Article Text


Raynaud’s phenomenon and antiphospholipid antibodies in systemic lupus erythematosus: is there an association?
  1. D Caccavo1,
  2. F Del Porto2,
  3. P Garzia3,
  4. A P Mitterhofer2,
  5. S Galluzzo2,
  6. A Rigon3,
  7. M Vadacca3,
  8. M F Navajas3,
  9. A Amoroso2,
  10. A Afeltra3
  1. 1Department of Clinical Medicine, Immunology, and Infectious Diseases, University of Bari, Bari, Italy
  2. 2Department of Clinical Medicine, University La Sapienza, Rome, Italy
  3. 3University Campus Bio-Medico, Rome, Italy
  1. Correspondence to:
    Professor A Afeltra, University Campus Bio-Medico, Viale E Longoni 83, 00155 Rome, Italy;


Objective: To evaluate the association of IgG and IgM antibodies directed against different negatively charged phospholipids (that is, anticardiolipin (aCL), antiphosphatidylinositol, antiphosphatidylserine, and antiphosphatidic acid) and anti-β2-glycoprotein I (aβ2GPI), with Raynaud’s phenomenon in patients with systemic lupus erythematosus (SLE).

Methods: Ninety three patients with SLE (81 female), 40 with and 53 without Raynaud’s phenomenon, were included in the study. IgG and IgM antiphospholipid antibodies and aβ2GPI were determined by enzyme linked immunosorbent assay (ELISA).

Results: Fifty patients (54%) were positive for IgG and/or IgM antibodies to one or more phospholipid antigens or to β2GPI. The prevalence of all autoantibodies evaluated, either IgG or IgM, was higher in patients without than in those with Raynaud’s phenomenon. A negative association was found between IgG aCL and Raynaud’s phenomenon (p=0.038), whereas autoantibodies other than aCL were not significantly associated with Raynaud’s phenomenon.

Conclusion: Our results demonstrate no positive association between antiphospholipid antibodies and Raynaud’s phenomenon in SLE and indicate that measurement of anti-negatively charged phospholipid antibodies other than aCL is not useful as a serological marker predictive for Raynaud’s phenomenon.

  • antiphospholipid antibodies
  • Raynaud’s phenomenon
  • systemic lupus erythematosus
  • 2GPI, anti-β2-glycoprotein I antibodies
  • aCL, anticardiolipin antibodies
  • aPA, antiphosphatidic acid antibodies
  • aPI, antiphosphatidylinositol antibodies
  • aPL, antiphospholipid antibodies
  • aPS, antiphosphatidylserine antibodies
  • CTD, connective tissue diseases
  • RP, Raynaud’s phenomenon
  • SLE, systemic lupus erythematosus

Statistics from

Raynaud’s phenomenon (RP) is a disorder characterised by recurrent reversible vasospasm of fingers and toes often induced by exposure to cold.1 It is classified as primary or secondary, according to the absence or the presence of a known associated disease, respectively. Secondary RP is frequently present in patients affected by connective tissue diseases (CTD), including systemic sclerosis, systemic lupus erythematosus (SLE), Sjögren’s syndrome, dermatomyositis or polymyositis, and rheumatoid arthritis.1

Antiphospholipid antibodies (aPL) are autoantibodies directed against negatively charged phospholipids and their presence is significantly associated with arterial and/or venous thrombosis and recurrent fetal loss, the so-called antiphospholipid syndrome.2,3 Antiphospholipid antibodies have also been found to be associated, to a lesser extent, with other clinical manifestations.2,3 Besides anticardiolipin antibodies (aCL), which have been the most extensively studied aPL, the prevalence and clinical relevance of autoantibodies against other negatively charged phospholipids, such as antiphosphatidylinositol (aPI), antiphosphatidylserine (aPS), antiphosphatidic acid (aPA), and antibodies to β2-glycoprotein I (aβ2GPI), a plasma protein with anticoagulant activities, have been evaluated in recent years.4–8

Only a few studies have been performed to examine the association between RP and aCL, and conflicting results have been reported.9–13 On the other hand, no data are available on the association between RP and different aPL, other than aCL.

Therefore, this study aimed at analysing the association between RP and aCL, aPI, aPS, aPA, and aβ2GPI in patients affected by SLE.


Ninety three patients, fulfilling four or more criteria of the American College of Rheumatology for the classification of SLE, updated according to Hochberg,14,15 were included in the study. Patients were divided into two groups according to the presence (SLE cases) or the absence (SLE controls) of RP. To determine the cut off level of aPL, serum samples from 50 healthy donors matched for sex and age, who showed normal coagulation assays and negative serological tests for syphilis, were used. Sera were stored at −80°C until used. Samples were coded and investigators performing the assays were unaware of the presence or absence of RP.

