Medikamenteninduzierte Ulcera/en: Unterschied zwischen den Versionen
Becher (Diskussion | Beiträge) (Die Seite wurde neu angelegt: „<h3>Introduction</h3> Wounds are caused by a variety of factors. In addition to leg ulcers, diabetic wounds and pressure ulcers, there is a wide range of other…“) |
Becher (Diskussion | Beiträge) |
||
(31 dazwischenliegende Versionen desselben Benutzers werden nicht angezeigt) | |||
Zeile 8: | Zeile 8: | ||
__TOC__ | __TOC__ | ||
− | <h3>Introduction</h3> | + | <h3 class="clear">Introduction</h3> |
Wounds are caused by a variety of factors. In addition to leg ulcers, diabetic wounds and pressure ulcers, there is a wide range of other factors which can cause chronic skin defects with poor or no healing tendency. In particular, very rare causes, if not recognised, lead to non-healing wounds if the underlying cause is not eliminated through ignorance. Drugs can also cause ulcers when administered orally or injected. There are many individual publications in the literature and, for some drugs, reviews with small case numbers. The data available on such wounds is generally very limited. In a survey of experts in Germany on the genesis of chronic venous leg ulcers involving more than 31,000 patients, the incidence of drug-induced ulcers was determined at 1% (1). | Wounds are caused by a variety of factors. In addition to leg ulcers, diabetic wounds and pressure ulcers, there is a wide range of other factors which can cause chronic skin defects with poor or no healing tendency. In particular, very rare causes, if not recognised, lead to non-healing wounds if the underlying cause is not eliminated through ignorance. Drugs can also cause ulcers when administered orally or injected. There are many individual publications in the literature and, for some drugs, reviews with small case numbers. The data available on such wounds is generally very limited. In a survey of experts in Germany on the genesis of chronic venous leg ulcers involving more than 31,000 patients, the incidence of drug-induced ulcers was determined at 1% (1). | ||
− | <h3>Definition | + | <h3>Definition of a drug-induced ulcer</h3> |
− | + | Drug-induced ulcers occur in temporal relation to the intake of a specific preparation. As a rule, but not always, the defect does not heal until after discontinuation. Local intensive wound treatment etc. can lead to a partial improvement of the findings, but not to healing. Ideally, evidence of drug-induced ulcus genesis is provided by conscious or frequent accidental re-exposure with renewed development of a skin ulcer after interim healing. | |
− | <h3> | + | <h3>Issue</h3> |
− | + | Many patients are multimorbid and take numerous drugs. A differentiation is then often hardly feasible. Individual case studies alone therefore hardly prove a drug-induced pathogenesis. They can only give indications of a possible connection. Only when many cases are examined together can a causal connection be proven. Histological samples from drug-induced ulcers provide no evidence of a drug-induced pathogenesis. Vasculitis, i.e. inflammatory processes in cutaneous or subcutaneous vessels, thrombus formation with occlusion of vessels or other pathomechanisms which lead to skin defects via secondary infarction, can be caused by a variety of causes in addition to drugs. | |
− | <h3> | + | <h3>Known examples of drug-induced ulcers</h3> |
<b><i>Hydroxyurea</i></b> | <b><i>Hydroxyurea</i></b> | ||
− | Hydroxyurea | + | Hydroxyurea is a hydroxylated urea derivative and has been used as a cytostatic drug since 1916 (2). Side effects are reported in the literature between 10 and 35% (3). The frequency of the occurrence of skin ulcers under hydroxyurea therapy is stated extremely differently in the literature. It varies from less than 0.1% (4) up to 12% (5). |
− | + | It is possible that there are still disease-related incidence differences. Najean and Chaine report that 8-9% of patients with myelodysplastic diseases and 29% of patients with sickle cell anaemia developed leg ulcera taking Hydroxyurea (6,7). | |
− | + | The initial description was in 1986 as part of a long-term therapy with Hydroxyurea in chronic myeloid leukaemia (8). | |
− | Hydroxyurea, | + | Hydroxyurea, also known as hydroxycarbamide, causes inactivation of ribonucleotide reductase with inhibition of cellular DNA synthesis and cell death in the S-phase. It is therefore used as a cytostatic agent for leukemias and myeloproliferative diseases. In addition, its use is known to be effective in sickle cell anaemia. It has also been used in the antiretroviral treatment of HIV infections. |
− | Best e. a. | + | Best e. a. describe in 1998 14 own patients with hydroxyurea-induced ulcers from a set of 115 patients with different myeloproliferative diseases with the intake of hydroxyurea. Their average age was 65 (41-74). Usually, the ulcers were located in the ankle joint region, but occasionally also above the tibia, the back of the foot or on the lower leg. In 10 cases on the inner ankle and 8 cases on the outer ankle, there was no localisation preference for the extremely painful ulcers. 64% of patients had multiple ulcerations. The mean cumulative dose before development of the ulcers was 2.3 kg (1.4-5.5 kg). The mean duration of administration was 6.1 years (2-15 years). The ulcers healed in 12 of 14 patients after discontinuation of treatment, in 2 patients an additional split skin transplant was necessary. In one case the dose was reduced from 1.5 to 0.5 hydroxyurea. Healing occurred within 5 months. Another patient had an improvement in the reduction of the therapy from 2 to 1 g, when the dose was increased again to 2.5 g the original ulcer deteriorated over 4 months and others appeared, so that the therapy had to be discontinued (5). From this and many other literature references it is clear that discontinuation of therapy for hydroxyurea-induced ulcers is necessary. Healing then usually occurs slowly but spontaneously. |
− | + | Histology does not provide decisive diagnostic changes. In the dermis, cell swelling, oedema, fatty degeneration of the vascular walls and perivascular lymphocytic infiltrations without vasculitis can be detected. Overall, the signs correspond to those of a cutaneous occlusive vasculopathy. In later stages dermal fibrosis is described (5). | |
