Precious Life, Precious Access
Catherine King MBChB, Timothy Evans MRCS & Yogita Aggarwal MRCP
Aim: We present the case of a young renal patient who generated significant life-threatening morbidity from her difficult and complicated renal access. The case highlights the limited life-span of temporary access (tunnelled or non-tunnelled central catheters) in renal patients who are not fit for other forms of definitive access, and how this may, in turn, limit their own life-span. We will discuss some of the complications of dialysis access, and how they can be managed.
History: A 36 year old Afro-Caribbean female presented with an acute history of severe unilateral hip and groin pain, on a background of having felt non-specifically unwell over the preceding 3 days. She had homozygous sickle cell disease, and had been admitted over 20 times in the past 12 months due to painful sickle crises, typically affecting her hips and limbs.
Past Medical History: She was well known to the renal services as she had anuric end-stage renal failure (presumed to be secondary to sickle cell disease) and been a haemodialysis patient for greater than 10 years. She was never fit nor well enough to be considered for transplantation, and had elected against peritoneal dialysis due to patient factors.
Of relevance, the patient had very difficult and ‘precious access’. She had had previous multiple failed arteriovenous fistulae (AVF) due to having small caliber veins. She also had numerous central tunnelled line insertions, and as a result had developed abnormal central venous vessels making further upper central venous catheter insertion virtually impossible. The left internal jugular vein and left subclavian vein were patent but drained into an occluded left brachiocephalic vein. The right internal jugular vein was of small caliber and grossly stenotic at the junction with the subclavian vein. Furthermore, the right subclavian vein was small and stenosed (Figure 1):
Examination: On admission she had the following observations and examination findings:
On further questioning, there had been difficulties in achieving good dialysis blood flows in the venous port for the last 4 dialysis sessions. She had not had any rigors on dialysis - which is often classical for a line related infection. The remainder of the examination was normal.
Impression: The patient was considered to have a sickle crisis secondary to infection, the most likely source being the tunnelled femoral dialysis line.
1) Sickle Cell Crisis: The patient was treated for a non-severe sickle cell crisis with opioid analgesia, oxygen and 250ml aliquot boluses of normal saline. Fentanyl was used in preference to morphine as it has no active metabolites, and thus is relatively safer in renal patients. Morphine is not recommended in dialysis patients due to the rapid accumulation of active, non-dialysable metabolites that are neurotoxic. The patient only received small boluses of fluid, instead of a maintenance infusion, as greater control can be maintained of a patient’s fluid balance by continual assessment of the volume state before and after each fluid bolus. Anuric patients can rapidly become fluid overloaded and in extremis with even 500mls of fluid especially if they are already slightly hypervolaemic. The patient also received a blood transfusion over dialysis so that the extra volume and salts (in particular potassium) could be removed. The patient’s admission potassium was high, and by giving the patient a blood transfusion would risk elevating the serum potassium further. Blood cells are rich in potassium which is released when cells are lysed.
2) Poor Flows onialysis: The potassium was higher than the usual pre-dialysis value which coincided with clearances of toxins from the blood being poorer. Poor dialysis blood and dialysate flows are often due to intra- and extra-luminal clot and biofilm[i] formation. Line salvage techniques, such as catheter lock thrombolytics and thrombolytic infusions, can be used before considering alternative access and line exchange. In this case, line salvage was preferred as the patient had precious access and a high potassium level. She received a thrombolytic line lock, after which the line worked sufficiently well for an adequate dialysis session and blood transfusion.
3) Central-Venous catheter (CVC) related bacteraemia: Before antibiotics were given, paired (peripherally and from the CVC) blood cultures were taken. There was no suggestion of endocarditis and thus only two sets of blood cultures were taken. In the absence of patient-specific microbiological past-history, the patient was started on broad spectrum Gram-negative and positive antibiotics, as per unit guidance. In this case, due to patient allergies, she was commenced on meropenem and teicoplanin. The blood cultures grew a combination of Gram-positive and Gram-negative organisms: 2 species of Enterococci, a Proteus and Klebsiella were cultured from the patient’s blood. Mixed growth bacteraemias can orginate from bowel pathology. The patient denied any bowel symptoms and an abdominal ultrasound and non-contrast abdominal CT scan were normal. The patient was not well enough for any other imaging during her inpatient stay.
