Despite being the third choice for type of vascular access for hemodialysis, 1 double lumen tunneled hemodialysis catheters are used for more than 25% of patients in several dialysis units around the world. 2 Well recognized complications 3–6 associated with catheters include infections, stenosis, and thrombosis of central veins, as well as mechanical dysfunction reducing dialysis delivery.
Less frequent complications such as local hematomas and external bleeding, 7 atrial thrombus, 8 excessive catheter adherence, 9 or accidental catheter removal have been also described. An even more infrequent complication has been catheter fracture as reported in less than 1% of subclavian venous catheters implanted for chemotherapy. 10,11 Such a rare complication has been reported only recently with hemodialysis catheters. 12–14
The authors report here another case of catheter tip fracture with a different mode of presentation with the purpose of increasing the awareness of potential hazards associated with hemodialysis catheters.
T.N. is a woman aged 44 years from Vietnam who weighs less than 50 kg. In 1980, she was diagnosed with Berger’s disease, which was proven by biopsy. She did not have any other known comorbidity. She began peritoneal dialysis in 1994 and had several episodes of bacterial peritonitis over the years. Peritoneal dialysis was pursued until 2001 when a rare complication, sclerosing peritonitis, occurred and for which an immunosuppressive regimen (steroids and mycophenolate mofetil) was initiated. It was then decided to transfer her to chronic hemodialysis. On July 17, 2001, a cuffed double lumen tunneled hemodialysis catheter (Opti-Flow, 19 cm straight, Bard Access Systems, Salt Lake City, UT) was inserted into the right internal jugular vein by a radiologist. The insertion had been easy with no reported complication, and the chest x-ray after the procedure was normal with adequate catheter positioning. A left brachiocephalic arteriovenous fistula was created a few weeks later, but unfortunately it never matured enough to be punctured.
Hemodialysis was prescribed three times per week for 4 hours per session using a Fresenius 2008H monitor (Fresenius, Walnut Creek, CA), a standard bicarbonate dialysate, and a F80 filter (Fresenius). A blood flow rate in the range of 300 to 400 ml/min was easily obtained from the Opti-Flow catheter most of the time with ports in the nonreversed position. Indeed, ports had been reversed on a few occasions only (on the following dates: February 21, March 2, March 28, and April 2, 2002). Standard heparinization was administered during hemodialysis (bolus of 3,200 units followed by 800 units per hour, for a total of 6,400 units of unfractionated heparin per session). The catheter was closed at the end of each treatment with normal saline followed by standard heparin 1,000 units/ml to fill the volume of each lumen (1.9 ml in the arterial line; 2.0 ml in the venous line). Only one treatment with intracatheter alteplase had been given on March 19, 2002. Hemodialysis was fairly well tolerated in this patient, and the catheter was never considered problematic in this patient.
On April 30, 2002, at the end of an otherwise uneventful dialysis session, while disconnecting the circuit from the catheter, the patient made a noise similar to a gasp and went into cardiorespiratory arrest. The catheter was clamped, but no syringes were connected to it yet. The chair was immediately placed in the Trendelenbourg position and the nurse started cardiopulmonary reanimation while waiting for further help.
On the monitor, ventricular fibrillation was seen; an electrical cardioversion was followed by intravenous epinephrine. She then presented sinus bradycardia that required intravenous atropine. Endotracheal intubation was difficult because of a very small mouth, and four attempts were required before it was successful. After these interventions, the patient remained relatively stable from a hemodynamic standpoint, and she was transferred to the intensive care unit. She nevertheless had been severely hypoxic.
After arrival in the intensive care unit, it was noted that the external venous clamp of the Opti-Flow catheter was broken; because this had not been noticed before, the hypothesis of a break during the reanimation process was entertained. The electrocardiogram showed sinus rhythm with no evidence of ischemia, and the creatine kinase and troponine concentrations remained within the normal range, rendering the hypothesis of a primary cardiac event unlikely. Biochemistry drawn during the arrest showed no clinically significant abnormality (Na 146, K 4.0, and Mg 0.95 mmol/L). Initial blood gas showed a mixed acidosis that quickly came back to normal. We were left with the hypothesis of an air embolism to explain the sudden cardiorespiratory arrest. The neurologic exam at the time of admission to the intensive care unit was not reassuring. The patient was in a deep coma with decerebration and anoxic myoclonia.
Because the venous clamp of the catheter was broken and the catheter had been used in a nonsterile manner to administer medications during the arrest, antibiotics were given as a prophylaxis, and it was decided to change the catheter over a guide wire the next morning. To our surprise, the routine chest x-ray postintubation showed that the tip of the Opti-Flow catheter was ruptured and visible in one relatively distal branch of the right pulmonary artery (Figure 1). The last chest x-ray available for comparison was from October 12, 2001, and it did not show a fractured catheter tip. The family and the manufacturer were both notified of the unexpected finding.