− | + | The size and extent of the ulcerations do not correlate with the amount of hydroxyurea. | |
− | In 2 | + | In 2 of the 14 patients Best describes the recurrence of ulcers after re-exposure. One re-exposure took place 3 years later, and a few months later multiple ulcers occurred in both ankle joints, which did not heal until 6 months after re-exposure. In the second patient, it took 4 months after re-exposure to develop new cutaneous ulcerations (5). |
− | + | Thus the genesis of hydroxyurea-induced leg ulcerations is confirmed. | |
<table border="1"> | <table border="1"> | ||
<tr> | <tr> | ||
− | <td> | + | <td>Substance</td> |
<td>Hydroxyurea</td> | <td>Hydroxyurea</td> | ||
</tr> | </tr> | ||
<tr> | <tr> | ||
− | <td> | + | <td>Appearance</td> |
− | <td> | + | <td>After a latency of several years</td> |
</tr> | </tr> | ||
<tr> | <tr> | ||
− | <td> | + | <td>Localisation</td> |
− | <td> | + | <td>Mainly malleolar, but also other localisations</td> |
</tr> | </tr> | ||
<tr> | <tr> | ||
− | <td> | + | <td>Complaints</td> |
− | <td>Extreme | + | <td>Extreme painfulness</td> |
</tr> | </tr> | ||
<tr> | <tr> | ||
− | <td> | + | <td>Therapy</td> |
− | <td> | + | <td>Discontinuation of the medication</td> |
</tr> | </tr> | ||
<tr> | <tr> | ||
− | <td> | + | <td>Healing</td> |
− | <td> | + | <td>Spontaneously, but slowly</td> |
</tr> | </tr> | ||
<tr> | <tr> | ||
− | <td> | + | <td>Safety of the context</td> |
− | <td> | + | <td>ensured</td> |
</tr> | </tr> | ||
</table> | </table> | ||
− | <b><i> | + | <b><i>Anagrelide</i></b> |
− | + | Anagrelide is an imidazoquinazoline, which was approved in Germany in 2005 for essential thrombocythemia. Its inhibitory effect on human platelet formation is achieved by delaying megakaryocyte maturation through inhibition of cyclic AMP phosphodiesterase III. Commonly described side effects are headache, diarrhoea, palpitation, weakness, pain, dizziness, abdominal pain and nausea. However, the vast majority of side effects resolve within several weeks (9). | |
− | + | Due to its excellent safety profile, Anagrelide has been shown to be a first-line treatment for myeloproliferative diseases associated with plateletosis, including primary thrombocythemia. | |
− | Ruiz-Argüelles e.a. | + | Ruiz-Argüelles e.a. describe a 74-year-old patient who had already developed a leg ulcer under hydroxyurea (1g/d), which healed within 4 months after discontinuation. 3 years later the patient received Anagrelide (1g/d), 5 months later extreme pain in the right lateral malleolus reappeared. Two further weeks later, an ulcer developed in exactly the same localisation as the hydroxyurea-induced ulcer that had occurred previously (10). |
− | Rappoport | + | Rappoport also describes a patient who, after 3 years of taking 0.5 g Hydroxyurea p/d for 6 weeks after conversion to 1g Anagrelide p/d, developed extremely painful erythema in the area of both lateral ankle joints, which subsequently became ulcerated. Basically, these ulcers could be the result of long-term use of Hydroxyurea or an expression of the newly initiated therapy with Anagrelide. However, as they only healed completely after the therapy with hydroxyurea, there is also the suspicion of Anagrelide-induced leg ulcers (9). |
− | + | There are further reports of Anagrelide-induced leg ulcers, but no large case numbers. Thus, even under the substitute drug for hydroxyurea, namely Anagrelide as a first-line as well as second-line therapy for myeloproliferative diseases, as well as for hydroxyurea, identical ulcers can develop, sometimes even in loco (10)! | |
<table border="1"> | <table border="1"> | ||
<tr> | <tr> | ||
− | <td> | + | <td>Substance</td> |
− | <td> | + | <td>Anagrelide</td> |
</tr> | </tr> | ||
<tr> | <tr> | ||
− | <td> | + | <td>Appearance</td> |
− | <td> | + | <td>After a latency of several month</td> |
</tr> | </tr> | ||
<tr> | <tr> | ||
− | <td> | + | <td>Localisation</td> |
− | <td> | + | <td>Mainly malleolar, but also other localisations</td> |
</tr> | </tr> | ||
<tr> | <tr> | ||
− | <td> | + | <td>Complaints</td> |
− | <td>Extreme | + | <td>Extreme painfulness</td> |
</tr> | </tr> | ||
<tr> | <tr> | ||
− | <td> | + | <td>Therapy</td> |
− | <td> | + | <td>Discontinuation of the medication</td> |
</tr> | </tr> | ||
<tr> | <tr> | ||
− | <td> | + | <td>Healing</td> |
− | <td> | + | <td>Spontaneously, but slowly</td> |
</tr> | </tr> | ||
<tr> | <tr> | ||
− | <td> | + | <td>Safety of the context</td> |
− | <td> | + | <td>ensured</td> |
</tr> | </tr> | ||
</table> | </table> | ||
− | <b><i> | + | <b><i>Coumarins</i></b> |
− | + | Coumarins are drugs used for anticoagulation. We distinguish between Phenprocoumon (Marcumar®, Falithrom®, Phenprogamma®, etc.), Warfarin (Coumadin®, etc.), and Acenocoumarol (Sintrom®). There are also reports of the use of other coumarins such as Acoumarol (11). | |
− | + | All coumarins may cause skin ulcerations. | |
− | + | The most common ones are phenprocoumon with a half-life of 10-14 days and warfarin with a half-life of 2 days. | |
− | + | Typically, coumarins decrease the activity and synthesis of coagulatory factors II, VII, IX, X and inhibit the formation of protein C and S, both of which have anticoagulatory effects. However, since these factors have different half-lives, there is an imbalance between anticoagulatory and procoagulatory factors. Factor VII and protein C have a short half-life. When warfarin is given initially, factor VII and protein C levels go down first. Factor IX decreases more slowly, followed by factor X, protein S and finally factor II. This can cause a temporary hypercoagulable state with inhibition of anticoagulant protein C production. In individuals with congenital protein C deficiency, this temporary hypercoaguability may be more pronounced. Patients with heterozygous protein C deficiency have a low level of protein C and thus a higher risk of suffering from WISN (warfarin-induced skin necrosis) when treated with Warfarin. Other hypercoaguable conditions including factor V Leiden mutation, antithrombin III deficiency and lupus anticoagulant detection are also associated with WISN. In contrast, other deficiencies of vitamin K such as under chemotherapy, liver disease, elimination of vitamin K-producing intestinal flora by antibiotics are rarely sufficient to cause WISN alone, but may increase the risk in hypercoagulable patients. The most likely mechanism of WISN is microvascular thrombosis, but hypersensitivity reactions to warfarin and direct damage to capillaries leading to dilation and rupture of vessels are also discussed as a possible pathomechanism. Typical affected patients are middle-aged, overweight women who have been treated for deep vein thrombosis, pulmonary embolism. The typical onset is day 3 - day 10 after starting warfarin therapy, but late onset up to 18 months, even 15 years, has been described (12). However, these late event data should be viewed critically, as a more detailed review of the literature may also involve discontinuation and then reinsertion of the drug (13). | |
