8.3 Anaemia, cytopenias, and thrombosis in palliative medicine

The illnesses, malignant or benign, that bring patients to a palliative care setting are often complicated by haematological problems such as anaemia, bone marrow failure, disseminated intravascular coagulopathy, and thrombosis. Supportive therapy for these problems can provide gratifying relief of symptoms and improvement in overall quality of life. This chapter emphasizes a practical approach to assessment and therapy intended to maximize quality of life.

Anaemia is present in 77% of men and 68% of women receiving palliative care (Dunn et al., 2003). Typical symptoms associated with a sudden drop in haemoglobin as seen with acute blood loss include tachycardia, orthostatic hypotension, and dyspnoea. If the onset of anaemia is more gradual, compensatory mechanisms can help lessen these symptoms by increasing the cardiac output, increasing plasma volume and shifting the haemoglobin dissociation curve.

Patients can often remain symptom free if mild to moderate anaemia evolves gradually. More commonly, however, a diminished overall well-being is seen manifesting as fatigue, decreased exercise capacity, and decreased appetite. Some patients can also complain of dizziness, headache, syncope, tinnitus, vertigo, and impaired cognitive function. Patients with underlying cardiac disease are more susceptible to anaemic symptoms as they are unable to compensate as readily (Mercadante et al., 2000).

Determining the cause of anaemia in the palliative care setting can be challenging, since the aetiology is often multifactorial. Disease-related causes include bone marrow infiltration, blood loss, haemolysis, and anaemia of chronic disease. Cancer-related treatment can also result in anaemia such as myelosuppression from chemotherapy and treatment-related myelodysplastic syndrome. Concomitant factors such as folate deficiency from malnutrition and gastrointestinal resections can also contribute to anaemia (Mercadante et al., 2000).

Anaemia of chronic disease (ACD) is a hypoproliferative anaemia caused by an immunological reaction to the presence of inflammation and malignancy. It results from the release of multiple cytokines from T cells and monocytes including interferon-gamma (INFγ), tumour necrosis factor α (TNFα), interleukin (IL)-1, IL-6, and IL-10. These cytokines stimulate the uptake into and storage of iron in macrophages and monocytes while also preventing the export of iron out of these cells. Hepcidin, a hepatic peptide stimulated by IL-6 and lipopolysaccharide, contributes to the retention of iron within the reticuloendothelial system. This results in a paradoxical situation of iron-deficient erythropoiesis occurring in a marrow replete with iron. Simultaneously, the cytokines also suppress the ability of the kidneys to produce erythropoietin, enhance red blood cell membrane damage, and prevent the differentiation and proliferation of red cell progenitors in the marrow. Together, erythropoiesis is reduced resulting in anaemia (Weiss and Goodnough, 2005).

The diagnosis of ACD is often based on the exclusion of other forms of anaemia but should always be considered in the palliative care setting. It is present in almost 50% of men and over 70% of women (Dunn et al., 2003). The anaemia is typically normocytic, normochromic, and is usually mild to moderate (80–95 g/L). The reticulocyte count will be low, reflecting a reduced marrow output. Serum iron studies can be helpful in differentiating ACD from iron-deficiency anaemia. Serum iron, total iron binding capacity (transferrin), and iron saturation are all low in ACD while total iron binding capacity is elevated in iron deficiency. Serum ferritin is a reflection of total body iron storage, and since iron stores within the reticuloendothelial system is abundant, ferritin is normal or elevated in ACD (Weiss and Goodnough, 2005).

Acute and chronic blood loss is common in a palliative patient, especially in gastrointestinal, head and neck, respiratory, uterine, and urinary cancers. Bulky sarcomas, hepatomas, melanomas, and ovarian cancers can bleed into the malignant masses as well (Mercadante et al., 2000). Even the loss of a few millilitres of blood a day can result in iron deficiency over time. In the early stage of iron deficiency, iron stores are low resulting in a reduced ferritin but normal serum iron, iron saturation, and haemoglobin levels. As iron deficiency worsens, serum iron and iron saturation falls but the haemoglobin is preserved. Anaemia is the end result of severe iron deficiency with the presence of microcytic hypochromic red blood cells. Abnormally shaped cells like target cells and pencil cells are seen as the severity of the iron deficiency advances (Beutler, 2010). In most cases, the chronic blood loss is obvious from the presence of blood in bodily fluids, but occasionally tests for occult blood or endoscopic examination is necessary to determine the presence and site of the bleeding.

