What we know and what we don’t know about catastrophic antiphospholipid syndrome

Abstract

Catastrophic antiphospholipid syndrome (CAPS) is a severe condition with high mortality. Since its description in 1992, an important effort has been made to improve and disseminate knowledge on CAPS. Most of our current knowledge comes from the studies performed using the CAPS Registry, a database created in 2000 to gather as many cases as possible in order to better define this disease. It has demonstrated that this condition has multiple faces and is often triggered by a precipitating factor that leads to a thrombotic microangiopathy and cytokine storm involving almost any organ of the body. Analysis of the CAPS Registry has also shown that patients receiving anticoagulation, glucocorticoids and plasma exchange and/or IVIG have a better prognosis. However, there are still many unresolved questions. In this review we summarize what is known and what is still a matter of research in this condition.

Introduction

APS is a systemic autoimmune disorder defined by the development of venous and/or arterial thromboses and pregnancy morbidity in patients with persistent circulating aPL [1]. Less than 1% of all APS patients develop a life-threating variant characterized by multi-organ thrombosis developing over a short period of time [2]. In 1992, Ronald A. Asherson described this very aggressive clinical variant of the syndrome and coined the term catastrophic APS (CAPS) [3]. Most of the knowledge on this condition has been provided by studying the patients included in the CAPS Registry. This is a database that includes cases published in the medical literature or reported directly to the CAPS Registry Project Group. It was created in 2000 by the European Forum on Antiphospholipid Antibodies, a group of experts devoted to performing international collaborative studies on aPL [4]. The CAPS Registry includes >700 cases from almost 200 published papers and 100 cases reported directly to the CAPS Registry Project Group. While significant progress has been made in understanding this syndrome, there are still several aspects that remain unknown. This article aims to review what we know and what we don't know about CAPS.

Epidemiology

CAPS is an extremely rare condition, but its current epidemiology is scarcely known. In the Europhospholipid Project, the largest epidemiological project on APS published to date, CAPS accounted for ≈1% of APS cases [2]. In this study, after 10 years of follow-up of 1000 patients with APS, only 9 cases developed CAPS [5]. This would indicate an incidence of 90 cases per 100 000 patients with APS per year. During this study, all cases occurred during the first 5 years of follow-up, and 5 of 9 patients (55.9%) died during the acute episode. It was the initial manifestation of APS in ≈50% of diagnosed patients [6]. Most patients with CAPS had primary APS, but some patients may have associated disorders, most often SLE, RA or other systemic autoimmune diseases. Unfortunately, due the extreme rarity of this disease, it has not been possible to perform other large studies to better clarify the exact incidence and prevalence of this disease.

Pathophysiology

The pathophysiological mechanisms involving CAPS are scarcely known. Today, CAPS is considered to have a multifactorial aetiology, including the interplay of genetic background and environmental factors, ending in a thrombotic storm. In general terms, <15% of general population patients with aPL develop thrombosis [7]. The reason why some patients with aPL do not develop thrombosis while others develop a single thrombotic event (classic APS) and others develop multiple vascular occlusions predominantly affecting small vessels in a short period of time (CAPS) remains a matter of research.

Although not clearly proven, some hypotheses have been proposed to explain the clinical manifestations observed in episodes of CAPS. Current understanding includes the notion of a two-hit hypothesis where several events take place in a sequential way leading to this catastrophic situation. This hypothesis states that a first event leads to a thrombophilic state where the production of circulating aPL is triggered in some subjects. Asherson and Shoenfeld [8] postulated the molecular mimicry theory, where they suggest that semblances of amino acid sequence found in several microorganisms and some peptides derived from β2-glycoprotein I (GPI), the so-called aPL co-factor, precipitate the production of aPL against β2-GPI. Skin, lung and urinary infections, as well as HIV and HCV infections, have been associated with an increase in aPL [9]. Additionally, a strong homology has been found between β2-GPI-related peptides and different common pathogens such as CMV, Haemophilus influenzae, Neisseria gonorrhoeae and tetanus toxoid [10, 11].

However, anti-β2-GPI antibodies do not bind to unstimulated endothelium in vivo [12]. Although Vega-Ostertag et al. [13] demonstrated that aPL are able to upregulate the production of tissue factor leading to a well-known thrombogenic factor, clotting will only take place in the presence of a second hit; this would be a thrombophilic state that complements the first thrombophilic condition leading to the thrombosis. In this sense, the presence of an environmental trigger (i.e. infections) has been reported in more than half of cases with CAPS [6].