− | + | Skin necrosis on coumarin administration is rare overall and occurs in about one in 5000 patients and about 3% of people with protein C deficiency. Typically affected are the fat-rich skin areas such as chest, hips and legs, very rarely also internal organs such as adrenal glands, uterus and the male genitals (14). | |
− | + | Although the occurrence of warfarin initiation has been described, the initiation phase appears to be only a partial factor in its genesis. Koduri and Steward each independently describe a patient who did not develop necrosis until the second administration of warfarin (13, 15). The multiple chronologically staggered occurrence is also described (15). | |
− | + | The first description was made by Flood in 1943 with the presentation of thrombophlebitis migrans disseminata with development of gangrene on the female breast (16). | |
<table border="1"> | <table border="1"> | ||
<tr> | <tr> | ||
− | <td colspan="2"><b> | + | <td colspan="2"><b>Comparison of bleeding with necrosis in patients receiving coumarin therapy (17)</b></td> |
</tr> | </tr> | ||
<tr> | <tr> | ||
− | <td> | + | <td>Bleeding</td> |
− | <td> | + | <td>Necrosis</td> |
</tr> | </tr> | ||
<tr> | <tr> | ||
− | <td> | + | <td>Affects men and women equally</td> |
− | <td> | + | <td>Female dominance</td> |
</tr> | </tr> | ||
<tr> | <tr> | ||
− | <td> | + | <td>Start not in relation to the start of therapy</td> |
− | <td> | + | <td>Start on the 3rd-6th day after starting coumarin therapy</td> |
</tr> | </tr> | ||
<tr> | <tr> | ||
− | <td> | + | <td>correction with the administration of vitamin K</td> |
− | <td> | + | <td>Increase despite administration of vitamin K </td> |
</tr> | </tr> | ||
<tr> | <tr> | ||
− | <td> | + | <td>Continuation of coumarin administration worsens the complication</td> |
− | <td> | + | <td>The continuation of coumarin administration has no effect on the extent of the complication</td> |
</tr> | </tr> | ||
<tr> | <tr> | ||
− | <td>Heparin | + | <td>Heparin worsens the complication</td> |
− | <td>Heparin | + | <td>Heparin can be helpful</td> |
</tr> | </tr> | ||
<tr> | <tr> | ||
− | <td> | + | <td>Missing necroses</td> |
− | <td> | + | <td>Existing necrosis</td> |
</tr> | </tr> | ||
<tr> | <tr> | ||
− | <td> | + | <td>Surgical interventions not necessary</td> |
− | <td> | + | <td>Surgical interventions often necessary</td> |
</tr> | </tr> | ||
</table> | </table> | ||
− | <b><i> | + | <b><i>Heparins</i></b> |
− | + | Allergic skin reactions or skin necrosis caused by heparins or low-molecular heparins are rarely but probably underestimated because too little is reported. The etiology of heparin-induced skin necrosis is unknown, the occurrence is often associated with HIT type II. This is an immunological reaction that results in binding of heparin to platelet factor 4 (PF-4), which leads to production of heparin-dependent IgG, IgM and IgA antibodies. Elevated IgG antibody titers are observed in patients with heparin-induced skin lesions, regardless of whether they develop thrombocytopenia. This suggests an immunological mechanism (18). | |
− | + | The broad spectrum of cutaneous reactions ranges from erythematous itching areas to large symptomatic plaques and heparin-induced skin necrosis. Women are more frequently affected than men. These are local necroses in the area of the application sites (19). But necroses outside the injection sites have also been described (20)! | |
− | |||
− | + | Heparin-induced skin necrosis starts between day 5 and day 10 after the start of intravenous subcutaneous heparin administration. A later onset until day 16 has been reported (20). | |
− | + | Skin necrosis can occur in patients with unfractionated heparin as well as in patients with LMW-heparin (21, 22). | |
− | + | The necroses begin with small erythematous painful lesions, which later spread to necrotic areas. Histologically, microvascular thrombi are found in small vessels with minimal inflammatory response (21). | |
+ | Cross-reactions between heparin and low molecular weight heparin are observed (19). | ||
− | + | Therapy recommendation: Therapy must be discontinued if systemic allergic reactions occur. The anticoagulant therapy should be changed to hirudin and/or coumarins. Appropriate skin tests should be performed. Depending on these tests, the anticoagulatory therapy can then be discontinued or modified. Alternatively, the new oral thrombin inhibitors etc. are alternative treatment methods. | |
<table border="1"> | <table border="1"> | ||
<tr> | <tr> | ||
− | <td> | + | <td>Substance</td> |
− | <td> | + | <td>Heparins</td> |
</tr> | </tr> | ||
<tr> | <tr> | ||
− | <td> | + | <td>Heparins</td> |
− | <td> | + | <td>Unfractionated, also LMWH, cave cross reactions</td> |
</tr> | </tr> | ||
<tr> | <tr> | ||
− | <td> | + | <td>Time occurrence</td> |
− | <td> | + | <td>Usually on day 5-10 after therapy start</td> |
</tr> | </tr> | ||
<tr> | <tr> | ||
− | <td> | + | <td>Localisation</td> |
− | <td> | + | <td>At the injection site, but also outside it!</td> |
</tr> | </tr> | ||
<tr> | <tr> | ||
− | <td> | + | <td>Therapy</td> |
− | <td> | + | <td>Discontinuation of the medication</td> |
</tr> | </tr> | ||
<tr> | <tr> | ||
− | <td> | + | <td>Healing</td> |
− | <td> | + | <td>Spontaneously, but slowly</td> |
</tr> | </tr> | ||
<tr> | <tr> | ||
− | <td> | + | <td>Safety of the context</td> |
− | <td> | + | <td>ensured</td> |
</tr> | </tr> | ||
</table> | </table> | ||
<b><i>Nicorandil</i></b> | <b><i>Nicorandil</i></b> | ||
