Diffuse calcinosis is a common manifestation of systemic sclerosis, mainly affecting the extremities.¹ The subcutaneous calcium deposits may erode the skin and predispose these areas to secondary infection, yet no currently approved medical treatment exists. Rare anecdotal reports suggest that low dose warfarin may have a role in reducing the size of calcinotic plaques.^2–,⁴ We describe three patients who had multiple calcinotic lesions secondary to systemic sclerosis and who were treated with warfarin for 1 year, with results varying in accordance with the age and size of the calcinotic lesions.

CASE REPORTS

DISCUSSION

Subcutaneous calcium deposition is a disturbing complication affecting patients who have connective tissue diseases—in particular, systemic sclerosis and dermatomyositis. Calcinosis may result in the formation of painful ulcers that can be easily infected. The mechanism of calcinosis formation is not entirely understood, and no effective treatment is available.¹ Throughout the years many treatment regimens have been tried and anecdotal case reports and small series have shown some or little improvement with diltiazem, adrenal steroids, carbon dioxide laser treatment, surgical treatment, and aluminium hydroxide. Better results have been obtained with aledronate, anti-tumour necrosis factor (anti-TNF), and warfarin (table 1⇓).^2–,¹⁶

In 1987 Berger et al offered a possible mechanism for the development of calcinosis.² Levels of the calcium binding amino acid γ-carboxyglutamic acid were found to be high in areas of calcinosis, so it was postulated that it might have a role in binding large amounts of calcium salts at the affected areas, leading to the formation of calcinosis.¹⁷

The carboxylation of glutamine, leading to the formation of the γ-carboxyglutamic acid, is coupled to the vitamin K cycle and is inhibited by warfarin.¹⁸ Thus it was suggested that carboxylation of glutamine to γ-carboxyglutamic acid, leading to calcinosis, might be inhibited by low doses of warfarin. At low doses, warfarin does not prolong the PT or PTT, does not cause an excess bleeding tendency, and thus requires no laboratory follow up of the patients’ coagulation status.

Two small clinical trials tested this hypothesis. The results of the first trial² demonstrated a radiological improvement in all four patients treated with warfarin, but a clinical reduction in calcinotic lesion size in only one of the four patients. The responsive patient had small calcinotic lesions with a low bone scan score, thought to represent a mild calcinotic process, whereas the other three patients had larger calcinotic plaques and high bone scan scores, thought to represent a more advanced and longer standing calcinotic process.

Lassoued et al reported their experience of using warfarin to treat six patients with diffuse calcinosis for 1 year.³ No significant clinical reduction in calcinotic lesion size occurred in any of the patients. All six patients had extensive calcinosis (Berger grade 4+) which had been present for a long time before the start of treatment.

Yoshida and Torikai described a patient with longstanding, yet mild, calcinosis (Berger grade 2) who was treated with low dose warfarin (1 mg). Complete clinical and radiographic response was noted.⁴

Our three patients had multiple calcinotic lesions and were treated with 1 mg of warfarin for a period of 1 year. The first two patients showed a rapid clinical response, with complete disappearance of all calcinotic lesions after 2 months of treatment. In the third patient, no change was noted in the size of the calcinotic lesions after a year of treatment. The two major differences between the responding and non-responding patients were the duration of the calcinotic lesions before the start of treatment, and their size. In the two responders the lesions developed relatively shortly before the start of treatment and were all smaller than 2 cm in diameter. In the non-responder the lesions were known to have been present for at least 5 years and were of larger diameter.

On the basis of our observations and the previously described cases,^2–,⁴ we conclude that two main factors influence the response to warfarin treatment in calcinosis: the size of the calcinotic lesions, and the time lapse between the formation of the calcinotic lesions and the start of treatment.

One possible explanation for the importance of these factors may be that relatively newer onset and smaller calcinotic lesions are more dependent for their growth on the amount and availability of γ-carboxyglutamic acid at the nidus site, and thus are more prone to inhibition by warfarin. The longer standing and larger lesions, on the other hand, are composed of greater amounts of surrounding ectopic calcium deposits, serving by themselves as a nidus for further calcification. As a result these lesions are less dependent for their growth on the availability of γ-carboxyglutamic acid and are more resistant to the inhibitory effects of warfarin.

A drawback of this treatment that still has to be examined is the potentially enhancing effect of warfarin on artery calcification as demonstrated in rat models.¹⁹ It is not clear yet whether this effect occurs in humans.

In conclusion, there still seems to be no highly effective treatment for calcinosis. We suggest that low dose warfarin may be effective in selected patients with systemic sclerosis who have early and mild calcinosis. At low doses this treatment is well tolerated, with a very low risk of bleeding. Larger, controlled studies are needed to validate this observation.