The formation of bronchocutaneous fistulae due to retained epicardial pacing wires: A literature review.

Temporary epicardial pacing wires during open-heart surgery are routinely used both for diagnostic and treatment purposes. In complicated cases where patients are unstable or the wires are difficult to remove, the pacing wires are cut at the skin level and allowed to retract by themselves. This procedure rarely causes complications. However, there have been cases reporting that retained pacing wires are linked to the formation of sterno-bronchial fistulae, which may present a while after the date of operation and are usually infected. This review aims to study the cases presenting sterno-bronchial fistulae due to retained epicardial pacing wires and to highlight the important factors associated with these. It is important to note these complications, as fistulae may cause a variety of problems to the patient if undiagnosed and left untreated. With the aid of scans such as fistulography, fistulae can be identified and treated and will improve the patients' health dramatically.

INTRODUCTION

The use of temporary epicardial pacing wires during open-heart surgery is usually a routine procedure, not commonly associated with a high morbidity.[1] Epicardial pacing wires are customarily used to diagnose and treat dysrhythmias during and following cardiac surgery.[2] These wires are usually removed within 5 postoperative days.[3] However, in complicated cases, where the removal of the wires may be difficult, they are cut at the skin level and left in situ.[3] Retained pacing wires rarely cause complications and patients usually tolerate them very well, relating to a low overall morbidity.[34] On the other hand, infections of sternal wounds account for some of the most serious complications postoperatively for mediastinal sternotomies.[5] They give morbidity rates at an incidence of 0.4–5%, and mortality rates within the hospital reach 7–29%, even after the correct treatment of mediastinitis is accounted for.[5] Taking into account the aforementioned, although retained pacing wires are part of a routine procedure, which is well tolerated, it is important to note that complications do occur and they need to be considered when they are left in situ. This literature review aims to look at the formation of fistulae from pacing wires and will particularly focus on the formation of bronchocutaneous fistulae as they are a very rare phenomenon, but one which has to be highlighted, as severe complications may arise from their presence. To the best of our knowledge, there are currently only three cases reported on the formation of sterno-bronchial fistulae[256] and one on the formation of cardio-cutaneous fistulae,[3] all thought to be caused by retained pacing wires.

RESULTS

Taking into account that sterno-bronchial fistulae are very rarely caused by pacing wires left in situ, to the best of our knowledge, only three articles were found describing such case studies. The case reports found were analyzed and compared to find similarities and differences, which may aid clinicians in future, considering the potential of such a severe complication of pacing wires left in situ. Mand’ák et al.[6] described a case of a 46-year-old male who initially presented with ischemic heart disease and 6 years post his coronary surgery developed fistulae. The patient underwent a standard “cardiac operation” with “cardiopulmonary bypass” in the year 1978, without any postoperative complications. Six years later, he presented with “an inflamed wound and free-draining pustules along the sternotomy incision.”[1] Due to the lack of evidence of any substernal communications on the fistulogram performed, a radical surgical treatment was used which included the full “removal of the wires, stitches, a surgical debridement, suction drainage, and closed irrigation of antimicrobial and antibiotic solution” and antibiotic administration.[1] On culturing, the bacteria such as Staphylococcus aureus, Pseudomonas, Staphylococcus epidermis, Acinetobacter, and Corynebacterium were found. The patient had two subsequent myocardial infarctions, which worsened his condition to the New York Heart Association II and an ejection fraction (EF) of 30%. In his final hospitalization in 1997, multiple skin and substernal fistulae were found, one of which communicated with the left bronchial tree. The fistulography showed “filling of the substernal cavity with contrast medium, the communication with bronchial tree and irregular bronchus affected by chronic inflammation, and the communication with the left main bronchus.”[1] The contrast computed tomography (CT) showed inflammatory changes too, and S. aureus and Corynebacterium diphtheriae were cultivated. Although a radical operation, wherein a total excision of the sternum, costal cartilages, and lung and chest wall resection would be preferred, this patient was treated conservatively with dressings and antibiotics due to his level of high risk. The patient died a year later due to suprarenal carcinoma with liver metastasis. A similar case reported by Sakellaridis et al.[5] described the formation of a fistula, communicating from the substernal space to the skin and to the left and right bronchial tree, which may be linked to retained epicardial pacing wires. Sakellaridis et al.[5] reported a case of a 70-year-old gentleman with a history of lung emphysema and an occluded left anterior descending artery (LAD), who underwent an off-pump coronary artery bypass graft (CABG), where atrial and ventricular temporary epicardial pacing wires were used as standard. The wires were cut at the skin level and left to retract before discharge. A year later, the patient presented with an “inflamed Type V wound”[5] at the lower part of the sternotomy, which led the patient to undergo an operation to remove the sternal wires, and have a surgical debridement. Antibiotic and antimicrobial closed circuit irrigation was applied, and systemic antibiotics were administered. The patient was fully discharged 10 days later, with no further signs of inflammation. However, 9 years later, the patient presented with a cough and a free draining pustule at the lower part of the sternotomy. Although the chest X-ray and CT appeared normal, the fistulography showed a cutaneous sternal fistula filling a small substernal cavity with contrast medium and communication with both bronchial trees, characteristic with the presence of retained pacing wires. The fistulae were at the superior division of the lingular bronchus on the left lung and at the bronchus of the right upper lobe.[5] The cultures of the patient presented S. aureus, and the patient was treated with antibiotics, which resolved the free drainage. Once again, although a radical surgical treatment would be preferred to remove the epicardial pacing wires and to have a debridement, due to the patient's unstable medical state (age, emphysema with forced expiratory volume in 1 s: 1, 2 L) and his refusal to undergo surgery, the patient was treated conservatively with antibiotics and showed no signs of infection 8 months later. Nowicki et al.[2] reported a case of a 53-year-old male who presented with a deep peristernal infection followed by the formation of a sterno-bronchial fistula post an on-pump CABG operation. Initially, the patient was admitted with unstable angina and was diagnosed with quadruple-vessel coronary artery disease, arterial hypertension, and hyperlipidemia, with an left ventricular EF of 55% and a logistic euroscore of 3.1%, which led to an emergency on-pump CABG operation. The operation involved one “arterial (left internal thoracic artery) to LAD and three venous grafts (saphenous vein graft to ramus intermediate, right coronary artery, and obtuse marginalis), aorto-coronary bypasses”[2] being implanted. Due to the patient's bradycardia and insufficient cardiac output, atrial and ventricular pacing wires were used. Postoperatively, the patient had several complications with low cardiac output, which required continuous infusion of inotropic agents, antiarrhythmics, and the establishment of an intra-aortic balloon pump. The patient stabilized and was discharged on the 9th following day with no signs of sternotomy infections. A month later, the patient was admitted with a “soft tissue abscess in the lower aspect of the wound,” fever at 39°C, and subscapular back pain. Within the following 12 months, the patient had several hospitalizations due to the persistent pain and fever, which led to a suspicion of pneumonia. During this time, several fistulae and multiple recurrent skin abscesses were observed, which were not cured by neither oral nor parenteral antibiotics. In one of the admissions, a year post, the initial operation of methicillin-sensitive S. aureus (MSSA) was cultured, which led the patient having an operation to remove two metallic sutures from the lower sternum, surgical debridement, and antibiotic therapy, which resolved the infection. In the following month, a follow-up CT showed purulent discharge around a correctly placed pacing wire on the right atrial wall, which formed visible gaseous bubbles, followed by a new onset of skin fistula a few days later, on the upper part of the sternum. The abscess irrigation performed made the patient expectorate which led to suspecting a bronchial tree fistula, confirmed by a fistulography. It is important to note that gastrofiberoscopy, CT, and bronchoscopy showed no evidence of the fistula. The shadow of the pacing wire could be observed in the “projection of the purulent canal communicating the skin with the bronchial tree.”[2] This led to explanting the wires under local anesthesia with the aid of CT, which resolved all infections and inflammations and led the patient being asymptomatic till the date of publication in 2011. All the three aforementioned cases focus on the formation of sterno-bronchial fistulae due to retained pacing wires. Although it is a very rare postoperative complication, it is very important to make a note of such cases as they are life-threatening and can be treated if acknowledged in time. As mentioned previously, only three such cases have been reported to the best of our knowledge until now. However, there have been a few more reports of pacing wires causing complications of a similar nature: Dobiáš et al.[3] reported a case of a cardio-cutaneous fistula post a CABG surgery treated with video-assisted right anterolateral minithoracotomy approach. Smith et al.[4] described a case of a focal retroaortic abscess presenting 3 years after surgery for tricuspid valve endocarditis. Finally, Kapoor et al.[1] reported a case of “right paracardiac mass due to organized pericardial hematoma around retained epicardial wires” presenting 3 months postoperatively. The last three cases showed a different side of complications caused by retained epicardial pacing wires; however, they can be associated with the fistulae as they come from the same source and are all extremely important to take into account when dealing with patients recovering postoperatively.