The bacteraemia was slow to respond despite sensitive anti-microbial therapy. The patient continued to spike temperatures and the CRP remained elevated and the blood cultures from the line continued to grow the aforementioned species. The source was either the line, septic embolic seeding, or both.
Alternative Access: The patient required the tunnelled line to be removed – however, as she had ‘precious access’, radiological guidance for future dialysis access insertions was sought. Further, in addition to the above mentioned access difficulties (Figure 1.), femoral lines were technically difficult to insert due to the patient having developed severe flexion deformities of the hip as a result of iatrogenic hip fusions following previous osteoporotic fractures.
The tunnelled line was removed and temporary dialysis lines (radiologically-inserted) were used as a bridge until a tunneled dialysis line could be re-inserted into the left femoral vein. The patient required temporary lines for 4 weeks before the CRP fell to a satisfactorily low level and line cultures were repeatedly negative. The temporary femoral lines were exchanged every 5 days in the radiology department in order to avoid significant colonisation of the temporary dialysis catheters. The right femoral vein could not be cannulated due to technical difficulties associated with the patient’s posture.
Septic Embolic Seeding: The patient was also assessed for possible septic seeding. She had a trans-thoracic echocardiogram which did not show any vegetations. A trans-oesophageal echocardiogram was delayed due to the patient not being fit for the procedure on numerous occasions. Her abdominal imaging was normal. She did not complain of back pain, and refused further CT and MRI imaging due to suffering from severe claustrophobia. After a further week of continued antibiotics and temporary non-tunnelled CVCs, the CRP continued to fall and the blood cultures remained negative. A new tunnelled line was inserted under radiological guidance and general anaesthesia. This proceeded without complications and the patient was discharged home. In total she was given 6 weeks of antimicrobial cover, under microbiological advice, as clinically she had behaved as if she had embolic seeding from the CVC.
1. Dialysis Access, whether it be in the form of a fistula or tunnelled line, often can be the last resort access point for a patient. Failure of this access can create significant clinical difficulties in providing life-maintaining treatment. The case highlights the importance of protecting access in renal patients. CVC failure is associated with increased use, and a failure to follow aseptic techniques.
So friendly tips would be:
- Do not cannulate an AVF, an arm with an AVF in it and do not take bloods from it.
- Do not use a tunnelled line to administer drugs or take bloods from – unless you are formally trained to use a renal tunnelled catheter or have been advised by the renal specialists to do so (which is very unlikely).
2. This case has also highlighted the importance of checking for seeding of infection in dialysis patients. This patient’s temperature, bacteraemia and CRP took some time to fall following the initiation of antibiotics, which were all suggestive of peripheral infective seeding from the CVC. Infective seeding can occur in any end-organ vascular bed such as the spine, peripheral limbs, heart valves, lungs, pulmonary trunk in the form of a PE, and the brain. Investigations need to be clinically guided in order to find peripheral seeding, as the management will not change with the regards to the duration of antibiotic therapy, but whether other forms of intervention are required, i.e. surgery for discitis or infective endocarditis.
[i] Dry weight – There is no standard definition for dry weight, but it is accepted to be the weight at which you would expect a person to be in the presence of normal physiological renal and cardiac function i.e. hydrate. Renal patients often have reduced urine outputs, and thus in order to maintain their dry weight per day, their fluid intake/day = urine output/day + insensible losses/day
[ii] Biofilm – is a film composed of cells of microorganism which attach to a surface and make a thin layer. It can be intra- and extra- luminal. In the biofilm phenotype, the microorganism shows greater resistance in its characteristics when compared to its free-swimming counterparts.