January 2015

The Perils of Anticoagulation in a Chronic kidney Disease (CKD) 5 Patient 

Eleanor Singer MBBS, Foundation Year 1.   (Edited by Yogita Aggarwal MRCP)


Introduction:  Patients with advanced chronic kidney disease (CKD) are prone to both a prothrombotic state, predisposing to thromboembolism, and a coagulopathy, which results in an increased tendency for bleeding (Jalal et al, 2010).  Whilst the mechanisms are not fully understood, what is known is that there are abnormalities associated with the coagulation cascade, platelet aggregation and function, and the regulation of the processes involved in haemostasis (Jalal et al 2010, Capodanno and Angiolillo 2012).   For those CKD patients that require anticoagulation, these physiological tendencies have to be taken into account when weighing up the benefits of therapy against the risk of thromboembolism in the absence of anticoagulation therapy and the risk of haemorrhage in the presence of anticoagulation therapy. 

The case presented is that of a pre-dialysis patient, established on warfarin, who suffered a spontaneous psoas muscle haemorrhage in the context of being converted to unfractionated heparin therapy peri-operatively.  The case highlights a recognised complication of anticoagulation, spontaneous haemorrhages, the risk of which is increased in advanced CKD due to a bleeding diathesis.   


Case history: 

  • The need for surgery

A 51 year old female of Indian ethnic origin had progressed over the course of a decade to heavily proteinuric CKD stage 5 secondary to poorly controlled hypertension.  Her current eGFR was 12ml/min/m3 and she had elected to have peritoneal dialysis as the mode of renal replacement therapy whilst a relative was being worked up as a live renal transplant donor.  Given her comorbidities, the patient was listed for a planned surgical peritoneal catheter insertion under direct visualisation in order to reduce the risk of iatrogenic complications. 

  •  Her medical background included the following:
  1. End-stage kidney disease secondary to hypertension – biopsy confirmed
  2. Metallic mitral and aortic valves, on warfarin with target INR 3-4
  3. Ischaemic heart disease with previous PCIon concomitant Aspirin and Simvastatin
  4. Atrial Fibrillation - non thrombotic to date
  5. Hypertension, on Amlodipine, Bisoprolol
  6. Glaucoma, on Latanoprost
  7. She had no prior history of bleeds.
  •  Admitted for conversion to a shorter acting anticoagulant

The patient was electively admitted a few days before her operation, so that she could be converted from warfarin to a shorter acting anticoagulant, which in this case was intravenous (IV) unfractionated heparin.  Advice was taken from the local Haematology team who suggested initiating unfractionated heparinisation, as per local Trust protocol, once the INR was less than 1.8. The antiplatelet therapy was discontinued 5 days prior to the operation.  

  •  A difficulty with intravenous heparin therapy-labile control

Unfractionated heparin was started when the INR was 1.78.  She received a loading bolus dose and maintenance infusion with a plan to check the APTTr in 6 hours’ time.  The initial 6 hour APTTr was low at 1.62 and the patient had an appropriate dose increase, as per protocol.  A subsequent APTTr was rechecked 6 hours after the dose increase.  The result was processed urgently within the hour, and demonstrated that the APTTr had increased to 3.4 and the infusion was stopped for an hour before a dose reduction was to be made, again as per unit protocol - see appendix 1. 

  •  Development of a suspected acute bleed

Within the next 2 hours, the patient subsequently developed back pain, radiating to her left hip, accompanied by a relative tachycardia of 90bpm (the heart rate may have been ‘capped’ due to the effect of beta-blockade), blood pressure of 98/60mmHg and a drop in haemoglobin (Hb) by 3g/dL.  On examination she had pain over her left renal angle and reduced movements of her leg at the left hip joint.  A retroperitoneal bleed was suspected. 

  • Management of a suspected acute bleed whilst on anticoagulation: stop anticoagulation, achieve haemodynamic stability, achieve robust venous access, consider reversal of anticoagulant effect, arrange close observation of the patient, consider invasive monitoring, and consider interventional radiology.

The IV heparin was stopped, a repeat APTTr was sent and the patient was haemodynamically stabilised with bolus fluid challenges and a blood transfusion (of 2 units) to maintain a Hb of 8g/dL. 

All anti-hypertensives were suspended and two large bore peripheral venous cannulae were maintained.  

The antidote to unfractionated heparin is protamine sulphate and can be difficult to tolerate due to its side effect profile.  The protamine sulphate was mobilised from the on-call emergency cupboard to activate if the patient did not stabilise or if the repeat APTTr was higher.  The action of unfractionated heparin lasts 4-6 hours as it has a short half-life. 