DISCUSSION

Temporary pacing wires are routinely used during cardiac surgery to diagnose or treat arrhythmias[1] and are often cut at the skin level and left to retract on their own. This process usually causes no further complications. Clipping the pacing wires remove the risk of bleeding related to excising the wires, but a foreign body is left in the mediastinum,[4] which may potentially create a number of risks for the patient. Complications are, however, rarely reported, especially those involving sterno-bronchial fistulae. Nevertheless, as seen above, cases of abscesses, hematomas, or fistulae of the cardio-cutaneous areas have also been reported and worth noting, even though they are still very rare.

Due to the rarity of the complications around epicardial pacing wires, especially the formation of sterno-bronchial fistulae, to the best of our knowledge, only three articles were found describing such cases. The pathophysiology of the sterno-bronchial fistulae remains unclear, and there are no supportive data to show that epicardial pacing wires of metallic sternal sutures can result in the formation of fistulae.[2] It is important to note that in all the above cases, a fistulography was required to show the fistulae and that CT and X-ray were often inadequate to provide the evidence needed. Another important observation is that in all the three cases of sterno-bronchial fistulae, S. aureus was cultured and was part of the etiology, and that although the peristernal tissue was involved, the sternal bone remained intact.[2] The removal of the pacing wires under local anesthesia can itself leave behind a communication between the pericardium and bronchial tree, which may lead to further complications and must be considered when dealing with patients who are at a higher risk and cannot undergo a radical operation. All the three cases showed an element of infection with the fistulae and a good recovery when the patients were treated up to the degree of follow-up allowed by the patient, although uncommon sterno-bronchial fistulae can occur and they can easily be missed if a fistulography is not performed. Treating an infected fistula with antibiotics may slow down the process of infection, but will not fix the fistula, which will require an operation. In patients who are generally stable, removal of the wires, debridement, and a course of antibiotics may resolve a problem which could potentially be life-threatening if not accounted for and treated. To conclude, sterno-bronchial fistulae formed due to retained pacing wires after a cardiac operation are relatively rare complications, which do, however, occur and may not be initially diagnosed, prolonging patient deterioration. Once diagnosed, they are treatable and resolvable and will allow the patient to fully recover. The presence of MSSA in cultures is important to note too, but has no supportive evidence as to how it may add to the formation of fistulae.

Financial support and sponsorship

Nil.

Conflicts of interest

There are no conflicts of interest.