eTable. An inventory of calciphylaxis patients with combined thrombophilias. They are categorized by number of hypercoagulable conditions with a description of the co-occurring thrombophilias.
Customize your JAMA Network experience by selecting one or more topics from the list below.
Dobry AS, Ko LN, St John J, Sloan JM, Nigwekar S, Kroshinsky D. Association Between Hypercoagulable Conditions and Calciphylaxis in Patients With Renal Disease: A Case-Control Study. JAMA Dermatol. 2018;154(2):182–187. doi:10.1001/jamadermatol.2017.4920
Are there hypercoagulable conditions that are risk factors for development of calciphylaxis?
In this matched case-control analysis of 152 individuals with chronic kidney disease, lupus anticoagulant, protein C deficiency, and combined thrombophilia were all found to be significantly associated with calciphylaxis development.
Clinicians should be aware of these associations in patients with impaired kidney function, and may consider increased screening and appropriate anticoagulation treatment to reduce the risk of calciphylaxis development.
Calciphylaxis is a rare skin disease with high morbidity and mortality that frequently affects patients with renal disease. Hypercoagulable conditions are frequently observed in both patients with calciphylaxis and those with chronic kidney disease (CKD), complicating our understanding of which hypercoagulable conditions are specific to calciphylaxis.
To identify hypercoagulable conditions that are risk factors for developing calciphylaxis while controlling for CKD.
Design, Setting, and Participants
This was a case-control study, comparing the hypercoagulability status of patients with calciphylaxis and with renal disease with that of a matched control population at 2 large urban academic hospitals in Boston, Massachusetts. Retrospective medical record review of laboratory values was performed to identify patients with hypercoagulable conditions. Case and control patients were further stratified based on both severity of CKD and warfarin. Patients with a dermatologic diagnosis of calciphylaxis between 2006 and 2014 and concomitant CKD were included as cases (n = 38). Three controls (n = 114) per case patient with CKD were included, and were matched by age, sex, and race.
Main Outcomes and Measures
The rate of various hypercoagulable states (ie, antithrombin III [ATIII] deficiency, protein C and S deficiency, factor V Leiden mutation, prothrombin gene mutation [G20210A], elevated factor VIII level, lupus anticoagulant, anti-IgG or IgM cardiolipin antibodies, heparin-induced thrombocytopenia antibodies, and elevation of homocysteine) in patients with calciphylaxis compared with their matched controls.
Of the calciphylaxis cohort, 28 (58%) were female and 18 (55%) were non-Hispanic white. Among all patients, lupus anticoagulant (13 [48%] positive in cases vs 1 [5%] in controls; P = .001), protein C deficiency (9 [50%] vs 1 [8%]; P = .02), and combined thrombophilias (18 [62%] vs 10 [31%]; P = .02) were found to be significantly associated with calciphylaxis. In a subanalysis of patients with stage 5 CKD, only lupus anticoagulant (12 [53%] vs 9 [0%]; P = .01) and combined thrombophilia (15 [63%] vs 1 [8%]; P = .004) remained significantly associated with calciphylaxis. In a separate subanalysis of warfarin-unexposed patients, only lupus anticoagulant (7 [50%] vs 1 [6%]; P = .01) and protein C deficiency (5 [46%] vs 10 [0%]; P = .04) remained significantly associated with calciphylaxis.
Conclusions and Relevance
Presence of lupus anticoagulant and combined thrombophilias are risk factors for the development of calciphylaxis in patients with late-stage renal disease. Clinicians should be aware of these associations in patients with impaired kidney function and may consider increased screening and appropriate anticoagulation treatment to reduce the risk of calciphylaxis development.
Calciphylaxis, also known as calcific uremic arteriolopathy, is a rare disease characterized by vascular calcification and cutaneous necrosis. These chronic, nonhealing wounds are often complicated by infection and sepsis, leading to high mortality. Calciphylaxis typically manifests in patients with late-stage renal disease, with incidence rates estimated to be 1% to 4% in this population.1,2 The etiology and pathogenesis of calciphylaxis remain poorly understood. Case-control studies have identified female sex, obesity, elevated calcium-phosphate product level, hyperphosphatemia, systemic corticosteroid use, and treatment with calcium salts and vitamin D as risk factors for calciphylaxis.3-6 Although many patients with renal disease present with these abnormalities, only a small number develop calciphylaxis. Thus, the rarity of calciphylaxis suggests an additional causative mechanism.