− | + | The product is a vasomotor substance for the treatment of angina pectoris and is marketed in Great Britain, Australia and many European countries under the name Ikorel™, in Switzerland under the name Dancor™ and in other non-European countries under the names Nicoran™, Aprior™, Nitorubin™ and Sigmart™ (according to Wikipedia). Besides the known side effects including skin flush, palpitations, weakness, headache, mouth ulcers, nausea and vomiting are described. In addition, perianal, perivulvar and peristomal ulcerations have been repeatedly described (23, 24, 25). They are all refractory to treatment and have no correlation with dose or time of use (23). McKenna describes the simultaneous development of a crural ulcer on the right anterior edge of the tibia in a 73-year-old female patient with an increasing ulceration in the perianal region developing over 11 months. Nicorandil was given at 2 times 20 mg p/d 2 years before the start of the ulceration. All local measures for healing failed. | |
− | + | After discontinuation of nicorandil, there was a dramatic and rapid and lasting improvement of the ulcerations in both localisations (26). | |
− | <h3> | + | <h3>Individual case reports</h3> |
− | + | For the drugs listed below, the development of skin ulcerations is described, which occurred in strict temporal relation to the specific drug intake. Healing occurred only after the medication was discontinued. As these are only individual reports and no drug re-exposure with recurrence of skin ulcerations took place, the connection can only be regarded as possible, but not certain. An exception is only the case with propylthiouracil, but it is still an individual case.</br> | |
− | <b>Diltiazem</b> | + | <b>Diltiazem</b> known as a calcium channel blocker in the treatment of conditions such as hypertension, angina pectoris and some forms of cardiac arrhythmias. It is also used as a preventive drug for migraine (27).</br> |
− | Leflunomid | + | <b>Leflunomid</b> is used in active moderate to severe rheumatoid arthritis and psoriatic arthritis as a pyrimidine synthesis inhibitor Arava® (28). </br> |
− | <b> | + | <b>Levamisole</b> (not available in Germany!) was originally used as an antiemetic in the 60'. Later, immunomodulatory effects were discovered and it was used in the treatment of nephrotic syndrome, colon cancer and rheumatoid arthritis. In the meantime there are further indications. Among these, severe cutaneous necroses also occurred, mimicking coagulopathy (29). </br> |
− | <b> | + | <b>Nifedipine</b> (Adalat® e.a.) is a drug from the group of calcium antagonists and is used therapeutically for the treatment of arterial hypertension and Raynaud's syndrome. There is also an individual case description with the occurrence of ulcers on the lower legs (30). </br> |
− | + | Further descriptions exist for propylthiouracil. <b>Propylthiouracil</b>is a thiourea derivative and belongs to the group of thyrostatic agents. Houston describes the case of an intermittent treatment over 13 years. After increasing the dose to 50mg, a local rash and ulcers in the area of the left ankle joint appear 2 months later in an extension of 4 by 5 cm. After discontinuation, spontaneous healing occurs within 5 months. Re-exposure 5 days later results in an identical reaction with local rash and ulcera (31). | |
− | <h3> | + | <h3>Diagnostics</h3> |
− | + | Unfortunately, there is no laboratory parameter or other marker that gives us any indication of drug-induced ulcers. Only the detection of antiphospholipid antibodies and the so-called antiphospholipid-antibody syndrome causes numerous skin symptoms, including pronounced ulcers and ischemia. Drugs mentioned in connection with the occurrence of these antibodies are chlorpromazine, phenothiazine, phenytoin, hydralazine, procainamide, streptomycin, etc (32). | |
− | <h3> | + | <h3>Summary</h3> |
− | + | There are a number of drugs, usually administered orally, which can cause ulcers or necroses as a rare side effect. As with coumarins, this can occur initially in the introductory phase or, in the case of other drugs such as hydroxyurea etc., as a cumulative dose with a wide range of variation. Unfortunately, for none of the drugs we have so far been able to work out an exact pathomechanism, a typical cumulative dose or other significant parameters. | |
− | + | In addition, there are numerous drugs that can cause necrosis when administered percutaneously. However, in this case the connection is much easier to establish anamnestically. | |
− | + | For the registration of drug-induced ulcers a data entry form for a register for drug-induced wounds can be requested from the author (fax no. +49 2151 32-2005) within the scope of the Wound-healing working group of the German Dermatological Society. Only by registering further possible correlations between drug intake and the occurrence of wounds will we be able to obtain valid data on the causal relationship, pathogenesis and other criteria such as cumulative doses etc. | |
− | <h3> | + | <h3>Literature</h3> |
<ul> | <ul> | ||
<li>Körber A, Klode J, Al-Benna S, Wax C, Schadendorf D, Steinstraesser L, Dissemond J. Etiology of chronic leg ulcers in 31,619 patients in Germany analyzed by an expert survey. J Dtsch Dermatol Ges 2011; Feb;. 9(2):116-21</li> | <li>Körber A, Klode J, Al-Benna S, Wax C, Schadendorf D, Steinstraesser L, Dissemond J. Etiology of chronic leg ulcers in 31,619 patients in Germany analyzed by an expert survey. J Dtsch Dermatol Ges 2011; Feb;. 9(2):116-21</li> |
Aktuelle Version vom 17. April 2020, 15:15 Uhr
Dr. med. Thomas Horn¹
Inhaltsverzeichnis
Introduction
Wounds are caused by a variety of factors. In addition to leg ulcers, diabetic wounds and pressure ulcers, there is a wide range of other factors which can cause chronic skin defects with poor or no healing tendency. In particular, very rare causes, if not recognised, lead to non-healing wounds if the underlying cause is not eliminated through ignorance. Drugs can also cause ulcers when administered orally or injected. There are many individual publications in the literature and, for some drugs, reviews with small case numbers. The data available on such wounds is generally very limited. In a survey of experts in Germany on the genesis of chronic venous leg ulcers involving more than 31,000 patients, the incidence of drug-induced ulcers was determined at 1% (1).