The next morning (May 1, 2002), the catheter was changed over a guide wire for another Opti-Flow by the angioradiologist. The broken catheter was sent for analysis to the manufacturer (Figure 2). The day after, a selective angiography of the right pulmonary artery was performed with the purpose of removing the fractured catheter tip. Unfortunately, the fractured catheter tip was too anchored to be removed. According to the radiologist, an endothelialization process had already started covering the inside of the tip, also possibly suggesting that the fracture had occurred more than 48 hours ago.
To better document the impact of the presence of the fractured catheter tip on pulmonary perfusion, an angioscan and a ventilation perfusion scintigraphy were requested. No deficit of pulmonary perfusion could be seen (Figure 3), probably because of the small size of the tip and of its distal position. A cardiac echography was also found to be normal. After consultation with thoracic surgery, it was decided that a thoracotomy to remove the tip would be major and extremely risky. The tip therefore remains in the right pulmonary vasculature of our patient.
The patient recovered neurologic function extremely slowly over the next 4 months. A slight impairment of intellectual functions and a mild paresis of her left arm are the main sequel. She was finally discharged from the hospital on August 21, 2002, and she was able to return home.
We revised her entire dialysis chart carefully; the only finding was on November 10, 2001, when the patient had complained of a vague thoracic pain during an episode of hypotension, and both the pain and the hypotension resolved spontaneously. An electrocardiogram at this time was normal, and the saturation was 100% at room air.
Tunneled central venous catheters have emerged as a valuable alternative for long term hemodialysis access. Despite sustained improvements in catheter design and material, particularly in recent years, 15 catheters are still associated with an increased morbidity and risk of death. 16,17
While searching the literature, we found very few cases of hemodialysis catheter fractures. 12–14,18,19 Anuradha et al.13 reported a rare case of spontaneous fragmentation and embolization of a double lumen hemodialysis catheter from the right femoral vein into the inferior vena cava, right atrium, and the superior vena cava. The catheter was subsequently removed percutaneously with success. Hu and colleagues 14 reported a case of accidental migration of double lumen catheter fragment into the internal jugular vein during a guide wire exchange; while the mid portion of the catheter was cut for guide wire introduction, the catheter fragment slipped under the neck skin. The catheter fragment was immediately retrieved by a radiologist using the femoral route. Verhage and van Bommel 18 reported a break in a dialysis catheter after several manipulations, but fortunately, there was no embolization. Although the catheter was inserted in the right jugular vein, a pinch-off sign was observed as described in subclavian. Székely and colleagues 19 noted tip rupture in three patients who were harboring single lumen subclavian cannulae of 2.8 mm thickness and 14 cm length, used for 1, 5, and 6 weeks of single needle hemodialysis. The broken pieces migrated in pulmonary vasculature and remained in place without associated complications for as long as 33 months.
The recent case report by Chawla et al.12 is probably the most similar to ours. They report a case of catheter tip fracture and migration causing cough; the catheter tip was found also in a distal segment of the right pulmonary vasculature. Those authors reviewed the FDA medical equipment malfunction reports database and found more than 35 cases of catheter rupture at that time, involving different catheter types and manufacturers. Such as the case reported by Chawla and colleagues, our case concerns the fracture of the catheter tip exclusively.
The macroscopic and microscopic analysis performed by Bard Access Systems upon our fractured catheter showed the following: “The venous staggered tip was missing from the catheter and both venous and arterial lumens were patent to infusion and under pressure, no leaks were observed. The staggered portion of the venous tip was severed from the assembly at the diameter transition. A cross sectional view of the severed site exhibited a jagged, irregular edge with a granular veneer. The characteristics of the damage found at the staggered venous transition site are consistent with a tensile break, however the exact mechanism of the damage is unknown.”20
In the past, Bard Access Systems has used a tip molding process that was later recognized as having the potential to create microfractures in the tip of the catheter, 20 and a product recall notice was largely distributed in March 2001. We were notified to return certain lots to the company in April 2001, and all identified lots were resent to Bard, which was several months before the insertion of the first Opti-Flow in our patient in July 2001. Therefore, we are fairly confident that she was not harboring a catheter with a potential defect. Nevertheless, since that time, we also carefully register the lot number of every catheter inserted in the chart of the concerned patient.
The first Opti-Flow catheter of our patient had been inserted under fluoroscopic guidance by an expert radiologist, and the procedure was smooth and uneventful; it is therefore unlikely that any damage or breakage was induced at that time. In addition, warning symptoms of imminent catheter fracture have been reported; they include malfunction, resistance to liquid injection, chest pain, palpitation, and cough. However, our patient did not present any of them.