Hypercoagulability has been posited as a potential risk factor for calciphylaxis.7 This hypothesis arose from numerous case reports and case series of patients with calciphylaxis with hypercoagulable conditions.8-12 Although descriptive studies have suggested that hypercoagulable states may predispose patients to calciphylaxis, lack of comparison to a control group has potential to overestimate the association. It is noteworthy that thrombophilias are also present at high rates in patients with chronic kidney disease (CKD), a group that significantly overlaps with patients with calciphylaxis, further complicating study of hypercoagulable risk factors.13,14 To date, rigorous case-control studies on thrombophilias in patients with calciphylaxis are limited. To better understand the relationship between hypercoagulability and calciphylaxis, a case-control study was performed in which 38 patients diagnosed as having calciphylaxis were compared with 114 control patients matched by age, sex, race, and diagnosis with CKD.
The Partners Healthcare Research Patient Data Repository (RPDR), a medical records database, was used to identify study cases and controls. Cases included individuals with both a clinical diagnosis of calciphylaxis by a dermatologist (given between 2006 and 2014 at either Massachusetts General Hospital or Brigham and Women’s Hospital) and CKD. Patients with calciphylaxis were identified in the RPDR using both the International Classification of Diseases, Ninth Revision, code 275.49, “Other Disorders of Calcium Metabolism,” and skin biopsy procedure codes, and administrative billing codes for CKD were used to capture those with concurrent renal dysfunction. Ultimately, 38 patients with both a calciphylaxis diagnosis by a dermatologist and a diagnosis of CKD were identified. Three controls per case patient (n = 114) were identified using the same search criteria for CKD within the RPDR and were further matched by age, sex, and race. The study was approved by the Partners Healthcare institutional review board.
A retrospective medical record review of all cases and controls were performed. All study data were securely collected and managed using REDCap Electronic Data Capture website.15 For each case, the index date was defined as the date when individuals presented with calciphylaxis. This index date was assigned to the corresponding controls. The following demographic data were abstracted: age at presentation, sex, race, and comorbidities prior to presentation (eg, coronary artery disease, diabetes mellitus types 1 and 2, alcohol abuse, hypertension, hyperparathyroidism, hepatobiliary disease, renal disease, obesity, and smoking history). Level of CKD was determined by value of glomerular filtration rate (GFR) at the date closest to index date. Stages 1 to 4 CKD (CKD1-4) were defined as a GFR of 15 mL/min or greater, while stage 5 CKD (CKD5) was defined as a GFR of less than 15 mL/min (inclusive of patients receiving dialysis). Patients who were prescribed warfarin within 6 months of index date were defined as having warfarin exposure.
Hypercoagulability status was determined based on selected laboratory results within each patient’s medical record. These included laboratory assessments for deficiencies of antithrombin III (ATIII) and proteins C and S, activated protein C resistance (APCR) screening with follow-up genotyping for factor V Leiden mutation, prothrombin mutation (G20210A), elevated factor VIII level, presence of lupus anticoagulant, presence of anti-IgG or IgM cardiolipin antibodies, presence of heparin-PF4 antibodies, and elevation of homocysteine levels. Medical records of patients who received evaluation of ATIII, protein C, and protein S were reviewed to ensure that they were not prescribed warfarin at the time of testing. If patients were prescribed warfarin at time of these tests, the ATIII, protein C, and protein S results were removed from the analysis. Because warfarin can continue to affect test results even if discontinued several months prior, case and control patients were further stratified based on whether they took warfarin within 6 months of presentation. Of note, hypercoagulability testing was clinician-determined, and as a result testing was not uniform. Indications for hypercoagulability testing likely varied between cases and controls.
Demographic data (ie, age, sex, race, and comorbidities) and anticoagulation exposure of calciphylaxis patients were compared with a control population by χ2 analysis or Fisher exact test using GraphPad Prism software (GraphPad Software Inc). Owing to rarity of hypercoagulable laboratory panels, cases and controls were compared by whole-group analysis rather than by matched case-control analysis. Hypercoagulable tests were assessed via Fisher exact test, with further subanalysis of 71 patients with CKD5 and 114 non–warfarin-exposed patients. For all of these statistical tests, a 2-sided P < .05 was considered significant.
The baseline demographics and comorbidities of patients with calciphylaxis were compared with those of control patients (Table 1). Of the calciphylaxis cohort, 28 [58%] were female, and 18 [55%] were non-Hispanic white. Patients with calciphylaxis were observed to have higher rates of coronary artery disease (32 [84%] vs 60 [53%]; P < .001), diabetes mellitus type 2 (25 [66%] vs 43 [38%]; P = .04), and hyperparathyroidism (27 [71%] vs 35 [31%]; P < .001) compared with controls, which is not surprising given that diabetes and hyperparathyroidism are known risk factors for calciphylaxis. There was no significant difference in body mass index, alcohol abuse, cigarette use, hypertension, diabetes mellitus type 1, or hepatobiliary disease between patients with calciphylaxis and control patients. Although all cases and controls had CKD, patients with calciphylaxis had greater severity of renal disease because they were more likely to have CKD5 compared with controls (28 [74%] vs 43 [50%]; P = .02).
