Systemic capillary leak syndrome: a rare but potentially life-threatening cause of protein loss and oedema in B cell prolymphocytic leukaemia

Abstract

A 64-year-old man had a several year history of B prolymphocytic leukaemia (PLL) which behaved indolently and had not required any treatment. Five years after diagnosis, he developed hypoalbuminaemia associated with severe lower-limb oedema, consistent with systemic capillary leak syndrome (SCLS). He recovered spontaneously but went on to have three further increasingly severe and protracted episodes over the subsequent 18 months. There was no identifiable precipitating factor for these episodes, but his peripheral lymphocyte count continued to increase slowly. The start of treatment for his PLL with chemoimmunotherapy was followed by a rapid resolution of residual oedema and normalisation of serum albumin. He has had no further attacks of SCLS in the 14 months since he started therapy for PLL. SCLS is a rare consequence of haematological malignancy which may show an excellent response to treatment of the haematological disease.

Background

Systemic capillary leak syndrome (SCLS) is rare but potentially life threatening. It is often idiopathic, but there have been case reports of its association with malignancy or infection. Our case highlights this association and demonstrates that treatment of the underlying disease can lead to resolution of the capillary leak.

Case presentation

A 64-year-old man presented to the haematology clinic with an isolated mild lymphocytosis, which was found to be due to B cell prolymphocytic leukaemia (B-PLL). Other than previous Barrett’s oesophagus under surveillance, he was very well and active.

His B-PLL was negative for deletion of the TP53 gene and behaved indolently, with a lymphocyte doubling time of approximately 2 years. He remained asymptomatic for 5 years, without complications relating to lymphoproliferation or bone marrow infiltration.

Five years from diagnosis, however, he developed hypoalbuminaemia with associated haemoconcentration and oedema to up his hips, which developed over about 7 days. This was accompanied by malaise and a retro-orbital headache. In retrospect, the patient believed he had experienced multiple short, milder episodes of these symptoms for several years prior to developing oedema. CT demonstrated only small-volume lymphadenopathy and mild splenomegaly. Bone marrow biopsy indicated 60%–70% infiltration with B cell prolymphocytes but normal maturation of other cell lines. The full blood count exhibited a severe lymphocytosis of 101×10⁹/L with a mild, isolated anaemia (haemoglobin 115 g/L).

After slowly recovering spontaneously over 3 months from the onset of his first episode, he went on to have three further episodes over a further 18 months, each more protracted than the last and with a less complete recovery (figure 1). Later episodes were also accompanied by hypogammaglobulinaemia associated with respiratory tract infections. As the patient felt well in between these periods of low albumin and oedema and had successfully avoided chemotherapy for nearly 7 years since diagnosis, he did not initially wish to commence any therapy for PLL. On each occasion, the development of hypoalbuminaemia and his subsequent recovery from it occurred without any medical intervention apart from intermittent use of diuretics for symptomatic treatment of oedema.

Figure 1.

Serial lymphocyte count (10⁹/L), serum albumin (g/L) and serum haematocrit (L/L) levels measured from the first symptomatic episode, 5 years following the patient’s diagnosis, to instigation of treatment and the time of writing. The lymphocyte count rose steadily over time without cytoreductive treatment to prevent expansion of the prolymphocytic leukaemia (PLL) clones. During three symptomatic episodes of malaise, headache, weaknesss and oedema, there was a significant fall in serum albumin, intially accompanied by a rise in haematocrit. In later episodes, the rise in haematocrit became less pronounced and generally fell as the PLL gradually impeded erythropoiesis. The first cycle of chemotherapy not only induced rapid resolution of the lymphocytosis but instigated the return of serum albumin and haematocrit levels to the normal expected range for the first time since the first episode. Normal levels have been maintained since.

Differential diagnosis

The differential diagnosis of severe oedema and hypoalbuminaemia was considered in detail.

The patient had a normal diet with a stable weight, but given the history of Barrett’s oesophagus, he underwent an oesophagogastroduodenoscopy. This failed to identify any cause for malabsorption or protein-losing enteropathy.

Echocardiogram was unremarkable, and there was no evidence for liver or renal failure. Multiple urine samples did not suggest renal protein loss.

There were no accompanying features to suggest sepsis, anaphylaxis or angioedema. Therefore, the diagnosis of exclusion SCLS secondary to leukaemia was considered.

Treatment

As the sensitivity of B-PLL to chemotherapy is uncertain, and the patient felt well between episodes, he was keen to delay treatment for as long as possible. Given the lack of an alternative explanation for the increasingly problematic protein loss, however, he later agreed to receive chemoimmunotherapy with fludarabine, cyclophosphamide and rituximab (FCR), starting approximately 7 years after diagnosis. He received a total of six cycles of FCR at monthly intervals.

Outcome and follow-up

After a single cycle of FCR, the patient reported feeling the best he had been since the episodes started 2 years prior, and his albumin climbed to normal levels (figure 1). He had no significant treatment-related toxicity and remains asymptomatic 1 year after completion of treatment. His blood count is normal, spleen size has reduced from 17 cm to 12 cm, and serum albumin has remained normal.

Discussion

B-PLL is a very rare B cell neoplasm characterised by infiltration of the bone marrow and spleen by B lymphocytes and a lymphocytosis in the peripheral blood. Patients may initially be asymptomatic but systemic symptoms such as night sweats, weight loss, fatigue or fevers are common. Treatment is indicated in the presence of these symptoms or if there is a rapidly rising white cell count, massive splenomegaly or evidence of bone marrow insufficiency due to the PLL.

SCLS is a reversible but potentially life-threatening phenomenon of capillary hyperpermeability, classically characterised by haemoconcentration (increased haematocrit, thrombocytosis and leukocytosis), hypoalbuminaemia and eventually hypotension in the absence of other causes of shock.¹ Massive third spacing of fluids leading to hypovolaemic shock may develop, with or without a prodromal illness of fevers, flushing and gastrointestinal upset. There is increasing risk of venous thromboembolism with progressive haemoconcentration. As vascular permeability spontaneously reverses, patients are at risk of flash pulmonary oedema secondary to overzealous intravenous fluid resuscitation. The frequency, duration and severity of episodes can vary widely between patients, but are usually stereotyped for each individual, feeling well in between.

First described in 1960, there have been fewer than 500 cases reported in the literature.¹ Although SCLS can be idiopathic, identified causes include trauma, sepsis, burns and cardiopulmonary bypass. There have been a number of case reports linking SCLS to cancer or anticancer treatment.² In the context of malignancy, it is most often attributed to therapies such as interleukin 2, granulocyte-colony stimulating factor and haematopoietic stem cell transplantation (both autologous and allogeneic).³ However, it can result from the cancer itself, either at presentation or later in the disease course. Most often it has been associated with haematological malignancies, including myelomas, lymphomas and leukaemias.³

The basic pathogenesis of SCLS remains largely unknown but appears to involve inflammation-driven damage to capillary endothelium resulting in loss of intravascular fluid and proteins into the interstitium. Pro-inflammatory mediators implicated in driving SCLS include Tumour Necrosis Factor (TNF), IL-1, IL-2 and IL-8, together with Vascular Endothelial Growth Factor (VEGF), which may play a role in increasing vascular permeability.⁴

Oedema due to SCLS is classically diuretic-resistant and aggressive vascular re-filling risks precipitating complications of fluid overload.¹ The mainstay of treatment rests with identifying and addressing any underlying cause. In patients with malignancy, this entails starting appropriate anticancer therapy, or stopping the offending treatment.

Other treatment strategies to improve disease course are yet to be identified. Immunomodulatory therapies studied thus far without demonstrable benefit include corticosteroids, intravenous immunoglobulins, plasmapheresis, beta-2 agonists and theophylline.⁴ Given the rarity of the condition, high-quality data is unsurprisingly lacking. The prognosis is uncertain and depends on the underlying cause.

Patient’s perspective.

• When I was diagnosed with B cell prolymphocytic leukaemia (B-PLL), I was told that my cancer was incurable and aggressive. However, I remained mostly well and symptom-free for 5 years.

• Three years ago, I began to have episodes of fluid retention, during which I was very weak and virtually immobile. Between episodes, with no treatment my strength returned, the majority of the fluid went and I resumed my running and gym classes.

• The haematologists were uncertain if these episodes were related to my B-PLL, but my blood counts continued to rise so chemotherapy was necessary.

• By day 2 of the first chemotherapy cycle, all of the remaining fluid had gone and I felt loads better—almost as good as 8 years prior to this. For it to happen so quickly and dramatically, I felt that I had taken part in a miracle.

• After 6 months of chemotherapy, my health and blood counts were all back to normal and I had no side effects. I continue to feel well.

Learning points.

• This is the first time that systemic capillary leak syndrome (SCLS) has been reported in the context of pro-lymphocytic leukaemia, to the best of our knowledge.

• This rare but often life-threatening phenomenon is undoubtedly under-recognised, considering the heterogeneity of presenting features among cases, of which many are present in multiple other causes of shock.

• SCLS should always be considered in the differential diagnosis of hypoalbuminaemia, fluid extravasation and haemoconcentration once more common causes are excluded.

• Where SCLS is suspected without an apparent cause, investigation for a culpable malignancy is essential.

• As SCLS can emerge at any point in the natural history of malignancy, and is often attributable to anticancer therapies, judicial observation of the temporal relationship between clinical features and treatment course is required.