The exact etiology of the arrest will never be fully understood. Was it caused by the tip embolization, or did an air embolism occur? Air embolism is a rare but expected complication of hemodialysis, despite modern monitors with secure alarms, 21,22 especially with catheters. 23,24
The present authors do not have a clear explanation for the catheter tip rupture in this patient nor for the exact etiology of her cardiac arrest. The authors conclude that, despite its rarity, tip rupture should be considered to be a potential complication of long term hemodialysis catheters. Clinicians must remain aware of possible occurrence and potential hazards.
1. National Kidney Foundation Dialysis Outcomes Quality Initiatives: Clinical Practice Guidelines for vascular access. New York, National Kidney Foundation, 1997, pp. 13–33.
2. Pisoni RL, Young EW, Dykstra DM, et al: Vascular access use in Europe and in the United States: results from DOPPS. Kidney Int 61: 305–316, 2002.
3. Butterly D, Schwab SJ: The case against chronic venous hemodialysis access. J Am Soc Nephrol 13: 2195–2197, 2002.
4. Jean G, Vanel T, Chazot C, Charra B, Terrat JC, Hurot JM: Prevalence of stenosis and thrombosis of central veins in hemodialysis with a tunneled jugular catheter. Néphrologie 22: 501–504, 2001.
5. Beathard GA: Catheter thrombosis. Semin Dial 14: 441–445, 2001.
6. Jean G: Incidence and risk factors for infections from hemodialysis catheters. Néphrologie 22: 443–448, 2001.
7. Huang EY, Cohen S, Lee J, et al: Diagnosis of unexplained bleeding from tunneled dialysis catheter. Am J Nephrol 21: 397–399, 2001.
8. Kingdon EJ, Holt SG, Davar J, et al: Atrial thrombus and central venous dialysis catheters. Am J Kidney Dis 38: 631–639, 2001.
9. Mahadeva S, Cohen A, Bellamy M: The stuck central venous catheter: beware of potential hazards. Br J Anaesth 89: 650–652, 2002.
10. Aitken DR, Minton JP: The “pinch-off sign”: a warning of impending problems with permanent subclavian catheters. Am J Surg 148: 633–636, 1984.
11. Noyen J, Hoorntje J, de Langen Z, Leemslag JW, Sleijfer D: Spontaneous fracture of the catheter of a totally implantable venous access port: case report of a rare complication. J Clin Oncol 5: 1295–1299, 1987.
12. Chawla LS, Chegini S, Thomas JW, Guzman NJ: Hemodialysis central venous catheter tip fracture with embolization into the pulmonary artery. Am J Kidney Dis 38: 1311–1315, 2001.
13. Anuradha S, Bansal R, Gupta A, Sinh NP, Kohli R, Agarwal SK: Embolisation of haemodialysis catheter. J Assoc Physicians India 45: 608–609, 2002.
14. Hu CC, Lin SC, Huang PH, Tseng SS, Tarng DC: A lady with a broken haemodialysis catheter fragment. Nephrol Dial Transplant 17: 1126–1128, 2002.
15. Ash SR: The evolution and function of central venous catheters for dialysis. Semin Dial 14: 416–424, 2001.
16. Butterly DW, Schwab SJ: Catheter access for hemodialysis: an overview. Semin Dial 14: 411–415, 2001.
17. Pastan S, Soucie JM, McCellan WM: Vascular access and increased risk of death among hemodialysis patients. Kidney Int 62: 620–626, 2002.
18. Verhage AH, van Bommel EF: Catheter fracture: an underrecognized and serious condition in haemodialysis. Nephrol Dial Transplant 15: 901–903, 2000.
19. Székely J, Racz L, Karatson A: Broken piece in the lungs: a complication of haemodialysis via subclavian cannulation. Int Urol Nephrol 21: 533–540, 1989.
20. Bard Access Systems. Product recall notice. Salt Lake City, UT, Bard Access Systems, March 27, 2001.
21. Anonymous: Air embolism associated with hemodialysis. Health Devices
18: 406–407, 1989.
22. Harvey GR, Lyness J, McGeown MG: Air embolism during hemodialysis. Lancet 7744: 259–260, 1972.
23. Ely WE, Hite D, Baker AM, Johnson MM, Bowton DL, Haponik EF: Venous air embolism from central venous catheterization: A need for increased physician awareness. Crit Care Med 27: 2113–2117, 1999.
24. Marin B, Chouvy V, Denier M, Debard I, Tran-Minh V: Gas in hepatic veins associated with tunneled femoral hemodialysis catheter. J Radiol 83: 47–48, 2002.