Definition of a drug-induced ulcer
Drug-induced ulcers occur in temporal relation to the intake of a specific preparation. As a rule, but not always, the defect does not heal until after discontinuation. Local intensive wound treatment etc. can lead to a partial improvement of the findings, but not to healing. Ideally, evidence of drug-induced ulcus genesis is provided by conscious or frequent accidental re-exposure with renewed development of a skin ulcer after interim healing.
Issue
Many patients are multimorbid and take numerous drugs. A differentiation is then often hardly feasible. Individual case studies alone therefore hardly prove a drug-induced pathogenesis. They can only give indications of a possible connection. Only when many cases are examined together can a causal connection be proven. Histological samples from drug-induced ulcers provide no evidence of a drug-induced pathogenesis. Vasculitis, i.e. inflammatory processes in cutaneous or subcutaneous vessels, thrombus formation with occlusion of vessels or other pathomechanisms which lead to skin defects via secondary infarction, can be caused by a variety of causes in addition to drugs.
Known examples of drug-induced ulcers
Hydroxyurea Hydroxyurea is a hydroxylated urea derivative and has been used as a cytostatic drug since 1916 (2). Side effects are reported in the literature between 10 and 35% (3). The frequency of the occurrence of skin ulcers under hydroxyurea therapy is stated extremely differently in the literature. It varies from less than 0.1% (4) up to 12% (5).
It is possible that there are still disease-related incidence differences. Najean and Chaine report that 8-9% of patients with myelodysplastic diseases and 29% of patients with sickle cell anaemia developed leg ulcera taking Hydroxyurea (6,7). The initial description was in 1986 as part of a long-term therapy with Hydroxyurea in chronic myeloid leukaemia (8). Hydroxyurea, also known as hydroxycarbamide, causes inactivation of ribonucleotide reductase with inhibition of cellular DNA synthesis and cell death in the S-phase. It is therefore used as a cytostatic agent for leukemias and myeloproliferative diseases. In addition, its use is known to be effective in sickle cell anaemia. It has also been used in the antiretroviral treatment of HIV infections. Best e. a. describe in 1998 14 own patients with hydroxyurea-induced ulcers from a set of 115 patients with different myeloproliferative diseases with the intake of hydroxyurea. Their average age was 65 (41-74). Usually, the ulcers were located in the ankle joint region, but occasionally also above the tibia, the back of the foot or on the lower leg. In 10 cases on the inner ankle and 8 cases on the outer ankle, there was no localisation preference for the extremely painful ulcers. 64% of patients had multiple ulcerations. The mean cumulative dose before development of the ulcers was 2.3 kg (1.4-5.5 kg). The mean duration of administration was 6.1 years (2-15 years). The ulcers healed in 12 of 14 patients after discontinuation of treatment, in 2 patients an additional split skin transplant was necessary. In one case the dose was reduced from 1.5 to 0.5 hydroxyurea. Healing occurred within 5 months. Another patient had an improvement in the reduction of the therapy from 2 to 1 g, when the dose was increased again to 2.5 g the original ulcer deteriorated over 4 months and others appeared, so that the therapy had to be discontinued (5). From this and many other literature references it is clear that discontinuation of therapy for hydroxyurea-induced ulcers is necessary. Healing then usually occurs slowly but spontaneously.
Histology does not provide decisive diagnostic changes. In the dermis, cell swelling, oedema, fatty degeneration of the vascular walls and perivascular lymphocytic infiltrations without vasculitis can be detected. Overall, the signs correspond to those of a cutaneous occlusive vasculopathy. In later stages dermal fibrosis is described (5).
The size and extent of the ulcerations do not correlate with the amount of hydroxyurea.
In 2 of the 14 patients Best describes the recurrence of ulcers after re-exposure. One re-exposure took place 3 years later, and a few months later multiple ulcers occurred in both ankle joints, which did not heal until 6 months after re-exposure. In the second patient, it took 4 months after re-exposure to develop new cutaneous ulcerations (5). Thus the genesis of hydroxyurea-induced leg ulcerations is confirmed.
Substance | Hydroxyurea |
Appearance | After a latency of several years |
Localisation | Mainly malleolar, but also other localisations |
Complaints | Extreme painfulness |
Therapy | Discontinuation of the medication |
Healing | Spontaneously, but slowly |
Safety of the context | ensured |
Anagrelide Anagrelide is an imidazoquinazoline, which was approved in Germany in 2005 for essential thrombocythemia. Its inhibitory effect on human platelet formation is achieved by delaying megakaryocyte maturation through inhibition of cyclic AMP phosphodiesterase III. Commonly described side effects are headache, diarrhoea, palpitation, weakness, pain, dizziness, abdominal pain and nausea. However, the vast majority of side effects resolve within several weeks (9).