The first step in management of haemorrhage is to control the bleeding lesion if practical (see Chapter 8.7). This may be accomplished surgically, endoscopically, or by radiation of the bleeding lesion (Videtic, 2013). Iron supplementation should be started concomitantly if iron deficiency is present. Treatment can be provided either orally in the form of simple iron salts or parentally as iron–carbohydrate complex. The oral route is preferred as parenteral iron is associated with more severe adverse effects. The dosage for adults should provide 150–200 mg of elemental iron a day. This is ideally taken orally in three or four separate doses 1 hour before meals. Mild gastrointestinal side effects like nausea, heartburn, constipation, or diarrhoea are common. For these patients, changing to another form of iron or reducing the dose initially can be helpful (Beutler, 2010).

Parenteral iron is useful for patients who are intolerant of oral iron, have intestinal malabsorption issues, or may be losing iron more quickly than can be replaced with oral supplementation. The total dose required can be calculated by this formula:

Dose of iron (mg) = whole-blood haemoglobin deficit (g/dL) × body weight (lb)

Iron sucrose contains 20 mg of elemental iron per millilitre and is recommended by the manufacturer to be administered at a maximum of 100 mg three times weekly. However, it has been safely given to chronic kidney disease patients at doses of 500 mg over 3 hours on 2 consecutive days. Typical adverse effects include hypotension, cramps, nausea, headache, vomiting, and diarrhoea (Beutler, 2010).

Iron dextran contains 50 mg of elemental iron per millilitre and is recommended by the manufacturer to be administered at a maximum of 100 mg per dose preceded by a 0.5 mL test dose. Larger doses are frequently used and considered safe with an increase in minor adverse effects. Iron dextran can result in a severe anaphylactic reaction occurring in less than 1% of patients. It is not dose dependent so can occur within the first few millilitres of the infusion. Typical presentation includes difficulty breathing, a choking sensation, becoming sweaty and anxious, nausea, and vomiting within the first few minutes of starting the iron dextran infusion. The infusion should be stopped immediately, resuscitation initiated, and epinephrine (adrenaline) be readily available. For patients with severe anaemia, transfusions may be necessary to help boost their haemoglobin. As each millilitre of blood contains 1 mg of elemental iron, blood transfusions will also help boost iron stores as well. Transfusion therapy is discussed later in this chapter in ‘Transfusions in the palliative care setting’ (Beutler, 2010).

Cancer cachexia is seen in about half of all cancer patients, characterized by anorexia and loss of adipose tissue and skeletal muscle mass (Suzuki et al., 2013), leading to progressive nutritional deficiency (Segura, et al., 2005). The incidence of vitamin B₁₂ deficiency was similar to that of the general elderly population at 7% but low serum folic acid level was significantly increased at 22%. Folic acid levels have been found to be an insensitive marker of occult folate deficiency, especially in those with weight loss. For example, 58% of patients with major weight loss in the palliative care setting have been found to have features of megaloblastic anaemia with a normal folate level (Dunn et al., 2003).

Folic acid is critical in the metabolism and DNA synthesis of all cells, especially cells that are highly proliferative including haematological cells. Fruits and vegetables are rich in folic acid and the recommended dietary intake for an adult is 0.4 mg. When folic acid intake is reduced to less than 5 micrograms, megaloblastic anaemia ensues in approximately 4 months. Folic acid is absorbed in the duodenum and proximal jejunum so patients with intestinal resections or masses involving this area are at high risk of developing megaloblastic anaemia.

Megaloblastic anaemia results in enlarged megaloblastic bone marrow precursor cells and consequently enlarged red blood cells increasing the red cell volume. Hypersegmented neutrophils with the presence of more than the usual three to five nuclei lobes are classically described in megaloblastic anaemia. All cell lines can be affected resulting in anaemia, thrombocytopenia, and/or neutropenia.

Dietary supplementation can be given orally, usually at a dose of 1 to 5 mg/day. Typically 1 mg a day is adequate to correct anaemia even if malabsorption is present (Green, 2010).

While anaemia from metastatic cancer is most commonly related to anaemia of chronic disease, iron deficiency, or other nutritional deficiencies, bone marrow infiltration from metastatic disease can occur. All malignancies can metastasize to the marrow, but the most common are from the lung, breast, and prostate. While not common, the classic clinical picture is that of a leucoerythroblastic blood picture with the presence of immature nucleated red cells, myeloid white cell precursors, and teardrop-shaped red cells. A bone marrow biopsy can help to confirm the diagnosis but since the majority of metastatic malignancies with marrow involvement are incurable, it is oftentimes not clinically necessary (Agarwal and Prchal, 2010).

