Vitamin C Deficiency in Patients With Acute Upper GI Bleeding

In This Article

Abstract and Introduction

Abstract

Background: Vitamin C is an essential dietary nutrient important for collagen synthesis, including within the gastrointestinal tract.

Aim: We aimed to document the prevalence of Vitamin C deficiency (VCD) in patients who present with upper gastrointestinal bleeding (UGIB) and its association with clinical outcomes.

Methods: We conducted a prospective cohort study of patients presenting with UGIB. Fasting Vitamin C levels were collected at admission. Primary outcomes were the prevalence of VCD (Vitamin C level <23 μmol/L, severe VCD < 12 μmol/L) and a composite outcome of adverse events, stratified by VCD status. Secondary outcomes were prolonged hospitalisation and the need for ICU admission.

Results: A total of 227 patients were included (mean age 64.5 years, males 63.9%). VCD was identified in 74 (32.6%) and severe deficiency in 32 (14.1%) patients. VCD was associated with a higher composite endpoint of AE (45.9% vs 24.8%, p < 0.01), higher in-hospital mortality (9.5% vs 1.3%, p < 0.01), increased prolonged admissions (62.2% versus 47.1%, p = 0.03) and increased rebleeding (17.6% vs 7.8%, p = 0.03), compared with patients with normal Vitamin C levels. Multivariate logistic regression models showed that VCD was independently associated with the composite endpoint of AE.

Conclusion: VCD is highly prevalent in patients with UGIB and associated with poorer outcomes, including higher mortality, rebleeding and length of stay. Interventional studies are required to determine the impact of early Vitamin C supplementation on clinical outcomes.

Introduction

Upper gastrointestinal bleeding (UGIB) remains the most common gastrointestinal emergency with a hospitalisation rate of 67 per 100,000 in the United States (US), and 134 per 100,000 in the United Kingdom (UK).^[1,2] Despite advances in management, UGIB continues to be associated with a significant mortality rate of 2%–10% in recent studies.^[3,4]

Vitamin C (ascorbic acid) is a water soluble vitamin and essential dietary micronutrient which plays an important role in connective tissue metabolism. Humans cannot synthesise ascorbic acid and rely on dietary sources for Vitamin C and its oxidation product, dehydroascorbic acid.^[5] Key dietary sources include citrus fruits, green vegetables, tomatoes and potatoes. Vitamin C deficiency (VCD) results in scurvy, a clinical syndrome which may manifest as cutaneous petechiae, ecchymoses, impaired wound healing, 'corkscrew hairs' and mucosal bleeding (particularly oral mucosa).^[6,7] In 1753, James Lind discovered that scurvy in sailors could be prevented by adding citrus fruits to their diet.^[8]

The relationship between VCD and gastritis, peptic ulcer disease and its complications was initially established in the 1930s.^[5] Indeed, VCD was associated with all forms of gastritis (autoimmune, chemical and infectious), and deficiency could be due to insufficient intake, increased metabolic requirements, or destruction within the gastrointestinal (GI) tract.^[9] Ascorbic acid supplementation was possibly associated with a decreased incidence of bleeding from peptic ulcer disease. In the late 1960s, three studies documented an association between VCD and/or low dietary intake of ascorbic acid and UGIB secondary to peptic ulcer disease.^[10–12] To our knowledge, no studies have investigated the prevalence or prognostic significance of VCD in patients with UGIB since that time. Furthermore, there are no studies of VCD in patients with variceal bleeding.

Currently, VCD is thought to be uncommon in developed countries, and largely limited to individuals with alcohol misuse, malnutrition, poverty and critical illness.^[13–15] A significant proportion of individuals presenting with UGIB share these risk factors for VCD, including multiple medical comorbidities, alcohol misuse and malnourishment.

We hypothesised that vitamin C deficiency may be prevalent in patients who present with UGIB, and that deficiency is associated with impaired mucosal healing which affect rebleeding risk and other bleeding complications. Thus, the study objectives were to: (1) determine the prevalence of VCD in patients presenting with acute UGIB and, (2) determine the association between VCD and patient outcomes.