However, this hypothesis is unable to explain why some patients develop only thrombosis in one site (classic APS) while patients with CAPS develop thrombosis simultaneously in different organs. It is well known that patients with sepsis have coagulation abnormalities [14]. Additionally, in 1998, Kitchens et al. [15] proposed that intravascular coagulation itself could increase the risk of developing another thrombosis. The authors suggested that the blood clot might promote thrombin formation and fibrinolysis would be impaired by an increase of plasminogen activator inhibitor.

Recently, the role of complement has been recognized in the pathophysiology of APS. Initially, it was more clearly shown in mouse models of obstetric APS [16, 17], but lately mouse models of thrombotic APS also underscore the role of complement in thrombotic APS [18]. In this sense, Stachowicz et al. [19] showed that blood clots from APS patients have a higher proportion of complement than clots from other patients. More recently, Chaturvedi et al. [20] were able to identify complement germline variants in 60% of patients with CAPS and in 51.5% with atypical haemolytic uraemic syndrome (HUS), while those were present in only 21.8% patients with thrombotic APS, 28.6% with SLE and 23.3% in controls. The authors suggested that patients with complement dysregulation due to genetic alteration might be predisposed to thrombosis when an environmental stressor takes place.

Conversely, some clinical manifestations might not be directly related to blood flow occlusion, but to cytokine overexpression in the ischaemic necrotic tissue, leading to the so-called cytokine storm.

Clinical features

CAPS mostly affects women (70%), mainly in the fourth decade of life, with a mean age of 38 years; however, cases have been described from birth to old age [6]. Nearly half of patients suffer from SLE or have at least some typical clinical or immunological manifestations of this condition. This association is more frequent in middle-aged patients, while it is less frequent in paediatric and elderly patients. Furthermore, some cases have been described in association with other autoimmune diseases. Interestingly, half of patients who develop the catastrophic variant have not previously presented other clinical manifestations of APS [6].

As a systemic disease, CAPS can affect any organ or system in the body. Therefore the clinical manifestations of this condition are very different depending on the organs affected by the thrombotic occlusions. Furthermore, some clinical manifestations are not driven by the thrombotic occlusion, but by the cytokine storm that develops from the inflammation that takes place. In this sense, clinical manifestations have been classically classified into those attributed to thrombosis itself and those attributed to the cytokine storm. However, sometimes it is difficult to differentiate if a clinical manifestation is attributable to one or the other cause and many times both pathways may work together [21]. Table 1 shows the prevalence of the clinical manifestations of patients included in the CAPS Registry [6]. CAPS patients frequently present with renal failure and variable degrees of hypertension, although hypotension does not exclude the diagnosis. Some patients present with proteinuria and sometimes with haematuria [6, 21–23]. The most common clinical renal manifestations are acute renal failure (45%), arterial hypertension (39%), proteinuria (52%) and haematuria (33%). Renal infarction is found in 8.7% of patients and renal vessel thrombosis in 4.8%. The most common pathological finding is acute renal thrombotic microangiopathy (TMA), found in 98% of samples.

Pulmonary manifestations are reported in two-thirds of cases. Lung thrombosis has been reported in up to 48.6% episodes of CAPS, presenting with dyspnoea and sometimes with pulmonary haemorrhage, while 28.3% develop diffuse alveolar haemorrhage (DAH). Recent studies have shown that the development of DAH is associated with laboratory features of TMA, pointing to a microangiopathic cause of DAH. Also, a significant relationship was found between triple aPL positivity and the finding of TMA in pathologic samples and, interestingly, was associated with hypocomplementemia. More rarely, some patients develop ARDS, classically attributed to the cytokine storm. Pulmonary emboli are associated with dyspnoea and sometimes the clinical picture is associated with pulmonary haemorrhage.

Central nervous system involvement has been reported in up to 56% of cases of CAPS, mainly due to stroke (40%) and/or encephalopathy (39%), manifested as consciousness deterioration. Encephalopathy could be ascribed to general hypoperfusion because of microthrombosis, generalized shock or cytokine storm. Many times it is not clear if brain involvement is attributable to encephalopathy or diffuse microangiopathy. Nevertheless, some patients present with neurological focal deficits with motor and/or sensitivity symptoms and an established stroke. Less often, physicians report of seizures and, when present, many times they are associated with other brain manifestations [6, 21, 23].