Mild to moderate marrow infiltration is often times asymptomatic with minimal blood count changes. With more significant marrow involvement, anaemia can be more severe and accompanied by an elevated white blood cell count. Platelets can be low, normal, or elevated. Treatment is focused at managing the underlying disease and symptom control.

Patients with haematological malignancies (Franchini et al., 2013) and bone marrow failure disorders like myelodysplastic syndrome (Foran and Shammo, 2012) have diseased marrows that are unable to produce adequate blood cells, often resulting in pancytopenia. While thrombocytopenia can result in easy bruising, it is unlikely to result in spontaneous bleeding at a platelet count of more than 20 × 10⁹ or in fatal intracranial haemorrhage at a platelet count of more than 5 × 10⁹ (Franchini et al., 2013).

Neutropenia in the palliative setting is most commonly due to bone marrow failure from myelosuppressive chemotherapy, disease infiltration of the bone marrow, or from intrinsic bone marrow failure. The definition of neutropenia is defined as an absolute neutrophil count (ANC) less than 1500 cells/microlitre with severe neutropenia being ANC less than 500 cells/microlitre. Patients with neutropenia are at high risk of developing severe bacterial and fungal infections. The Infectious Diseases Society of America defines fevers in neutropenic patients as a single oral temperature of greater than 38.3° Celsius or a sustained temperature of greater than 38.0° Celsius for more than 1 hour (Freifeld et al., 2011).

Patients at highest risk of severe febrile neutropenia are those with ANC less than 500 cells/microlitre for more than 7 days. Those with active comorbidities or significant hepatic or renal dysfunction are also considered to be high risk of severe bacterial infections.

Patients with prolonged neutropenia with ANC less than or equal to 100 cells/microlitre for more than 7 days should be considered for fluoroquinolone prophylaxis with levofloxacin or ciprofloxacin. In patients with fevers and neutropenia, the decision about management is dependent on the patient’s values and disease stage. If life prolongation is deemed appropriate, then a discussion about the extent of investigation and management should be discussed on a per case basis.

Standard investigations for febrile neutropenia include complete blood cell (CBC) count with differential leucocyte count and platelet count, serum electrolytes, hepatic enzymes, total bilirubin, and creatinine. At least two sets of blood cultures are recommended, one from each lumen of a central venous catheter and one from a peripheral vein. Cultures from other sites should be taken as appropriate and a chest radiograph should be performed if respiratory findings are present.

High-risk patients with febrile neutropenia should be hospitalized for intravenous empiric antibiotic therapy with an anti-pseudomonal beta-lactam agent such as cefepime, meropenem, imipenem–cilastatin, or piperacillin–tazobactam. Vancomycin should be added for empiric treatment of suspected catheter-related infection, skin or soft-tissue infection, pneumonia, or haemodynamic instability. If fevers persist or new infectious symptoms develop, additional investigations and alterations in the antimicrobials may be necessary. Management should be guided by clinical microbiological data derived from investigations undertaken. Treatment is generally continued until neutrophil recovery (ANC ≥500 cells/microlitre) plus whatever duration is appropriate for the site of infection and organism identified. For patients without neutrophil recovery, antimicrobials should be continued until all signs and symptoms of a documented infection have resolved.

While hematopoietic growth factors such as granulocyte colony-stimulating factor (G-CSF) are frequently used to prevent infections associated with chemotherapy-related neutropenia, it is generally not used to prophylactically raise neutrophil counts in the palliative population. In addition, G-CSF is not recommended for routine use as adjuncts to antimicrobials in febrile neutropenia either. While the days of neutropenia, duration of fever, and length of hospital stay have been minimally reduced in some trials, no survival benefit has been seen.

Disseminated intravascular coagulopathy (DIC) results from an overproduction of procoagulants, which overwhelms the anticoagulant mechanism resulting in a systemic generation of intravascular microthrombi. Two-thirds of DIC cases are a result of underlying severe infection or malignancy (Levi and Seligsohn, 2010). The incidence of DIC in solid tumours is 7% but is much higher in acute leukaemia, up to 15–20%, which can increase further with chemotherapy treatment. Patients with acute promyelocytic leukaemia have a very high risk of DIC with more than 90% diagnosed at some point in the course of the disease (Levi, 2009). The risk of DIC increases with advanced age, more advanced stages of cancer, and the use of chemotherapy or anti-oestrogen therapy. Triggers for DIC include sepsis, immobilization, and liver metastases (Levi, 2013).