Table 1. Baseline characteristics of patients with upper gastrointestinal bleeding
	Serum vitamin C (μmol/L)			p-value
	Total	≥23	<23
	n = 227	n = 153	n = 74
Age, mean (SD), years	64.5 (16.8)	63.7 (17.2)	66.3 (15.9)	0.28
Male sex, n (%)	145 (63.9)	96 (62.8)	49 (66.2)	0.61
Dementia, (%)	6 (2.6)	3 (2.0)	3 (4.0)	0.36
Diabetes mellitus, n (%)	54 (23.8)	29 (19.0)	25 (33.8)	0.01
Ischaemic heart disease, n (%)	50 (22.0)	36 (23.5)	14 (18.9)	0.62
Congestive heart failure, n (%)	22 (9.7)	11 (7.2)	11 (14.9)	0.07
Stroke or transient ischaemic attack, n (%)	17 (7.5)	13 (8.5)	4 (5.4)	0.41
Peripheral vascular disease, n (%)	7 (3.1)	4 (2.6)	3 (4.1)	0.56
Chronic obstructive pulmonary disease, n (%)	18 (7.9)	10 (6.5)	8 (10.8)	0.26
Chronic liver disease, n (%)	57 (25.1)	34 (22.2)	23 (31.1)	0.15
MELD, median (IQR)	15 (12, 19)	15 (12, 19)	16 (13, 21)	0.70
Child-Pugh, median (IQR)	8 (7, 9)	8 (7, 8)	8 (7, 9)	0.45
Cancer, n (%)	32 (14.1)	19 (12.4)	13 (17.6)	0.30
ACCI, mean (SD)	3.7 (2.3)	3.4 (2.2)	4.3 (2.4)	<0.01

Abbreviation: ACCI, age-Adjusted Charlson Comorbidity index.

Table 2. Clinical characteristics of patients with upper gastrointestinal bleeding
Clinical parameters	Serum vitamin C (μmol/L)			p-value
	Total	≥23	<23
	n = 227	n = 153	n = 74
Haematemesis, n (%)	74 (32.6)	44 (28.6)	30 (41.1)	0.05
Coffee ground vomiting, n (%)	74 (32.6)	61 (32.5)	13 (33.3)	0.35
Melaena, n (%)	74 (32.6)	26 (44.8)	48 (28.4)	0.03
Syncope, n (%)	29 (12.8)	16 (10.5)	13 (17.6)	0.13
Haemoglobin, mean (SD), g/dl	9.3 (2.8)	9.5 (2.9)	8.8 (2.7)	0.05
Albumin, mean (SD), g/L	31.5 (6.9)	32.5 (6.5)	29.5 (7.3)	<0.01
Platelets, mean (SD), ×10⁹/L	209 (97)	212 (94)	202 (103)	0.48
AIMS65 score ≥2, n (%)	83 (36.6)	47 (30.7)	36 (48.7)	<0.01
Glasgow-Blatchford score, mean (SD)	10.0 (4.1)	9.6 (4.1)	10.9 (4.0)	0.02
Endoscopic findings
Peptic ulcer disease, n (%)	100 (44.1)	68 (44.4)	32 (43.2)	0.43
Gastric	50 (22.0)	34 (22.2)	16 (21.6)	0.92
Duodenal	55 (24.2)	39 (25.5)	16 (21.6)	0.52
Variceal bleed, n (%)	44 (19.4)	24 (15.7)	20 (27.0)	0.04
Gastric	6 (2.6)	5 (3.3)	1 (1.4)	0.40
Oesophageal	42 (18.5)	22 (14.4)	20 (27.0)	0.02
Portal hypertensive gastropathy	17 (7.5)	12 (7.8)	5 (6.8)	0.77
Malignancy, n (%)	7 (3.1)	5 (3.3)	2 (2.7)	0.82
Mallory-Weiss tear, n (%)	6 (2.6)	4 (2.6)	2 (2.7)	0.97
Angiodysplasia/Dieulafoy, n (%)	11 (4.9)	9 (5.9)	2 (2.7)	0.50
Oesophagitis, n (%)	21 (9.3)	16 (10.5)	5 (6.8)	0.37
Gastritis/duodenitis, n (%)	17 (7.5)	13 (8.5)	4 (5.4)	0.41
Hiatus hernia/Cameron's erosions, n (%)	6 (2.6)	5 (3.3)	1 (1.4)	0.40
Gastric antral vascular ectasia, n (%)	3 (1.3)	2 (1.3)	1 (1.4)	0.98
No cause found, n (%)	24 (10.6)	15 (9.8)	9 (12.2)	0.59

Table 3. Clinical outcomes of patients with upper gastrointestinal bleeding
Clinical outcomes	Serum vitamin C (μmol/L)			p-value
	Total	≥23	<23
	n = 227	n = 153	n = 74
Primary outcomes
Composite endpoint of AE	72 (31.7)	38 (24.8)	34 (45.9)	<0.01
Individual components of composite endpoint
Inpatient death, n (%)	9 (4.0)	2 (1.3)	7 (9.5)	<0.01
Rebleeding, n (%)	25 (11.0)	12 (7.8)	13 (17.6)	0.03
Angioembolisation, n (%)	5 (2.2)	3 (2.0)	2 (2.7)	0.72
Surgical intervention, n (%)	2 (0.9)	2 (1.3)	0 (0)	0.32
Transfusion ≥4 units red cells, n (%)	55 (24.2)	32 (20.9)	23 (31.1)	0.09
Surgical intervention, n (%)	2 (0.9)	2 (1.3)	0 (0)	0.32
Transfusion ≥4 units red cells, n (%)	55 (24.2)	32 (20.9)	23 (31.1)	0.09
Secondary outcomes
Length of stay >7 days, n (%)	118 (52.0)	72 (47.1)	46 (62.2)	0.03
ICU admission, n (%)	27 (11.9)	16 (10.5)	11 (14.9)	0.34

Table 4. Univariable logistic regression for the composite endpoint of AE
Univariable regression	Odds ratio	95% CI	p value
Vitamin C deficiency (main factor)	2.57	1.43–4.62	<0.01
Diabetes	1.52	0.80–2.88	0.20
Chronic liver disease	1.85	0.99–3.45	0.05
Congestive heart failure	0.79	0.30–2.11	0.64
ACCI > 3	1.32	0.76–2.32	0.33
AIMS65 score ≥2	3.26	1.82–5.83	<0.01
Glasgow-Blatchford Score ≥7	8.83	2.64–29.56	<0.01
Variceal bleeding	1.65	0.84–3.26	0.15
Serum albumin	0.91	0.87–0.95	<0.01
Haemoglobin	0.64	0.55–0.74	<0.01

Abbreviation: ACCI, age-adjusted Charlson Comorbidity index.

Table 5. Multivariable logistic regression with vitamin C deficiency as factor of interest
Model	Covariates	VCD odds ratio	95% CI
	Vitamin C deficiency (univariable)	2.57	1.43–4.62
Model 1	Albumin, haemoglobin	2.18	1.12–4.25
Model 2	AIMS65 ≥ 2, haemoglobin	2.20	1.18–4.28
Model 3	AIMS65 ≥ 2, ACCI > 3	2.26	1.22–4.17
Model 4	AIMS65 ≥ 2, ACCI > 3, Haemoglobin	2.24	1.14–4.38
Model 5	AIMS65 ≥ 2, variceal bleeding	2.17	1.18–4.00
Model 6	AIMS65 ≥ 2, chronic liver disease	2.18	1.18–4.01
Model 7	Glasgow-Blatchford Score ≥7, albumin	2.30	1.22–4.31
Model 8	Albumin, INR	2.20	1.18–4.11
Model 9	Albumin, INR, haemoglobin	2.35	1.19–4.64

Abbreviation: ACCI, age-adjusted Charlson Comorbidity index.

My Alerts

Add to Email Alerts

Your Name is required.

Subject is required.

Please enter a Recipient Address and/or check the Send me a copy checkbox.

Your email has been sent.

is an Invalid Email Address.

Prevalence and Prognostic Significance of Vitamin C Deficiency in Patients With Acute Upper Gastrointestinal Bleeding

A Prospective Cohort Study

Abstract and Introduction

Abstract

Introduction

Prevalence and Prognostic Significance of Vitamin C Deficiency in Patients With Acute Upper Gastrointestinal Bleeding

A Prospective Cohort Study

Abstract and Introduction

Abstract

Introduction

Tables

Tables

Tables

Tables

Tables

References

Authors and Disclosures

Authors and Disclosures

Email This

Feedback