Heart involvement is also frequently reported (50%), mainly due to myocardial infarction (30%), angina, cardiac valvulopathy (28%) or Libman–Sacks endocarditis (13%), sometimes with cardiogenic shock. However, in some cases it is difficult to differentiate the cause of this cardiogenic shock with depressed ejection fraction between myocardial infarction due to microangiopathy, Takotsubo myocardiopathy, myocarditis or other causes of myocardial dysfunction. Performing a myocardial biopsy is frequently not possible and only very few cases report data on MRI and/or ventriculography.

When valve involvement is reported, the main valves affected are the mitral and aortic valves. This condition is mainly reported as valvular insufficiency and, rarely, it leads to the requirement of valvular replacement. Intracavitary thrombosis has seldom been reported in patients with CAPS. However, it would be interesting to have more transoesophageal echocardiographies performed to better characterize valve involvement in these cases.

Skin complications are also a frequent clinical issue in CAPS, reported in up to 47% of cases [6]. Often they are in the form of livedo reticularis/racemosa (45%); however, few of these cases develop skin necrosis with ulcers and digital ischaemia (28%) [24]. A small percentage develop subungual splinter haemorrhages. These clinical features have been shown to be related to microthrombi in dermal capillaries.

Other organs affected are the liver, peripheral vessels, intestine, spleen and adrenal glands. There are few reports of thrombosis in the pancreas, retina and bone marrow, as well as testicular/ovarian infarction, necrosis of the prostate and acalculous cholecystitis. However, the attribution of these clinical manifestations to CAPS is sometimes difficult [6].

Laboratory features

Patients with CAPS often develop anaemia and thrombocytopenia. These features are present in the early phase of the CAPS episode. In fact, anaemia, elevated lactate dehydrogenase and proteinuria have been found to be able to identify patients with early CAPS [25]. Sometimes schistocytes can be identified and clear features of TMA are demonstrated, but also features of disseminated intravascular coagulation (DIC) have been reported [6].

The presence of aPL is a hallmark of the disease and some patients present triple positivity [6]. However, triple aPL positivity has been shown not to be a reliable marker to differentiate between those patients with APS at risk of developing the catastrophic variant [25]. Some patients only have positivity for one or two antibodies, with lupus anticoagulant and IgG ACL followed by IgG anti-β2-GPI antibodies being more often reported. Almost 60% have ANA, but only 30% also have anti-dsDNA antibodies, usually fulfilling classification criteria for SLE, while 10% fulfil the classification criteria for other systemic autoimmune diseases [6]. Hypocomplementemia has been found in 60% of patients with CAPS; however, 55% of them have been shown to fulfil clinical criteria for SLE [26]. Furthermore, hypocomplementemia, although it seems to be an early biomarker for CAPS, was found not to be a good biomarker for TMA in these patients [25, 26].

Classification

Preliminary classification criteria for CAPS were proposed during the 10th International Congress on aPL in 2002 [27]. However, in the real world, the diagnosis of CAPS may be very difficult and therefore it is important to actively look for some clinical and laboratory features that may be clues for the identification of a patient with CAPS.

From the clinical point of view, it is important to suspect this condition in a patient with multisystem microangiopathic involvement. The diagnosis will be easier if the patient has a known history of APS or persistent aPL positivity. Furthermore, given that approximately half of CAPS patients have an associated autoimmune disease, in a patient with multi-organ involvement and a history of another autoimmune disease (particularly SLE), the clinical suspicion of CAPS is of paramount importance.

Differential diagnosis

Most cases of CAPS present as microangiopathic storm rather than large vessel occlusion. Given the fact that several thrombophilic situations also lead to thrombosis in multiple sites throughout the body, the differential diagnosis of patients with multiple thrombosis is usually very difficult [28].

When thrombosis presents in large vessels, the search for classic well-known risk factors for thrombosis should be undertaken. Typical risk factors for thrombosis include immobility, obesity, pregnancy, oral contraception, malignancy, surgery and hereditary and acquired thrombophilia [29]. Nevertheless, the thrombotic storm that characterizes CAPS should point to a differential diagnosis, including TMA. Classically, the differential diagnosis of CAPS includes severe infections, with or without DIC, TMA, malignancies, heparin-induced thrombocytopenia (HIT), HELLP (haemolysis, elevated liver enzymes and low platelets) syndrome and scleroderma renal crisis, among others [21].