The production of intravascular microthrombi results in multiorgan failure, while the consumption of platelets, fibrinogen, and other coagulation factors increases bleeding. Hence, this results in a contradictory increased risk of both bleeding and clotting. Solid organ tumours are more prone to thrombosis while leukaemia patients are more prone to bleeding. Mucocutaneous bleeding is classic with cutaneous purpura, haemorrhagic bullae, oral and intestinal mucosal bleeding, and bleeding from intravenous sites. Other sites including intracranial haemorrhage, pulmonary haemorrhage, and haemorrhagic necrosis of the adrenals can also be seen. Ischaemic organ failure from microthrombi classically results in renal insufficiency, liver failure, intestinal ischaemia, cutaneous focal necrosis, acral gangrene, and respiratory failure. With severe DIC, cardiovascular shock can ensue (Levi and Seligsohn, 2010).

While DIC from infection and acute leukaemia typically presents very acutely, solid-organ cancer-related DIC can be more subacute and less fulminant with a more chronic presentation (Levi, 2009).

Laboratory findings classically reveal thrombocytopenia and prolonged prothrombin time/international normalized ratio (INR) or activated partial thromboplastin time and a suppressed fibrinogen level. Since fibrinogen acts as an acute-phase reactant and is elevated in the presence of infection or malignancy, it can remain within normal range despite an active DIC process. The most important diagnostic tool is to maintain a high clinical suspicion for DIC in patients with bleeding or clotting symptoms (Levi, 2013).

As the mainstay of treatment for DIC is supportive care while treating the underlying disease, its management becomes very difficult in palliative patients with incurable cancer. Prognosis is poor with a high mortality rate, especially with more severe DIC. Mortality rates range from 31% to 86% depending on the severity of DIC (Levi, 2013).

Supportive care of DIC in palliative cancer patients involve cardiovascular support and replacement of coagulation factors by using blood products. Cryoprecipitate contains large quantities of fibrinogen and factor VIII while fresh-frozen plasma helps replace the remaining coagulation factors. Platelet transfusions are sometimes necessary as well. In patients with clotting as the main presentation, anticoagulation should be initiated cautiously as the risk of bleeding is high (Levi, 2013).

The most common indication for blood transfusion in the setting of advanced cancer is fatigue and dyspnoea. Unfortunately, data regarding the relationship between fatigue and dyspnoea and the presence of anaemia remains controversial. In addition, there are currently no randomized controlled trials assessing the effectiveness of blood transfusions in this population. A Cochrane review found a subjective improvement in fatigue and dyspnoea at a rate of 31% to 70% with red blood cell transfusions (Preston et al., 2012).

Blood products are scarce and costly and hence use in the terminally ill evokes ethical discussions involving the principles of autonomy, beneficence, nonmaleficence, and justice. There are currently no guidelines on transfusions for palliative care patients, and each case should be assessed on an individual basis (Smith et al., 2013).

Red cell transfusion rapidly improves tissue oxygenation in the presence of anaemia or acute blood loss. Unfortunately, there is no evidence regarding red cell transfusions in the palliative population. Among non-palliative care patients, a Cochrane review comparing restrictive to liberal transfusion strategies did not find any adverse health outcomes and concluded that in patients without acute coronary artery disease, red cell transfusions is generally not necessary until haemoglobin levels approach 70–80 g/L as long as there is no evidence of bleeding (Carson et al., 2012).

Thrombocytopenia in the palliative setting is most commonly seen in patients with haematological malignancy and bone marrow failure. They can also be seen in advanced liver disease and splenomegaly. Platelet transfusions are scarcer and have a shorter duration of effect in raising the platelet count. They are generally indicated in patients with thrombocytopenia and active bleeding. Extrapolating from surgical patients, a platelet count above 50,000/microlitre should be adequate for bleeding control. Prophylactic platelet transfusions are considered for patients with platelet count less than 10,000/microlitre as they are at higher risk of severe spontaneous bleeding (Kaufman, 2013).

Blood cell transfusions like red cells and platelets have associated with it potential adverse events including transmission of blood-borne viral, bacterial, parasitic, and prion infections as well as transfusion reactions. With improved blood donor screening and blood testing strategies, the associated risks of receiving a transfusion-transmitted infection can been significantly reduced (Katz and Menitove, 2013).