Table 2 demonstrates results of both laboratory results that test for presence of hypercoagulable conditions as well as recent anticoagulation medication exposure. Patients with calciphylaxis were significantly more likely than controls to have lupus anticoagulant (13 [48%] vs 1 [5%]; P = .001), protein C deficiency (9 [50%] vs 1 [8%]; P = .02), and combined thrombophilia (18 [62%] vs 10 [31%]; P = .02). Of those who had combined thrombophilias, patients with calciphylaxis had a range of 2 to 6 positive hypercoagulable test results (eTable in the Supplement). There was no significant difference in rates of ATIII deficiency, anticardiolipin antibodies, factor V Leiden, heparin platelet-factor 4 (PF4) antibodies, elevated homocysteine level, protein S deficiency, and prothrombin gene mutation between cases and controls. Patients with calciphylaxis had a slightly higher incidence of having any positive hypercoagulable condition compared with controls, but this difference was not significant (26 [74%] vs 35 [64%]; P = .36). Individuals with calciphylaxis were significantly more likely to have been exposed to warfarin (19 [50%] vs 9 [9%]; P < .001) in the 6 months prior to diagnosis compared with controls. Only patients with medication records available during that period were included.
Table 3 demonstrates a subanalysis of cases and controls with CKD5 (n = 28 for cases, n = 43 for controls). Within this analysis, patients with calciphylaxis were significantly more likely than controls to have lupus anticoagulant (12 [52%] vs 9 [0%]; P = .01) and combined thrombophilia (15 [63%] vs 1 [8%]; P = .004), while there was no longer a significant difference in incidence of protein C deficiency (8 [50%] vs 4 [0%]; P = .12).
Table 4 displays a subanalysis of patients who were not prescribed warfarin in the 6-month period prior to index date (n = 19 for cases, n = 95 for controls). Lupus anticoagulant (7 [50%] vs 1 [6%] for controls; P = .01) and protein C deficiency (5 [46%] vs 10 [0%]; P = .04) were significantly associated with calciphylaxis. However, after adjusting for severity of CKD, only lupus anticoagulant was found to be significantly associated with calciphylaxis (P = .04).
In this case-control study evaluating hypercoagulable conditions in 38 patients with calciphylaxis with renal disease, we discovered significant associations between development of calciphylaxis and presence of lupus anticoagulant, protein C deficiency, and combined thrombophilia. Subgroup analysis of patients with CKD5 demonstrated that presence of either lupus anticoagulant or a combined thrombophilia remained significantly associated with development of calciphylaxis. An alternative subgroup analysis of patients without recent warfarin exposure showed that lupus anticoagulant and protein C deficiency remained significantly associated with calciphylaxis. In these analyses, there was no significant difference in the other laboratory tests for hypercoagulable conditions, such as ATIII deficiency, anticardiolipin antibodies, elevated factor VIII, and heparin-PF4 antibodies, or mutations in prothrombin or factor V Leiden genes.