A CT angiogram +/- embolisation intervention demonstrated a large acute retroperitoneal haematoma within the psoas muscle which was no longer actively bleeding.  The patient did not require embolisation.

 The patient was transferred to a renal HDU bed and monitored closely overnight. 

 The heparin remained suspended overnight, as per Haematology advice – this decision was based on a risk versus benefit decision.  Her condition remained stable and serial VBGs showed no further drop in her haemoglobin.

  •  Subsequent sub-acute anticoagulation consideration – Risk of developing thromboembolism becoming greater than risk of a further acute bleed

Given the patient had advanced CKD, an acute immobilising event, mitral and aortic mechanic valves, and atrial fibrillation which all increased her risk of developing thromboembolism; and because clinically the patient was no longer displaying signs of ongoing haemorrhage, the intravenous unfractionated heparin (which had been suspended for 12 hours) was reintroduced at a low dose continuous infusion and was gradually increased to achieve a therapeutic APTTr over the next 24-48 hours as per advice taken from cardiology and haematology specialists.  Through this period, her Hb was monitored on a 4-6 hourly basis and her observations were taken hourly – both demonstrated that the patient was not bleeding and that she had become haemodynamically stable.

  •  Proceeding to surgery – ‘no rush’

After 36-48 hours, the patient had further imaging which demonstrated that the retroperitoneal haematoma was no larger.  A decision was made that the patient was fit for a peritoneal catheter insertion under GA which was successfully completed, under laparoscopic guidance, a week after the initial psoas muscle haemorrhage.

  •  Case outcome

6 hours after the operation, the unfractionated heparin was re-continued.  Once the PD catheter was confirmed to be functional, she was then re-started back onto warfarin 48 hours after the operation. 

Amazingly, the patient did not have any further loss of her eGFR during this acute stage. 


Learning points from the case:

  1. Advanced CKD patients have altered haemostasis putting them at increased risk of thromboembolism and acute bleeds.  The use of anti-platelets and anticoagulation increased the risk of bleeds and must be weighed up against the risk of thromboembolism formation i.e. strokes and myocardial infarctions.
  2. Patients are often admitted a few days before elective surgery for the conversion of warfarin anticoagulation to unfractionated intravenous heparin.  However, novel anticoagulants are being used, which, despite having limited trial data in CKD 4 and 5 patients, are more selective in their inhibition of the coagulation cascade and do not require as intensive routine monitoring as warfarin or intravenous unfractionated heparin.  Widespread use is limited by cost, trial data and the lack of antidotes.
  3. The case also covers some of the important considerations when managing a patient with an acute haemorrhage.



1.  Altered haemostasis in CKD:

Disorders of coagulation are common in CKD.  Under normal circumstances, vessel wall injury initiates a tightly controlled cascade which leads to the formation of clot and minimises tissue damage.  The process involves coordinated:

  • platelet accumulation and activation, and
  • thrombin and fibrin formation, and
  • other ill-defined control mechanisms.

Patients with CKD have a spectrum of abnormalities which, whilst not fully understood, centre around a prothrombotic state due to platelet hyper reactivity, and increased fibrin and thrombin formation; and a bleeding diathesis due to abnormalities in platelet function and coagulation cascade deficiencies (Jalal et al, 2010).

In the early stages of CKD, there is an association with a pro-thrombotic tendency, which in the advanced stages of CKD, such as end-stage renal failure (ESRF), manifests as a thrombotic state as well as a bleeding diathesis (Jalal et al, 2010).


2.  CKD patients on anticoagulation:

In patients with advanced CKD, the risk of bleeding from further anticoagulation and the baseline state of having a bleeding diathesis must be weighed up against the benefits of being anticoagulated.  The benefit should outweigh the harm.  Validated risk stratifying tools have been developed to assess the risk of thromboembolism and the risk of haemorrhage if anticoagulated.  CHADS2 and CHA2DS2-VASc have been developed to assess the risk of stroke in the presence of various comorbidities and AF in the general population.  HEMORR2HAGES, Outpatient Bleeding Risk Index (OBRI) and HAS-BLED are bleed risk scoring systems however HAS-BLED has not been validated in renal patients.


3.  Peri-operative bridging anticoagulation:

  • Risk assessment:

In the peri-operative situation, the risk of thrombosis if anticoagulation is completely withdrawn versus using “bridging” anticoagulation – which is often shorter acting and allowing for tighter control, with room for more rapid reversibility immediately pre- and post- operatively – must be weighed up against each other.