Transfusion reactions range from frequent to rare, and from mild to severe (see Table 8.3.1). The most common are febrile reactions and mild allergic reactions. Febrile non-haemolytic transfusion reactions are likely related to cytokines released in response to antibody–leucocyte or antibody–platelet interactions. Fevers usually respond to antipyretics such as acetaminophen, which does not have the anti-platelet effect of aspirin and non-steroidal anti-inflammatory drugs. Prophylactic antipyretics are not necessary unless there is a history of febrile reactions. Rigor can be ameliorated with a small dose of meperidine, which can stop the symptom almost immediately. In patients with persistent febrile or allergic reactions despite premedication, hydrocortisone 1–2 hours before a transfusion can be used. Leucoreduction of blood products if available can significantly reduce the frequency of febrile reactions as well (Choate et al., 2013).

Mild allergic transfusion reactions are believed to be a result of plasma proteins in the blood products to which the patient has formed antibodies. They are quite common, occurring in approximately 1% of all transfusions. At the onset of pruritus and hives, the transfusion should be stopped and a dose of 25–50 mg of diphenhydramine given. Once the rash improves and the patient fells well without signs of fever, chills, or vasomotor instability, transfusion of the same unit can be re-initiated. In patients with frequent mild allergic transfusion reactions, pre-medication with diphenhydramine can be utilized (Choate et al., 2013).

Transfusion-associated circulatory overload (TACO) results from excess fluid accumulation in the lungs. Transfused blood products rapidly increase a patient’s intravascular volume and can exceed the ability of the cardiovascular system to compensate. Patients with pre-existing cardiovascular disease are especially at risk, and transfusing at a rapid rate in the absence of acute haemorrhage is an additional risk factor. Reducing the rate of transfusion in those at risk for fluid overload and the administration of diuretics before a transfusion can be preventative (Choate et al., 2013).

Acute intravascular haemolytic transfusion reactions are a result of the acute haemolysis of ABO incompatible red blood cells in a patient with the corresponding antibody. The reactions are very rapid, occurring within minutes after initiating the transfusion. The transfusion needs to be stopped immediately, cardiorespiratory support initiated, and a crystalloid solution infused to maintain a good urine output (Choate et al., 2013).

Erythropoietin is a hormone naturally produced by the kidney and is an essential growth factor for red cell progenitors in the bone marrow. It is traditionally used to manage anaemia associated with advanced renal disease (Marks et al., 2013) but has become increasingly common in cancer-related anaemia as well. The use of erythropoietin-stimulating agents (ESAs) like epoetin and darbepoetin in patients on myelosuppressive chemotherapy has been shown to reduce the need for red cell transfusions, but can also increase the risk of hypertension, thromboembolic events, and death. There is some suggestion that ESAs may improve quality of life but the data is not definitive. Also controversial is whether ESAs worsen tumour control (Tonia et al., 2012).

The American Society of Hematology/American Society of Clinical Oncology 2010 guideline recommends that ESAs be considered in patients undergoing myelosuppressive chemotherapy who have a haemoglobin level less than 100 g/L and cautions against ESA use under any other circumstances of cancer-related anaemia. They also recommend using the lowest ESA dose possible and targeting a haemoglobin level to avoid transfusions. If no response is seen after 6–8 weeks, the ESAs should be discontinued (Rizzo et al., 2010).

Venous thromboembolism (VTE) is a category of disease comprised of deep vein thrombosis (DVT) and pulmonary embolism (PE). Although venous thrombosis can occur in any site, the vast majority of what we know is focused on lower extremity DVT and PE. VTE is associated with significant morbidity and mortality. The link between cancer and thrombosis has long been established and was first described by French internist Armand Trousseau in 1865. Cancer patients experience a higher rate of VTE (Blom et al., 2006). It is estimated that one in 1000 adults per year develop VTE in the general population. About 15% of patients with cancer develop symptomatic VTE, 15–20% of patients with VTE have cancer, and about 10% of patients presenting with idiopathic VTE will develop cancer within a year (Lee, 2004). Cancer is also associated with both a higher rate of anticoagulant failure and a two- to sixfold higher rate of major bleeding (Palareti et al., 2000; Prandoni et al., 2002). Cancer patients with thrombosis also have a shortened life expectancy (Sorensen et al., 2000; Chew et al., 2006).

The pathophysiology of thrombosis in cancer is often complex and multifactorial but can be generally thought of in a simple framework developed in 1856 by the German physician Rudolf Virchow. Together, venous stasis, endothelial injury, and hypercoagulability are now commonly known as Virchow’s triad. Large lymphadenopathy or tumours causing local compression, paralysis from spinal cord compression, and hospitalization from complications of cancer can contribute to venous stasis. Many chemotherapeutic agents, surgical interventions, and central venous access can contribute to endothelial injury. Cancer causes various procoagulant changes in the blood, including, for example, an increase in circulating levels of tissue factor, and thrombogenicity often depends on tumour type and stage (Otten et al., 2004; Lee et al., 2006).

VTE continues to be a significant problem throughout a cancer patient’s care, from pre-diagnosis to diagnosis to treatment to palliation and death. Older studies carried out in a time when autopsies were done more frequently estimate that up to 50% of patients with cancer have VTE at the time of death. Despite recognizing this link between cancer and thrombosis, there is a paucity of data in the palliative care setting and there still remains the question of how aggressively we should try and prevent, diagnose, and treat thrombosis in the end stages of life.

To help answer this question, it is important to understand what treatment of VTE accomplishes for patients. The initial phase is targeted at decreasing symptoms (leg swelling and pain, chest pain, shortness of breath), preventing clot extension, preventing embolic events, preventing early recurrence, and decreasing upfront mortality. Longer-term treatment is targeted at decreasing the risk of recurrent VTE and mitigating post-thrombotic symptoms (Lee, 2004). In most patients, the risk of haemorrhage is outweighed by the thromboembolic risk in the first few weeks to months after a new VTE is diagnosed and, in the absence of major contraindications, prompt investigations and initiation of treatment is usually recommended. Duration of treatment hinges on the fine balance between thrombotic complications and haemorrhagic complications, both of which can change over time (Noble et al., 2008).

Treatment of VTE is based largely on anticoagulant therapy. Although new oral anticoagulant agents are being developed, none have currently been studied in the cancer or palliative setting. At present, the mainstay of treatment consists of administering a fast acting agent such as intravenous (IV) unfractionated heparin (UFH) or subcutaneous (SC) low-molecular-weight heparin (LMWH). This is frequently followed by initiation of oral warfarin therapy to target an INR of 2–3 for ongoing maintenance therapy. An overlap of about 5–7 days is required between the parenteral and oral medications as there is a delay in the onset of anticoagulant effect of warfarin.

More recent data has supported the ongoing use of LMWH rather than warfarin in cancer patients, particularly in those receiving active chemotherapy (Lee et al., 2005). Prophylactic anticoagulants for VTE prevention is mostly comprised of low-dose SC UFH or LMWH. All anticoagulants are associated with a risk of major or fatal haemorrhage. IV or SC modalities are additionally cumbersome for patients and may cause discomfort and decrease quality of life. Other considerations for those undergoing anticoagulant therapy include the need for continued monitoring of laboratory parameters including potentially INR, CBC count (to look for thrombocytopenia), and liver and kidney function tests.

The line between active treatment and palliation can often be unclear. Palliative care is targeted at symptom control, maintaining quality of life and psychosocial supports, and decreasing suffering. In the treatment and prevention of VTE, there are many overlapping features. Although anticoagulant therapy is in part used to decrease mortality, much is also used to decrease clot burden and improve symptoms (McLean et al., 2010). Investigation for VTE will often require a hospital visit and potential need for admission. Administration of anticoagulant therapy, as described earlier, is associated with its own challenges (Soto-Cardenas et al., 2008; McLean et al., 2010). Given this, it is prudent to individually tailor treatment plans and care plans for different patients. For example, for a patient who is palliative but still ambulatory and living independently at home, it is very reasonable to consider investigation for and treatment of newly diagnosed VTE, or prophylactic anticoagulation in case of temporary immobilization, as controlling the complications of VTE can help maintain their state of independence.

At the extreme of life, a terminally ill patient who is bedridden and approaching death may benefit more from symptom control only, regardless of aetiology, rather than being subjected to hospital transport for further investigation, uncomfortable administration of medications with frequent monitoring, and potentially the need for further intervention if complications such as bleeding occur (Tran, 2010).

Haematological complications are common and significant issues in cancer patients and those who eventually require palliative care. The complications of investigation and treatment make devising care decisions for such patients a challenging process. Frank and open discussions with each patient regarding the overall goals of care can help guide management. It is important to remember that a patient’s status and thus these very goals will change over time and constant re-evaluation is essential. This will help shape plans that avoid unnecessary testing and intervention, as well as avoid depriving a patient of the compassionate care needed in the end of life.