Antiphospholipid antibody quantification

The measurement and characterisation of IgG and IgM aPL were performed by enzyme linked immunosorbent assay (ELISA) using commercially available kits (ORGenTec-Diagnostika GmbH, Mainz, Germany). In brief, serum samples diluted 1:100 were first screened in microplates coated with a mixture of negatively charged phospholipids—that is, cardiolipin, phosphatidylinositol, phosphatidylserine, and phosphatidic acid (AntiPhospholipid Screen, ORGenTec-Diagnostika). The wells were saturated with purified human β2GPI as cofactor for antibody binding. All samples were assayed in duplicate. Combined IgG and IgM calibrators ranging from 0 to 120 and from 0 to 80 arbitrary U/ml, respectively, were supplied by the manufacturer, and a standard curve was established. The assay system was calibrated against the internationally recognised aCL reference sera from EN Harris, Lousville, USA. Values ⩾10 arbitrary U/ml (a value higher than the 99th centile of the control sera for both IgG and IgM antibodies) were considered positive. Sera positive in this preliminary screening were tested against single phospholipid antigens coated in the presence of β2GPI as cofactor, or against purified β2GPI which was coated on high binding γ irradiated wells. Sera showing optical density values exceeding the detection limit of the standard curve were further diluted and assayed.

Statistical analysis

Two-group comparisons were analysed by Fisher’s exact test (two tailed) for categorical variables and by Student’s t test or Mann-Whitney U test for continuous variables according to Gaussian or non-Gaussian distribution, respectively. Values of p<0.05 were considered to be significant.


The study group comprised 93 patients, of whom 81 were women. Their mean age was 38.5 (SD 11.3 years; range 17–69). Forty patients (43%) had RP never complicated by digital necrosis. Table 1 shows the demographic and salient immunological features of the two patient groups. No significant differences in age, sex distribution, disease duration, presence of antinuclear antibodies and anti-dsDNA antibodies, and treatment were present between groups.

Table 1

Comparison of demographic and salient immunological data for Raynaud’s phenomenon positive and Raynaud’s phenomenon negative patients

Table 2 shows the prevalence, antigen specificity, and isotype distribution of aPL in the total study group.

Table 2

Antiphospholipid antibody (aPL) in the whole group of patients with SLE (n=93): prevalence, antigen specificity, and isotype distribution

Association between aPL and Raynaud’s phenomenon

Table 3 compares the prevalence of different aPL in patients with and without RP. Overall, RP was observed in a higher percentage of aPL negative than in aPL positive patients. This finding was observed, to a different extent, for either IgG or IgM aCL, aPI, aPS, aPA, and aβ2GPI. However, the negative association between aPL and RP reached significance (p=0.038) only for IgG aCL. No significant difference was shown between the median titre of aPL between patients with and without RP, irrespective of antigen specificity and isotype (data not shown).

Table 3

Raynaud’s phenomenon: comparison between patients with SLE with and without aPL

To determine whether the degree of aCL positivity differed significantly between RP positive and RP negative patients, aCL positive patients were subdivided into two groups—namely, patients with low (<20 U/ml) and patients with medium-high (⩾20 U/ml) antibody titre. Twenty eight of 50 aCL positive patients (56%) showed medium-high antibody titre. The distribution of patients with medium-high aCL titre between RP positive and RP negative patients was 10/40 (25%) among RP positive and 18/53 (34%) among RP negative patients, and the difference was not significant (p=0.37).


In this study we provide evidence that secondary RP is not positively associated with the presence of aPL in patients with SLE and we found a significant negative association between IgG aCL and RP. The results so far reported on the association between RP and aPL are contradictory, and this could be due, at least in part, to the different study groups evaluated in these investigations.

Vayssairat et al found a significantly higher prevalence of aCL in patients with RP than in healthy controls, especially in patients with CTD.9 However, because the prevalence of either aPL or RP is significantly higher in patients with CTD than in the normal population this finding is not surprising. On the other hand, in a case-control study limited to patients with clinical disorders such as thromboses, repeated abortions, thrombocytopenia without SLE, a significant positive association was reported between RP and aCL.10 However, it should be pointed out that this last study was performed on a relatively small number of patients.

By contrast, in a cooperative study conducted on a large number of patients with different CTD, also including 165 patients with undifferentiated CTD, a negative association between the presence of aCL and RP was found.11 This finding is in agreement with our results, even though it is hard to provide a possible explanation for this negative association. On the other hand, it should be pointed out that no significant association was observed in our study when the comparison was restricted to patients with a medium-high aCL titre. Another group of investigators evaluated the presence of aCL in patients affected by primary or secondary RP and found that, among 17 patients with SLE and RP, 7 (41%) were aCL positive.12 Of note, this percentage is almost the same found in our study, where 17/40 (43%) patients with SLE and RP showed aCL positivity. Finally, Bongard et al demonstrated a relationship between aCL and nailfold capillary changes in patients with SLE, but they failed to find an association between aCL and RP.13

Our further objective was to investigate a putative association between RP and aPL directed against other negatively charged phospholipids or β2GPI, because in recent years conflicting results have been reported on either the clinical relevance or clinical associations of these autoantibodies.7,8 We found IgG and/or IgM aPI, aPS, aPA, and aβ2GPI in a lower percentage of patients in comparison with aCL and we recorded no significant association between RP and these different aPL or aβ2GPI.

In our opinion, the lack of association between aPL and RP is not surprising because vasospasm of the small muscular arteries and arterioles of the digits and not thrombosis is responsible for the syndrome. However, thrombotic events may complicate severe forms with sustained vasospasm, and patients unresponsive to conventional vasodilator treatment can be successfully treated with long term anticoagulation.1

In conclusion, our results, obtained in an adequate number of patients, argue against a positive association between RP and antibodies directed against negatively charged phospholipids or β2GPI, at least in SLE. The importance of the negative association between IgG aCL and RP, if any, needs to be confirmed in a larger number of patients.


View Abstract

Request permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.