  •  Type of bridging anticoagulation:

The traditional method of peri-operative anticoagulation has been to use intravenous unfractionated heparin which, despite being shorter acting and allowing for tighter/closer manipulation, is not without its complications.   

  •  The complications with IV unfractionated heparin include the following.  It is:

labour intensive for the health care professional as it requires frequent blood tests and dose manipulations till the therapeutic APTTr is achieved.

Restrictive to the patient as they are connected to a continuous infusion.

It requires co-ordination between the ward team and the theatre teams in order to allow the infusion to be stopped 4-6 hours pre-operatively and thus allow the clotting to normalise minimising the risk of intraoperative bleeding.

Often requires a longer pre-operative admission period so that the patient can be converted and stabilised onto intravenous heparin.

Associated with risk of developing type 2 heparin-induced thrombocytopenia and thrombosis (HITT) – which is rare but occurs due to the development of platelet factor 4 antibodies that cause the over activation of platelets and subsequent thrombus formation.

  • Alternatives to unfractionated heparin:

Alternatives to unfractionated heparin include low molecular weight heparins (LMWH) such as dalteparin and enoxaparin, heparinoids such danaparoid, and direct thrombin inhibitors such as hirudin, bivalirudin, argatroban and dabigatran. 

Many of the LMWH and heparinoid alternatives to unfractionated heparin are primarily metabolised by the kidneys.  They have a longer half-life in patients with poor renal function especially when the GFR <30ml/min/m3, thus requiring a dose reduction in order to avoid accumulation and an increased risk of haemorrhage (Capodanno and Angiolillo, 2012; Grand’Maison et al, 2005).  If it is Trust policy to use a LMWH in advanced CKD, then one with a shorter half-life maybe preferred.  Dalteparin is a favourable choice, due to its half-life of 2.4 hours, compared to 4.3 for enoxaparin (White and Ginsberg 2003).  

  • Monitoring the therapeutic range of heparin

Heparins work by binding to and enhancing the anticoagulant properties of antithrombin, primarily via clotting factors Xa and Thrombin (IIa)(Barras 2013).  Anti-factor Xa levels can be used to determine the therapeutic range for unfractionated and LMWH.  It is also a more accurate test than the APTTr but costs more and is not routine available. 

  • Limited data on optimal anticoagulant in advanced CKD population

**Having discussed these options for management of anticoagulation, it is important to recognise that there is a paucity of evidence from randomised controlled trials as to gold standard therapy14.  This is partly due to the fact that patients with renal impairment are often excluded from studies.



  1. Barras M (2013). Anti-Xa assays. The Australian Prescriber. June. 36;3 pp 98-101.
  2. Capodanno D, Angiolillo DJ (2012). Antithrombotic therapy in patients with chronic kidney disease. Circulation. May 29;125(21):2649-61
  3. Grand'Maison A, Charest AF, Geerts WH. (2005)  Anticoagulant use in patients with chronic renal impairment. Am J Cardiovasc Drugs. 5(5):291-305.
  4. Jalal DI, Chonchol M, and Targher G. (2010)  Disorders of hemostasis associated with chronic kidney disease.  Semin Thromb Hemost. Feb;36(1):34-40.
  5.  UHB clinical guidelines (2012). IV Unfractionated Heparin. Clinical Guidelines Procedure Document Control Number: CG003. Other similar guidelines are available in different trusts and online.
  6. White RH, Ginsberg JS (2003). Low-molecular-weight heparins: are they all the same? British Journal of Haematology.  Apr;121(1):12-20
  7. Wells PS, Forgie MA, Simms M, Greene A, Touchie D, Lewis G, Anderson J, Rodger MA. (2003) The outpatient bleeding risk index: validation of a tool for predicting bleeding rates in patients treated for deep venous thrombosis and pulmonary embolism.  Arch Intern Med. Apr 28;163(8):917-20. 


Appendix 1. example guideline (9)

Initial dose

Bolus: 80 units/kg (max 5000 units)

Then 15 units/kg/hr

APTTR < 1.2 x control

Additional bolus: 80 units/kg

Rate increase by 0.3 mL/hr

APTTR 1.2-1.5 x control Rate increase by 0. 2mL/hr
APTTR 1.5-2.4 x control NO CHANGE
APTTR 2.5-3.0 x control Rate decrease by 0.1 mL/hr
APTTR 3.1-5.0 x control

Stop infusion for 1 hour. 

Rate decrease by 0.3 mL/hr

APTTR <5.1 x control

Stop infusion for 2 hours

Rate decrease by 0.5 mL/hr

APTTR = activated partial thromboplastin time ratio













Back To Index