Due to its excellent safety profile, Anagrelide has been shown to be a first-line treatment for myeloproliferative diseases associated with plateletosis, including primary thrombocythemia.
Ruiz-Argüelles e.a. describe a 74-year-old patient who had already developed a leg ulcer under hydroxyurea (1g/d), which healed within 4 months after discontinuation. 3 years later the patient received Anagrelide (1g/d), 5 months later extreme pain in the right lateral malleolus reappeared. Two further weeks later, an ulcer developed in exactly the same localisation as the hydroxyurea-induced ulcer that had occurred previously (10).
Rappoport also describes a patient who, after 3 years of taking 0.5 g Hydroxyurea p/d for 6 weeks after conversion to 1g Anagrelide p/d, developed extremely painful erythema in the area of both lateral ankle joints, which subsequently became ulcerated. Basically, these ulcers could be the result of long-term use of Hydroxyurea or an expression of the newly initiated therapy with Anagrelide. However, as they only healed completely after the therapy with hydroxyurea, there is also the suspicion of Anagrelide-induced leg ulcers (9).
There are further reports of Anagrelide-induced leg ulcers, but no large case numbers. Thus, even under the substitute drug for hydroxyurea, namely Anagrelide as a first-line as well as second-line therapy for myeloproliferative diseases, as well as for hydroxyurea, identical ulcers can develop, sometimes even in loco (10)!
Substance | Anagrelide |
Appearance | After a latency of several month |
Localisation | Mainly malleolar, but also other localisations |
Complaints | Extreme painfulness |
Therapy | Discontinuation of the medication |
Healing | Spontaneously, but slowly |
Safety of the context | ensured |
Coumarins Coumarins are drugs used for anticoagulation. We distinguish between Phenprocoumon (Marcumar®, Falithrom®, Phenprogamma®, etc.), Warfarin (Coumadin®, etc.), and Acenocoumarol (Sintrom®). There are also reports of the use of other coumarins such as Acoumarol (11).
All coumarins may cause skin ulcerations.
The most common ones are phenprocoumon with a half-life of 10-14 days and warfarin with a half-life of 2 days. Typically, coumarins decrease the activity and synthesis of coagulatory factors II, VII, IX, X and inhibit the formation of protein C and S, both of which have anticoagulatory effects. However, since these factors have different half-lives, there is an imbalance between anticoagulatory and procoagulatory factors. Factor VII and protein C have a short half-life. When warfarin is given initially, factor VII and protein C levels go down first. Factor IX decreases more slowly, followed by factor X, protein S and finally factor II. This can cause a temporary hypercoagulable state with inhibition of anticoagulant protein C production. In individuals with congenital protein C deficiency, this temporary hypercoaguability may be more pronounced. Patients with heterozygous protein C deficiency have a low level of protein C and thus a higher risk of suffering from WISN (warfarin-induced skin necrosis) when treated with Warfarin. Other hypercoaguable conditions including factor V Leiden mutation, antithrombin III deficiency and lupus anticoagulant detection are also associated with WISN. In contrast, other deficiencies of vitamin K such as under chemotherapy, liver disease, elimination of vitamin K-producing intestinal flora by antibiotics are rarely sufficient to cause WISN alone, but may increase the risk in hypercoagulable patients. The most likely mechanism of WISN is microvascular thrombosis, but hypersensitivity reactions to warfarin and direct damage to capillaries leading to dilation and rupture of vessels are also discussed as a possible pathomechanism. Typical affected patients are middle-aged, overweight women who have been treated for deep vein thrombosis, pulmonary embolism. The typical onset is day 3 - day 10 after starting warfarin therapy, but late onset up to 18 months, even 15 years, has been described (12). However, these late event data should be viewed critically, as a more detailed review of the literature may also involve discontinuation and then reinsertion of the drug (13).
Skin necrosis on coumarin administration is rare overall and occurs in about one in 5000 patients and about 3% of people with protein C deficiency. Typically affected are the fat-rich skin areas such as chest, hips and legs, very rarely also internal organs such as adrenal glands, uterus and the male genitals (14).
Although the occurrence of warfarin initiation has been described, the initiation phase appears to be only a partial factor in its genesis. Koduri and Steward each independently describe a patient who did not develop necrosis until the second administration of warfarin (13, 15). The multiple chronologically staggered occurrence is also described (15). The first description was made by Flood in 1943 with the presentation of thrombophlebitis migrans disseminata with development of gangrene on the female breast (16).
Comparison of bleeding with necrosis in patients receiving coumarin therapy (17) | |
Bleeding | Necrosis |
Affects men and women equally | Female dominance |
Start not in relation to the start of therapy | Start on the 3rd-6th day after starting coumarin therapy |
correction with the administration of vitamin K | Increase despite administration of vitamin K |
Continuation of coumarin administration worsens the complication | The continuation of coumarin administration has no effect on the extent of the complication |
Heparin worsens the complication | Heparin can be helpful |
Missing necroses | Existing necrosis |
Surgical interventions not necessary | Surgical interventions often necessary |
Heparins Allergic skin reactions or skin necrosis caused by heparins or low-molecular heparins are rarely but probably underestimated because too little is reported. The etiology of heparin-induced skin necrosis is unknown, the occurrence is often associated with HIT type II. This is an immunological reaction that results in binding of heparin to platelet factor 4 (PF-4), which leads to production of heparin-dependent IgG, IgM and IgA antibodies. Elevated IgG antibody titers are observed in patients with heparin-induced skin lesions, regardless of whether they develop thrombocytopenia. This suggests an immunological mechanism (18).
The broad spectrum of cutaneous reactions ranges from erythematous itching areas to large symptomatic plaques and heparin-induced skin necrosis. Women are more frequently affected than men. These are local necroses in the area of the application sites (19). But necroses outside the injection sites have also been described (20)!
Heparin-induced skin necrosis starts between day 5 and day 10 after the start of intravenous subcutaneous heparin administration. A later onset until day 16 has been reported (20). Skin necrosis can occur in patients with unfractionated heparin as well as in patients with LMW-heparin (21, 22). The necroses begin with small erythematous painful lesions, which later spread to necrotic areas. Histologically, microvascular thrombi are found in small vessels with minimal inflammatory response (21). Cross-reactions between heparin and low molecular weight heparin are observed (19).
Therapy recommendation: Therapy must be discontinued if systemic allergic reactions occur. The anticoagulant therapy should be changed to hirudin and/or coumarins. Appropriate skin tests should be performed. Depending on these tests, the anticoagulatory therapy can then be discontinued or modified. Alternatively, the new oral thrombin inhibitors etc. are alternative treatment methods.
Substance | Heparins |
Heparins | Unfractionated, also LMWH, cave cross reactions |
Time occurrence | Usually on day 5-10 after therapy start |
Localisation | At the injection site, but also outside it! |
Therapy | Discontinuation of the medication |
Healing | Spontaneously, but slowly |
Safety of the context | ensured |
Nicorandil The product is a vasomotor substance for the treatment of angina pectoris and is marketed in Great Britain, Australia and many European countries under the name Ikorel™, in Switzerland under the name Dancor™ and in other non-European countries under the names Nicoran™, Aprior™, Nitorubin™ and Sigmart™ (according to Wikipedia). Besides the known side effects including skin flush, palpitations, weakness, headache, mouth ulcers, nausea and vomiting are described. In addition, perianal, perivulvar and peristomal ulcerations have been repeatedly described (23, 24, 25). They are all refractory to treatment and have no correlation with dose or time of use (23). McKenna describes the simultaneous development of a crural ulcer on the right anterior edge of the tibia in a 73-year-old female patient with an increasing ulceration in the perianal region developing over 11 months. Nicorandil was given at 2 times 20 mg p/d 2 years before the start of the ulceration. All local measures for healing failed. After discontinuation of nicorandil, there was a dramatic and rapid and lasting improvement of the ulcerations in both localisations (26).
Individual case reports
For the drugs listed below, the development of skin ulcerations is described, which occurred in strict temporal relation to the specific drug intake. Healing occurred only after the medication was discontinued. As these are only individual reports and no drug re-exposure with recurrence of skin ulcerations took place, the connection can only be regarded as possible, but not certain. An exception is only the case with propylthiouracil, but it is still an individual case.
Diltiazem known as a calcium channel blocker in the treatment of conditions such as hypertension, angina pectoris and some forms of cardiac arrhythmias. It is also used as a preventive drug for migraine (27).
Leflunomid is used in active moderate to severe rheumatoid arthritis and psoriatic arthritis as a pyrimidine synthesis inhibitor Arava® (28).
Levamisole (not available in Germany!) was originally used as an antiemetic in the 60'. Later, immunomodulatory effects were discovered and it was used in the treatment of nephrotic syndrome, colon cancer and rheumatoid arthritis. In the meantime there are further indications. Among these, severe cutaneous necroses also occurred, mimicking coagulopathy (29).
Nifedipine (Adalat® e.a.) is a drug from the group of calcium antagonists and is used therapeutically for the treatment of arterial hypertension and Raynaud's syndrome. There is also an individual case description with the occurrence of ulcers on the lower legs (30).
Further descriptions exist for propylthiouracil. Propylthiouracilis a thiourea derivative and belongs to the group of thyrostatic agents. Houston describes the case of an intermittent treatment over 13 years. After increasing the dose to 50mg, a local rash and ulcers in the area of the left ankle joint appear 2 months later in an extension of 4 by 5 cm. After discontinuation, spontaneous healing occurs within 5 months. Re-exposure 5 days later results in an identical reaction with local rash and ulcera (31).
Diagnostics
Unfortunately, there is no laboratory parameter or other marker that gives us any indication of drug-induced ulcers. Only the detection of antiphospholipid antibodies and the so-called antiphospholipid-antibody syndrome causes numerous skin symptoms, including pronounced ulcers and ischemia. Drugs mentioned in connection with the occurrence of these antibodies are chlorpromazine, phenothiazine, phenytoin, hydralazine, procainamide, streptomycin, etc (32).
Summary
There are a number of drugs, usually administered orally, which can cause ulcers or necroses as a rare side effect. As with coumarins, this can occur initially in the introductory phase or, in the case of other drugs such as hydroxyurea etc., as a cumulative dose with a wide range of variation. Unfortunately, for none of the drugs we have so far been able to work out an exact pathomechanism, a typical cumulative dose or other significant parameters. In addition, there are numerous drugs that can cause necrosis when administered percutaneously. However, in this case the connection is much easier to establish anamnestically.
For the registration of drug-induced ulcers a data entry form for a register for drug-induced wounds can be requested from the author (fax no. +49 2151 32-2005) within the scope of the Wound-healing working group of the German Dermatological Society. Only by registering further possible correlations between drug intake and the occurrence of wounds will we be able to obtain valid data on the causal relationship, pathogenesis and other criteria such as cumulative doses etc.
Literature
- Körber A, Klode J, Al-Benna S, Wax C, Schadendorf D, Steinstraesser L, Dissemond J. Etiology of chronic leg ulcers in 31,619 patients in Germany analyzed by an expert survey. J Dtsch Dermatol Ges 2011; Feb;. 9(2):116-21
- Dissemond J., e.a. Leg ulcer in a patient associated with hydroxyurea therapy. Int J Derm 2006; 45: 154-160.
- Bravo Blanco AM, Moral JL, Tembrás S, Gómez R, Ulibarrena C, Obdulia Vazquez M. Hydroxyurea-induced skin ulcerations in patients with chronic myeloproliferative disorders. J Hematother Stem Cell Res. 2003 Aug;12(4):461-3.