The finding of aPL in a patient with multiple thromboses has been described as the hallmark of CAPS, allowing a correct differential diagnosis of microangiopathic disease. However, aPL are not pathognomonic of CAPS. Indeed, aPL have been reported in many other conditions, including infections, malignant diseases and autoimmune conditions [30]. However, they appear in low titres and are usually transient.

Severe systemic infections, presenting with sepsis or DIC, can develop symptoms that are similar to CAPS. Sometimes, however, the situations take place together, with the first (infection) acting as a trigger for the second (CAPS). There are some reports showing activation of inflammation and coagulation during severe infections leading to thrombosis and, at the same time, infections like coronavirus disease 2019 have been shown to develop transient aPL [2, 31]. However, although transient aPL positivity is thought to have no role in the risk of thrombosis, its significance in the context of CAPS is unclear. Today, high levels of aPL are considered a highly specific finding for CAPS, helping in the differential diagnosis between these two clinical situations [6].

Disorders more often associated with DIC are infections, severe trauma, malignancies and obstetric complications [32, 33]. DIC presents with thrombosis and bleeding driven by coagulation factor consumption, leading to prolongation of coagulation times and fibrinogen consumption [34]. However, features of DIC have been reported in patients with CAPS [35]. Thus, sometimes it might not be possible to differentiate between DIC and CAPS and both situations might take place together.

TMAs, such as thrombotic thrombocytopenic purpura (TTP), and HUS are the main conditions that are part of the differential diagnosis with CAPS and, indeed, differentiating between them represents a challenge for the clinician. Renal and neurological clinical manifestations with anaemia, thrombocytopenia and the presence of schistocytes in peripheral blood smears can be found in both TTP and CAPS. However, CAPS often presents with other clinical features not as frequently described in other TMAs. The presence of low levels of ADAMTS-13 activity might provide the clue for the diagnosis of TTP; in children, Escherichia coli toxin in stool should point to HUS; and high levels of aPL should point to the diagnosis of CAPS.

The increased risk of thrombosis in patients with malignancies is well established [36–38]. In some cases, microangiopathy and cancer are diagnosed concurrently [38]. Malignancy-related TMA is mostly found in disseminated solid cancers but has also been reported in haematological malignancies [37, 38]. Gastric, breast, prostate, lung and unknown primary cancers are the malignancies most frequently associated with TMA [37]. Furthermore, aPL are frequently found in patients with malignancies [39]. However, aPL found in patients with solid malignancies do not seem to be related to an increased thrombotic risk, while they seem to be related in patients with haematological malignancies [21, 40, 41].

Conversely, malignancies are one of the most frequent triggers of CAPS, reported in up to 16% of these patients [6]. Haematological malignancies, and especially Hodgkin’s lymphoma, are most frequently reported in CAPS patients [42].

HIT is a rare but sometimes severe complication of heparin treatment that occurs 4–10 days after the initiation of therapy [41]. Type II HIT is characterized by the formation of heparin-anti-platelet factor 4 (PF4) antibodies complex that bind to platelets, leading to thrombosis [43]. Patients with APS are often treated with heparin during bridging therapies to perform exploratory tests, and aPL have been found in about one-third of patients with HIT and up to 10% of heparin-naïve aPL-positive patients were found to be positive for anti-PF4 antibodies [44]. Furthermore, both arterial and venous thrombosis are seen in HIT patients. A history of heparin administration and the presence of anti-PF4 antibodies can help physicians to distinguish between these clinical situations, although the presence of anti-PF4 antibodies has not been studied in a systematic way in CAPS [21]. Sometimes functional platelet aggregation assays can be helpful for correct diagnosis [45].

HELLP syndrome is an endothelium disease that affects small vessels in the liver. HELLP syndrome normally occurs at the end of pregnancy [46]. The association of APS and early-onset HELLP syndrome is well known. Some series have found that about half of cases with an early-onset second trimester HELLP are associated with APS. A review of patients included in the CAPS Registry found 15 cases that occurred during pregnancy and post-partum and about half of them developed HELLP syndrome simultaneously [47]. Indeed, HELLP syndrome has been considered an expression of CAPS in some cases or a trigger of CAPS [48]. Probably, both clinical situations might favour each other. However, the optimal diagnostic approach and management of patients with concurrent CAPS and HELLP syndrome is still unclear.

