Reduced Vitamin C Levels in Patients with Bleeding of Unknown Cause and Generalized Joint Hypermobility

Introduction

Vitamin C is essential for the synthesis of collagen. Ehler-Danlos syndrome (EDS) is a group of collagen disorders associated with generalized joint hypermobility (HM), bleeding and bruising. The genetic cause of the HM EDS subtype (hEDS) is unknown. Referral to haematology clinics may result in the diagnosis: bleeding of unknown cause (BUC), where no specific haemostatic treatment is available. However, it is a clinical experience, published in many textbooks, that bleeding, and bruising associated with EDS improve by oral intake of vitamin C. No previous studies examined vitamin C levels in patients with BUC. The aim of the study is to investigate the association of vitamin C levels with HM and vascular fragility in patients with BUC.

Methods

Patients with bleeding tendency are referred to our Haemophilia Center. Following ISTH-BAT score, normal coagulation test (INR, APTT, fibrinogen, F8, F9, F11, F13, VWFAg, VWFRist) and inconclusive results of flow cytometry analysis of platelet function, patients are diagnosed with BUC. Patients with a high ISTH-BAT score > 10 were included in this study. We performed clinical examinations by a tourniquet test for capillary fragility and a Beighton score for generalized HM. A skin extensibility test was performed by pulling on the skin off the forearm and the lateral side of the throat until resistance was met. The test was considered positive (abnormal) if the skin extended more than 2 cm in both sites when pulled.

Germline DNA was analyzed by whole genome sequencing (WGS) analysis and an in silico panel of 170 genes associated with bleeding and generalized HM was examined. Additionally, homozygous polymorphism in vitamin C genes (GULOP, SLC23A1, SLC23A2, SVCT1, HP, GST and SOD2) known to cause reduced vitamin C levels, were assessed

Vitamin C analysis: The participants received oral and written guidelines regarding intake of vitamin C in relation to the blood sample. The guidelines included no vitamin C supplements for 3 days before the blood sampling and no intake of vitamin C rich food on the day of the appointment. Plasma samples were immediately acidified by addition of 10% meta-phosphoric acid to ensure the stability of ascorbic acid, and subsequently analyzed by high-performance liquid chromatography. Severe vitamin C deficiency was defined as <11µM, deficiency as < 23 µM, suboptimal status as 23-49 µM and adequate status as >49 µM.

Results

Sixty probands and 20 healthy age- and gender-matched controls (HC) were included. Most patients were women (56/60, 93%), median age 48 (range 21- 75) and median BAT score 13 (range 10-21). By WGS a 67-year woman with Beighton score 7 was diagnosed with Noonan syndrome due to a pathogenic variant in PTPN11 (c.922A>G). Also, a 62-year woman without HM was diagnosed with mild FV deficiency due to a heterozygous variant in F5 (c.449A>G). No other causal variants were identified. In a patient with a homozygous variant in SLC23A1, vitamin C level was suboptimal 42µM. Suboptimal vitamin C levels were found in 19/60 patients and 2/20 HC. One patient was severely vitamin C deficient (7µM). Vitamin C levels were significantly lower in patients compared to HC (median 56 µM vs 76 µM; p = 0.012). Beighton score was positive in 29/60 patients and 1/20 HC (p<0.001). In 18/60 patients versus 0/20 HC the skin extensibility test was abnormal (p = 0.01). In 12/29 patients with a positive Beighton score the test for skin extensibility was also positive indicating a diagnosis of hEDS. Tourniquet test was positive (> 10 petechiae) in 18/60 patients and 4/20 HC (p>0.5). Patients with positive Beighton scores had significantly higher BAT scores than patients without HM (P= 0.014). No significant associations were found between vitamin C levels, a positive Beighton score, a positive Tourniquet test or a high BAT score respectively.

Conclusions

We found that a high proportion of patients with BUC have generalized HM, abnormal skin extensibility and positive tourniquet test. Indeed, patients with HM had increased ISTH-BAT scores compared to patients with negative Beighton scores. Our data support that hEDS may be an underlying cause of BUC. Further studies are needed to determine the genetic background of hEDS. Levels of vitamin C were reduced in patients with BUC compared to age and sex matched control. Our results could pave the way for a randomized study of vitamin C supplements versus placebo for bleeding and bruising in patients with generalized HM and BUC.