- Ravandi-Kashani F, Cortes j, Cohen P, Talpaz M, O´Brien S, Markowitz A, Kantarjian H. Cutaneous ulcers associated with hydroxyureatherapy in myeloproliferative disorders. Leuk Lymphoma 1999;35: 109-18
- Best PJ, Daoud MS, Pittelkow R, Petitt RM. Hydroxyurea-induced leg ulcerations in 14 patients. Ann Intern Med 1998; 128: 29-32
- Najean Y, Rain JD. Treatment of polycythaemia vera: the use of hydroxyurea and pipobroman in 292 patients under the age of 65 years. Blood 1997; 90: 3370-77.
- Chaine B, Neonato MG, Girot R, et al. Cutaneous adverse reactions to hydroxyurea in patients with sickle cell disease. Arch Dermatol 2001; 137: 467-70.
- Montefusco E, Alimena G, Gastaldi R, Carlesimo OA, Valesini G, Mandelli F. Unusual dermatologic toxicity of long-term therapy with hydrxyurea in chronic myelogenous leukemia. Tumori 1986; 72:317-321
- Rappoport L, Körber A, Grabbe S, Dissemond J. Auftreten von Ulcera crurum in Zusammenhang mit der Einnahme von Anagrelid. DMW 2007; 132:319-321.
- Ruiz-Argüelles GJ, Ruiz-Delgado GJ, Ruiz-Reyes G, Chernoff SG. Anagrelide-induced relapse of a hydroxyurea-induced leg ulcer in a patient with primary thrombocythemia. Mayo Clin Proc 1998 Nov;73(11):1125.
- Nalbandian RM, Mader IJ, Barrett JI, Pearce JF, Rupp EC. Petechiae, Ecchymosis, and Necrosis of Skin induced by Coumarin Congeners. JAMA 1965; vol 192, 7, 107-112.
- Chacon G, Nguyen T, Khan A, Sinha A, Maddirala S. Warfarin-induced skin necrosis mimicking calciphylaxis: a case report and review of the literature. J Drugs Dermatol. 2010 Jul; 9(7):859-63.
- Koduri PR, Parveez R.Lack of compliance and late-onset warfarin-induced skin necrosis. Am J Hematol 1999 Jul; 61(3):218-9.
- Gallerani M, Manfredini R, Moratelli S Non-haemorrhagic adverse reactions of oral anticoagulant therapy. Int J Cardiol. 1995; Mar 24;49(1):1-7.
- Stewart A, Penman I, Cook M, Ludlam C. Warfarin-induced skin necrosis. 1998, 233-235
- Flood EP, Reddish MH, Bociek SJ, Shapiro S. Thrombophlebitis migrans disseminate: a report of a case in which gangraene of the breast occurred. NY S J Med 1943; 43: 1121-24.
- Brooks LWjr, Blais FX. Coumarin-induced skin necrosis. J Am Osteopath Assoc 1991; 91: 601-05
- Warkentin TE. Heparin-induced skin lesions. B J Haematol 1996; 92 (2): 494-7.
- Harenberg J, Hoffmann U, Huhle G, Winkler M, Bayerl C, , Cutanous reactions to anticoagulants – recognition and management. Am J Clin Dermatol 2001; 2: 69-75
- Tietge UJ, Schmidt HH, Jäckel E, Trautwein C.Low molecular weight heparin-induced skin necrosis occurring distant from injection sites and without thrombocytopenia. J Intern Med 1998; 243: 313-15
- Jörg I, Fenyvesi T, Harenberg J. Anticoagulant-related skin reactions.Expert Opin Drug Saf. 2002 Sep; 1(3):287-94.
- Ojeda, E, Perz MC, Mataix R, Arbelo A, Jiminez S, Campo C, Balda I. Skin necrosis with a low molecular weight heparin. Brit J Haem 1992; 82: 620.
- Abdelrazeq AS, Owen C, Smith L, McAdam JG, Pearson HJ, Leveson SH. Nicorandil-associated para-stomal ulceration: Case series. Eur J Gastroenterol Hepatol 2006; 18(12):1293-5.
- Baker RP, Al-Kubati W, Atuf M, Phillips RK. Nicorandil-induced severe perianal ulceration. Tech Coloproctol. 2007; Dec 11(4):343-5. Epub 2007 Dec 3.
- Claeys A, Weber-Muller F, Trechot P, Cuny JF, Georges MY, Barbaud A, Schmutz JL. Cutaneous, perivulvar and perianal ulcerations induced by nicorandil. Br J Dermatol. 2006; Aug 155(2):494-6.
- McKenna DJ, Donnelly J, Armstrong DK. Nicorandil-induced leg ulceration. Br J Dermatol. 2007; Feb156(2):394-6.
- Carmichael AJ, Paul CJ.Vasculitic leg ulcers associated with diltiazem. BMJ. 1988; 297(6647):562.
- Jakob A, Porstmann R, Rompel R.Skin ulceration after leflunomide treatment in two patients with rheumatoid arthritis. J Dtsch Dermatol Ges. 2006 Apr;4(4):324-7.
- Farhat EK, Muirhead TT, Chaffins ML, Douglass MC. Levamisole-induced cutaneous necrosis mimicking coagulopathy. Arch Dermatol 2010 Nov;146(11):1320-1.
- Luca S, Romeo S. Edema ed ulcere cutanee degli arti inferiori quale effeto collaterale della nifedipina. Minerva Cardioangiolog 1999; 47: 219-22.
- Houston BD, Crouch ME, Brick JE, DiBartolomeo AG. Apparent vasculitis associated with propylthiouracil use. Arthritis Rheum. 1979 Aug; 22(8):925-8.
- Nahass GT. Antiphospholipid antibodies and the antiphospholipid antibody syndrome. 1997; 36, 150-168.