This study further evaluates why some patients with renal disease develop calciphylaxis while others do not. Although the pathogenesis of calciphylaxis is poorly understood, the current theory is that the disease is caused by cutaneous arteriolar stenosis with subsequent thrombotic occlusion, leading to ischemic skin necrosis.6 Arteriolar calcification is thought to be a result of bone mineral derangements, which are frequently present in patients with severe renal disease.16 However, the cause of the secondary thrombotic occlusion is less clear and may be caused by altered blood flow within calcified vessels, a genetic or acquired state of hypercoagulability, or a combination of the 2. An association between hypercoagulability and calciphylaxis has been previously suggested by a case series that showed thrombophilic conditions (specifically antithrombin deficiency, lupus anticoagulant, and protein S deficiency) co-occur with calciphylaxis at high frequencies, although these patients were not matched to a comparison group.12 The only hypercoagulable laboratory values evaluated to date in case-control studies have been protein C activity, protein S antigen, ATIII activity, and homocysteine levels, all of which had no significant association with calciphylaxis, similar to our findings (we similarly did not find an association between protein C deficiency and calciphylaxis in patients with CKD5).17
Our study suggests that both lupus anticoagulant and combined thrombophilia may uniquely contribute to the underlying pathogenesis of calciphylaxis in patients with CKD5. Lupus anticoagulant has been suggested to exert its prothrombotic effects by inducing vitamin K deficiency, and can be associated with antiphospholipid syndrome. Interestingly, patients with antiphospholipid syndrome typically develop deep vein thrombosis, whereas patients with calciphylaxis develop arterial thrombosis.18 Although lupus anticoagulant can be associated with lupus erythematosus, we did not observe a significant difference in rates of lupus erythematosus between cases and controls (1 case and 1 control carried this diagnosis, with another control having concern for drug-induced lupus). Combined thrombophilias were also significantly associated with calciphylaxis development, perhaps because multiple points of the coagulation cascade are affected, creating a greater overall likelihood of thrombosis. In addition, although protein C deficiency was significantly associated with calciphylaxis when comparing patients of all stages of CKD, this association no longer held when comparing patients with CKD5, suggesting that protein C deficiency may be a function of variable alterations in protein C with hemodialysis.19 As expected, we also observed a significant association between warfarin exposure and calciphylaxis, which is in line with the current understanding of warfarin as a well-known risk factor, with warfarin-associated calciphylaxis being described as a distinct type of calciphylaxis.20-22 Subset analysis of the calciphylaxis cohort not receiving warfarin in the 6 months prior to disease presentation revealed that lupus anticoagulant and protein C deficiency conferred increased likelihood of developing calciphylaxis. Overall, these results indicate that presence of lupus anticoagulant is strongly associated with the development of calciphylaxis, even in individuals with severe renal dysfunction who are not using warfarin.
The major strength of our study is that to our knowledge it is the most comprehensive case-control evaluation of hypercoagulable traits to date with matching by diagnosis with CKD, a diagnosis within itself that can be associated with alterations in coagulability.14 Although case-control evaluations are the first step in identifying potential risk factors, we recognize that both larger, prospective case studies and improved pathophysiological understanding are needed to confirm the hypercoagulable conditions associated with calciphylaxis in this study as true risk factors. A limitation of this study is that we matched by diagnosis with CKD rather than by CKD5 or dialysis dependence. We remedied this via a subgroup analysis of CKD5 case patients and controls. Another limitation is that controls were not as frequently tested for hypercoagulable conditions as case patients. Patients with calciphylaxis are more often tested for hypercoagulable states based on current recommendations, whereas case patients were likely tested only if there was clinical suspicion of a prothrombotic state. As a result, hypercoagulability laboratory tests were collected at lower rates in the control patients. This may have reduced our ability to capture true differences and led to an underestimation of the association between certain hypercoagulable conditions and calciphylaxis. However, documented rates of thrombophilias in the general population with CKD are similar to rates observed in this study, suggesting that information bias is more limited.13
Our findings suggest that presence of lupus anticoagulant and combined thrombophilia are both risk factors for the development of calciphylaxis in patients with CKD5. Prospective screening for these factors may help identify patients with CKD at increased risk for developing calciphylaxis who would benefit from tighter metabolic control and increased screening. In addition, these data may provide insight for improved prevention and treatment of calciphylaxis, possibly by better managing thrombophilic states in appropriate patients. Specifically, preliminary data have suggested that use of novel oral anticoagulants confer a survival advantage in patients with calciphylaxis.23 Clinicians should be aware of these associations in patients with impaired kidney function, and can consider modifying anticoagulation treatment so as to reduce the risk of calciphylaxis development.
Corresponding Author: Daniela Kroshinsky, MD, MPH, Department of Dermatology, Massachusetts General Hospital, 50 Staniford St, Second Floor, Boston, MA 02114 (firstname.lastname@example.org).
Accepted for Publication: September 26, 2017.
Published Online: December 13, 2017. doi:10.1001/jamadermatol.2017.4920
Author Contributions: Drs Kroshinsky and Dobry had full access to all of the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis.
Study concept and design: All authors.
Acquisition, analysis, or interpretation of data: All authors.
Drafting of the manuscript: All authors.
Critical revision of the manuscript for important intellectual content: All authors.
Administrative, technical, or material support: St John, Kroshinsky.
Study supervision: Kroshinsky.
Conflict of Interest Disclosures: None reported.
Additional Contributions: We are indebted to Sowmya R. Rao, PhD, and Hang Lee, PhD, from the Harvard Catalyst Biostatistics Service for assistance with statistical analysis. Dr Rao was paid for her time and assistance; Dr Lee provided a free statistical consultation.
Create a personal account or sign in to: