blistering at the site of PICC insertion was seen in two patients. Early postprocedural dislodgement of the PICC occurred in two patients on days 2 and 15, respectively. In the first patient, the line had been secured using a nonadhesive skin dressing. After dislodgement, a femoral line was placed because of the inability to find a suitable noninfected place on the skin to place another PICC. The second patient had the line secured with a suture. Late postprocedural complications included four accidental dislodgements, one partial line migration, one upper extremity swelling, and one case of phlebitis that resulted in catheter removal.
Three other procedures were performed. One inferior vena cava filter (Gunther Tulip, Cook) was placed for pulmonary embolism prophylaxis in a patient presenting with acute deep vein thrombosis. A subsequent cavogram showed extensive collaterals and persistent entrapped thrombus. Attempts at mechanical thrombolysis (Oassis Hydrolyser, Boston Scientific) were unsuccessful; therefore, the filter was left in situ. In another patient, a nasojejunal tube was placed over a floppy guidewire (0.035 inch [0.09 cm] Bentson wire) to minimize trauma to the pharynx and esophagus. One percutaneous ultrasound-guided liver biopsy was performed to assess iron overload in a child who required ongoing blood transfusions. No procedural complications were noted.
The complications were categorized according to the Society of Interventional Radiology grading system as follows: Minor A (n = 7), Minor B (n = 8), Major C (n = 10), Major D (n = 3), Major E (n = 0), and Major F (n = 0). This calculation yields a minor complication rate of 17.2% (15 of 87 procedures), and a major complication rate of 14.9% (13 of 87 procedures).
Discussion
Epidermolysis bullosa is a multisystem disorder involving the skin and gastrointestinal, respiratory, and urinary tracts [13]. Because there is no cure, clinical measures are directed toward management of symptoms, nutritional support, and skin care [1, 5]. In the interventional radiology literature, there are a few reports concerning patients with epidermolysis bullosa, mostly dealing with technical aspects of gastrointestinal procedures (e.g., esophageal dilatations or gastrostomy insertion) [7, 13–21].
In our study, 27% of patients diagnosed with epidermolysis bullosa needed 87 interventional procedures. It is likely that the role of interventional procedures may increase even more in the future, as clinical supportive measures and minimally invasive techniques evolve, and the prognosis and life expectancy of patients with epidermolysis bullosa improve. The minimally invasive nature of interventional radiology and the use of image guidance increase the accuracy of the procedures, enhance the speed of the procedures, and reduce the time and depth of anesthetic required, thereby minimizing trauma to the patient's skin, oropharynx, and subcutaneous tissues.
The most common interventional procedure undertaken for children with epidermolysis bullosa is esophageal dilatation. The majority of patients with recessive dystrophic epidermolysis bullosa are likely to have esophageal strictures, located mostly in the upper esophagus [1]. The strictures are associated with respiratory problems, such as recurrent aspiration and pneumonia, and gastrointestinal problems, such as dysphagia or phagodynia and inadequate oral intake [5–7, 13]. Esophageal strictures can be managed by surgical, endoscopic, or fluoroscopic approaches. An important part of any treatment for patients with epidermolysis bullosa is the prevention of further epithelial or skin damage, and this influences the choice of treatment approach. Endoscopic dilatations may cause a significant degree of pharyngoesophageal trauma while advancing the endoscope through the pharynx and upper esophagus. Balloon catheters, on the other hand, have the advantage of a much smaller caliber than an endoscope when deflated, yet are capable of achieving a large diameter when inflated [7, 13, 14]. Fluoroscopic dilatation may be less traumatic with generous lubrication of all devices and gentle manipulation of catheters over soft floppy-tipped wires. Access through the gastrostomy stoma further decreases the likelihood of oropharyngeal trauma [21]. When dilatation is performed through the stoma, the skin surrounding the stoma should be protected during the exchange and passage of catheters and tubes. For patients without a gastrostomy tube, gentle access through the oropharynx with protection of the face with petroleum jelly or gel should be performed. Some authors advocate the use of steroids at the time of esophageal dilatation, although we have not done so [14]. Five of seven of our patients needed multiple dilatations. We achieved a mean interval of 16 months between dilatations, which compares favorably with or exceeds the treatment intervals quoted in previous studies [14, 15, 20]. Although we have had recurrent strictures, we have not had resistant strictures. The factors that influence the degree and duration of response are unclear. The appearance on barium swallow of a long smooth stricture frequently did not correlate with the more focal waist seen at the time of balloon dilatation. This most likely is due to an element of under-filling of the esophagus, because the patients frequently cannot initiate a large bolus as a result of their acquired microstomia and ankyloglossia. The only reasonable explanation for the major complication of acute dysphagia and hold-up above the level of the treated stricture is inadvertent trauma due to the wire (floppy-tipped Bentson).
The anesthesiologist usually adopts a minimalist approach during esophageal dilatations. In our study, patients were not intubated. This approach, however, carries the potential risk of aspiration of contrast material from the upper esophagus during the dilatation. One of our patients required emergent intubation during the procedure because of such an episode. It is important that the radiologist, anesthesiologist, and family are all aware of the risk of aspiration in patients with epidermolysis bullosa who undergo an esophageal dilatation, partly because the patient is not intubated, but also because of the high level of many of the strictures (i.e., cervical and upper thoracic esophagus in four of six patients). We insist on overnight observation after dilatation in children with epidermolysis bullosa, because of the associated comorbidity and potential increase risk for complications.
In approximately 25% of the patients, esophageal dilatations and other aggressive supportive dietary therapies are inadequate to provide sufficient nutritional status, and placement of gastrostomy tube is required [14, 21]. The need for gastrostomy feeding may be related not only to anatomic causes of recurrent and repeated strictures but also to odynophagia, stomatitis, feeding aversion, and increased metabolic demands. Okada et al. [16] used the nonendoscopic percutaneous imaging-guided approach, also called the “push technique,” to insert gastrostomy tubes in patients with epidermolysis bullosa. They concluded that this minimally invasive technique reduces the risk of procedure-related morbidity and leads to overall improvement in the quality of life. We use a similar technique and stress the importance of assessment of skin integrity before the procedure, gentle skin preparation, significant lubrication of catheters and wires, protection of the skin from friction while dilating the tract, and placing the tube and appropriate securement of the tube with gauze covered by a self-adhesive gauze dressing, avoiding any adhesive materials. The fact that four of 10 patients underwent an early gastrostomy tube check during their first week, which is a rate far higher than usual, reflects the level of physician concern about the potential for complications in these patients. Yet all of these checks were normal, and none of our patients developed peritonitis. Another concern is that stomal problems would be likely to occur following gastrostomies. Stehr et al. [21] reported easily managed granulation formation in one of five patients with epidermolysis bullosa. In our patient group, two patients encountered stomal problems (early leakage and granulation tissue formation); the others had a satisfactory stoma.
Poor oral intake caused by painful mouth lesions or poor wound healing, probably secondary to malnutrition, are two common concurrent reasons for needing total parenertal nutrition [1, 5]. Scarring, thickening, contractures, and superinfections of the skin make peripheral venous access in these children very difficult. Central venous catheter placement for long-term therapy may be a good option to avoid recurrent skin trauma from repeated attempts to place IV canulae. The three main challenges of central venous access in epidermolysis bullosa, however, are identifying an uninvolved and uninfected portion of skin to place the line, securing the line without adhesive dressings, and avoiding infection. Preferred sites for percutaneous venous access (e.g., the antecubital fossa, groin, and neck) are frequently badly scarred and severely contracted, rendering access very difficult at these sites. We prefer the use of cuffed lines, both CVLs and PICCs, because initially the suture and then later the in-growth of skin around the cuff after about 2 weeks makes them more secure than uncuffed lines in this challenging group of patients.
Despite the minimally invasive nature of the interventional procedures and modifications implemented, complications still occur in this medically fragile group of patients. In our study, approximately half the complications were minor (Society of Interventional Radiology grades A and B, n = 15), and the other half were major (grades C and D, n = 13); however, we erred on the side of overestimating the severity of issues. The complications may be inherent to the underlying condition (e.g., blistering) or related to the procedure (e.g., esophageal tear, aspiration, or thrombophlebitis).
A multidisciplinary approach to treating patients with epidermolysis bullosa is necessary, from the time of referral to assessment and the decision to proceed. Advice regarding available options for dressings and securing techniques is valuable. With the advice of the dermatology team, we have implemented an epidermolysis bullosa kit for use in the interventional department. This kit (the Epidermolysis Bullosa Box) contains all the dressings, soft cotton rolls, permissible adhesives, and a variety of self-adhesive gauze dressing sizes that may be required. Parental involvement in dressings and securing of devices is also very helpful, because they know what works for their child. We frequently invite parents in at the end of the procedure to participate in the child's dressing.
There are several limitations to this study, including its retrospective descriptive nature and the small number of patients. Gaps in documentation in the hospital's medical records regarding dressings and home care are inevitable, because a high proportion of care is delivered by family and caregivers in the community. There are also gaps in the medical records about fine details of securing devices (i.e, gastrostomy tubes and vascular access).
In conclusion, this review of interventional procedures in this high-risk group of medically fragile patients suggests that interventional imaging-guided procedures can be successfully and safely performed when utmost care is given to skin and mucosal protection. Wound healing is adequate and should not deter intervention. Awareness of the inherent risks is important. A multidisciplinary approach is paramount.
Gastrostomy tube insertion |
| Avoid endotrachial intubation |
| Use lubricated nasogastric tube |
| Use self-adhesive gauze dressing to secure gastrostomy tube |
| Avoid excessive traction |
| Low index suspicion for peritonitis |
| Consent expanded |
| Peripherally inserted central catheters and central venous lines |
| Conduct preprocedure clinic visit and skin inspection |
| Use cuffed peripherally inserted central catheter with suture |
| Gauze wrap for securement |
| Avoid standard tourniquet if possible |
| Consent expanded |
| Esophageal dilatation |
| Approach via gastrostomy stoma if present |
| Avoid intubation |
| Lubrication |
| Overnight stay |
| Consent expanded |
| Enterostomy maintenance (no modifications are used) |
| Anesthesia |
| Avoid intubation |
| Avoid adhesive tape |
| Modified monitoring devices |
| Other (nasojejunal tube, inferior vena cava filter, liver biopsy) |
| Consent expanded |
Note—Each procedure is followed by a list of the techniques or modifications used for patients with epidermolysis bullosa.
Footnotes
Address correspondence to E. Mavili ([email protected]).
B. Connolly holds the position of Earl Glenwood Coulson Chair in Interventional Radiology at The Hospital For Sick Children.
References
1.
Freeman EB, Köglmeier J, Martinez AE, et al. Gastrointestinal complications of epidermolysis bullosa in children. Br J Dermatol 2008; 158:1308 –1314
2.
Fine JD, Eady RA, Bauer EA, et al. The classification of inherited epidermolysis bullosa (EB): Report of the Third International Consensus Meeting on Diagnosis and Classification of EB. J Am Acad Dermatol 2008; 58:931 –950
3.
Heymann WR. Classifying epidermolysis bullosa. J Am Acad Dermatol 2008; 59:1075 –1076
4.
Woodley DT, Chen M. Epidermolysis bullosa: then and now. J Am Acad Dermatol 2004; [suppl 1]51:S55 –S57
5.
Mellerio JE, Weiner M, Denyer JE, et al. Medical management of epidermolysis bullosa: Proceedings of the IInd International Symposium on Epidermolysis Bullosa, Santiago, Chile, 2005. Int J Dermatol 2007; 46:795 –800
6.
Fine JD, Johnson LB, Weiner M, Suchindran C. Gastrointestinal complications of inherited epidermolysis bullosa: cumulative experience of the National Epidermolysis Bullosa Registry. J Pediatr Gastroenterol Nutr 2008; 46:147 –158
7.
Azizkhan RG, Denyer JE, Mellerio JE, et al. Surgical management of epidermolysis bullosa: Proceedings of the IInd International Symposium on Epidermolysis Bullosa, Santiago, Chile, 2005. Int J Dermatol 2007; 46:801 –808
8.
Sy K, Dipchand A, Atenafu E, et al. Safety and effectiveness of radiologic percutaneous gastrostomy and gastro jejunostomy in children with cardiac disease. AJR 2008; 191:1169 –1174
9.
Lewis EC, Connolly B, Temple M, et al. Growth outcomes and complications after radiologic gastrostomy in 120 children. Pediatr Radiol 2008; 38:963 –970
10.
Sacks D, McClenny TE, Cardella JF, et al. Society of Interventional Radiology clinical practice guidelines. J Vasc Interv Radiol 2003; 14:S199 –S202
11.
Connolly BL. Gastrointestinal interventions – emphasis on children. Tech Vasc Interv Radiol 2003; 6:182–191
12.
Chait PG, Temple M, Connolly B, John P, Restrepo R, Amaral JG. Pediatric interventional venous access. Tech Vasc Interv Radiol 2002; 5:95 –102
13.
Fujimoto T, Lane GJ, Miyano T, et al. Esophageal strictures in children with recessive dystrophic epidermolysis bullosa: experience of balloon dilatation in nine cases. J Pediatr Gastroenterol Nutr 1998; 27:524 –529
14.
Azizkhan RG, Stehr W, Cohen AP, et al. Esophageal strictures in children with recessive dystrophic epidermolysis bullosa: an 11-year experience with fluoroscopically guided balloon dilatation. J Pediatr Surg 2006; 41:55 –60
15.
Castillo RO, Davies YK, Lin YC, Garcia M. Young management of esophageal strictures in children with recessive dystrophic epidermolysis bullosa. J Pediatr Gastroenterol Nutr 2002; 34:535 –541
16.
Okada T, Sasaki F, Shimizu H, et al. Effective esophageal balloon dilation for esophageal stenosis in recessive dystrophic epidermolysis bullosa. Eur J Pediatr Surg 2006; 16:115–119
17.
Inal M, Soyupak S, Akgül E, Aksungur EH, Akinoglu A. Fluoroscopically guided endoluminal balloon dilatation of esophageal stricture due to epidermolysis bullosa dystrophica. Dysphagia 2002; 17:242 –245
18.
Kay M, Wyllie R. Endoscopic dilatation of esophageal strictures in recessive dystrophic epidermolysis bullosa: new equipment, new techniques. J Pediatr Gastroenterol Nutr 2002; 34:515–518
19.
Karakan T, Dogan I, Cindoruk M, Dumlu S, Gorgul A, Unal S. Fluoroscopically guided endoluminal balloon dilation of oesophageal stricture due to dystrophic epidermolysis bullosa in two sisters. Acta Gastroenterol Belg 2006; 69:327 –329
20.
Heyman MB, Zwass M, Applebaum M, Rudolph CD, Gordon R, Ring EJ. Chronic recurrent esophageal strictures treated with balloon dilation in children with autosomal recessive epidermolysis bullosa dystrophica. Am J Gastroenterol 1993; 88:953–957
21.
Stehr W, Farrell MK, Lucky AW, Johnson ND, Racadio JM, Azizkhan RG. Non-endoscopic percutaneous gastrostomy placement in children with recessive dystrophic epidermolysis bullosa. Pediatr Surg Int 2008; 24:349 –354
Information & Authors
Information
Published In
Copyright
© American Roentgen Ray Society.
History
Submitted: May 3, 2009
Accepted: December 30, 2009
First published: November 23, 2012
Keywords
Authors
Metrics & Citations
Metrics
Citations
Export Citations
To download the citation to this article, select your reference manager software.