Scleroderma renal crisis has been included as a clinical situation that needs to be taken into the differential diagnosis for patients with CAPS. A previous history of RP and the presence of usually diffuse sclerodermic cutaneous findings makes the differential diagnosis easier. However, although rarely, the development of CAPS in patients with SSc has been reported. In these patients, the determination of aPL levels and a renal biopsy is recommended to establish the diagnosis.

Treatment

Due to the rarity of CAPS, its treatment is largely empirical. There are no randomized controlled trials to guide the efficacy of the therapies; however, analysis of the cases reported in the CAPS Registry has allowed some general guidelines to be proposed [49]. Furthermore, Fig. 1 shows our proposed algorithm for the management of a case with the suspicion of CAPS [50]. CAPS has a bad prognosis when no treatment is prescribed and therefore prompt and aggressive treatment is justified [34]. Classically, three aspects have been claimed as the basis for treatment. First, the so-called supportive general measures; second, aggressive treatment of any identifiable trigger; and finally, specific treatment.

Since often patients with CAPS end up in an intensive care unit because of multi-organ failure, appropriate supporting care is mandatory. Patients might need mechanical ventilation support due to pulmonary failure or renal replacement therapy when renal failure is an issue. Sometimes cases require inotropic drug support.

As this condition is often preceded by a trigger, early identification of that trigger is important since it might allow an early intervention to decrease the thrombogenic state. The most frequent precipitating factors are infections, followed by surgical procedures and malignancies. Thus, looking for airway, urinary and skin infections is encouraged, especially if there is fever. Perioperative prophylaxis with heparin is advocated in aPL carriers with or without previous thrombosis, since several cases developed in this setting.

Specific treatment is based on the pathophysiologic events that take place during the episode. Current treatment guidelines for CAPS were proposed based on the analysis of the first 130 patients reported with this condition, and more recent studies performed using an evidence-based medicine approach have confirmed the validity of those initial guidelines [49–51]. These studies show that the highest survival rate is achieved with a combination of anticoagulation, glucocorticoids and plasma exchange and/or IVIG. In detail, analysis of the CAPS Registry showed that triple therapy was prescribed in 189 (40.1%) episodes, other combinations in 270 (57.3%) and none of those treatments in 12 (2.5%) episodes; the mortality rate was as high as 75% in those episodes that did not receive these treatments, 41.1% in those that received another combination and 28.6% in patients that received triple therapy [50]. Interestingly, the authors did not find any statistical differences in the survival rate between those patients who received triple therapy with plasma exchange or IVIG [50].

Some authors have used CYC with good results, especially in the setting of SLE [51]. More recently, some biologic drugs have been included in the therapeutic armamentarium of these patients. Our group was able to show the benefit of rituximab in some cases with CAPS, although the evidence is still scarce [52]. Additionally, increasing evidence on the role of complement in the pathophysiology of CAPS has led many authors to consider the use of eculizumab, especially in patients with TMA features. A recent review of these cases showed that 29 of 39 patients who received eculizumab recovered from the CAPS episode, while 9 worsened and 5 died despite treatment [53].

However, to date all these treatments are empirical. The doses, schedule and adverse effects are unclear and more studies are needed to better define the optimal therapeutic approach to this condition.

Prognosis

The long-term prognosis of patients who survive the initial catastrophic event was analysed in a study of the CAPS Registry that demonstrated that 66% of them remained symptom free with anticoagulation during an average follow-up of 67.2 months [54]. However, 26% of surviving patients developed further APS-related thrombosis after the initial CAPS event. Furthermore, 15% of patients were functionally impaired because of CAPS, including cardiac failure, chronic renal insufficiency, gait abnormalities and visual symptoms. However, prognostic factors that predict the long-term outcomes of patients who suffer a CAPS event are unknown.

Conclusion

While significant progress has been made in understanding the clinical features of CAPS, there are still several gaps in our knowledge [55, 56]. Ongoing research is crucial to further elucidate the underlying pathophysiological mechanisms, improve diagnostic and predictive tools, refine treatment strategies and enhance long-term management and prevention approaches (Table 2).

Data availability

No new data were generated or analysed in support of this research.

Authors’ contributions

All three authors contributed equally to the elaboration of this review article.

Funding

No specific funding was received from any bodies in the public, commercial or not-for-profit sectors to carry out the work described in this article.

Disclosure statement: The authors have declared no conflicts of interest.

References