Increased homocysteine plasma level is associated with shortened prothrombin time in HIV-infected patients

Abstract

Objective: To find factors associated with increased homocysteine plasma level in HIV-infected patients.

Methods: Cross-sectional study, carried out as a supplementary task to the standard care of HIV-infected patients. The possible association of increased homocysteine plasma level with blood analyses results was assessed with a multiple linear regression analysis, using the automatic linear modeling available in SPSS version 22.

Results: A total of 145 patients were included. Creatinine was higher than normal in 7 patients (5%), prothrombin time was shortened in 36 patients (25%), and a monoclonal gammopathy was detected in 2 patients (1%). In the regression analysis, an association was found between high homocysteine plasma level and the following variables: low prothrombin time (β coefficient −0.286, confidence interval −1.1854 to −0.754, p < 0.001), high creatinine (coefficient 9.926, confidence interval 6.351–15.246, p < 0.001), low folic acid (coefficient −0.331, confidence interval −0–483 to −0.187, p < 0.001), and low vitamin B₁₂ (coefficient −0.007, confidence interval −0.01 to −0.001, p = 0.005).

Conclusion: An association was found between increased homocysteine plasma level and shortened prothrombin time.

Keywords:

• Prothrombin time
• homocysteine
• HIV-infection

Introduction

Combined antiretroviral therapy has resulted in a substantial decrease in morbidity and mortality related to human immunodeficiency virus (HIV) infection, particularly in developed countries.^1–4 However, many HIV-infected patients have dyslipidemia, hyperglycemia, and central obesity which may be favored by side effects of antiretroviral medication, effect of ongoing replication of the virus, or both.⁵ In addition, a high prevalence of smoking and other cardiovascular risk factors has been found in these patients.^6,7 As a consequence, cardiovascular disease is being increasingly recognized in HIV-infected subjects, and the magnitude of the problem is expected to significantly increase in years to come.⁸ For that reason, the HIV-infected people constitutes an appropriate population to study the influence of risk factors in the development of cardiovascular disease, and therefore to refine the knowledge in the field.

Homocysteine is an amino acid produced by the catabolism of methionine, an essential amino acid found in high concentrations in eggs, fish, meat, and other foods. Genetic and nutritional factors determine homocysteine plasma levels. They are increased by an excessive methionine intake and by dietary deficiencies of folate, vitamin B₆, vitamin B₁₂, and other vitamins.⁹ Increased homocysteine plasma levels was considered for years as a risk factor for cardiovascular disease,^10,11 and epidemiological studies suggested a beneficial role of homocysteine-lowering strategies.¹² However, large randomized clinical trials failed to demonstrate a clear effect of such strategies on clinical outcome in patients with atherosclerosis,^13–16 and as a result, the role of the amino acid as a cardiovascular risk factor has been seriously questioned.^17,18 Moreover, it is unclear by now whether hyperhomocysteinemia is a cause, a consequence, or just a marker of cardiovascular disease. Therefore, a better understanding of the authentic role of homocysteine in the pathogenesis of cardiovascular disease, as well as the determinants of increased plasma levels of the amino acid, is needed.

In a previous study, we not only confirmed the well-known association of high homocysteine plasma level with low serum level of folic acid and vitamin B₁₂,⁹ but we also found an association of high homocysteine plasma level with a family history of early coronary disease, sexual HIV risk behavior, hepatitis C virus co-infection, higher height, and higher diastolic blood pressure, in a group of HIV-infected patients.¹⁹ With this new study, in the same line of research, we aimed to determine more factors associated with high homocysteine plasma level in the HIV infected population. For that purpose, we assessed the possible association of multiple blood and urine analyses results with homocysteine plasma level.

Method

Study design and patients

This was a cross-sectional study, carried out in the HIV outpatients’ clinic of the Hospital General Universitario of Castellon, a public institution associated with the University Jaume I of Castellon, belonging to the National Health Service of Spain and the Generalitat Valenciana. The organization provides free of charge comprehensive medical care to most HIV-infected people in the province of Castellon.

We aimed to enroll all patients older than 18 years of age belonging to one of the practices in the clinic. But we excluded pregnant women and subjects who were taking vitamin supplements because those circumstances have been demonstrated to influence homocysteine plasma levels.^9,20,21

Procedures and variables included

This study was designed as a supplementary task to the usual follow-up required by HIV-infected patients. According to guidelines,^3,4 these subjects must be monitored every three to six months, in terms of clinical, immunologic, and virologic status, and they must receive antiretroviral treatment and prophylaxis against Pneumocystis jiroveci pneumonia, generally with trimethoprim/sulfamethoxazole, and other opportunistic infections when indicated. The study was completed throughout two consecutive visits of every patient. All participants were attended at both visits by one of the authors of this paper.

In the first visit, demographic, clinical and HIV-related characteristics from each participant were recorded. In the second visit, results of blood and urine analyses that had been taken after overnight fasting, one to three weeks earlier, were recorded. Plasma homocysteine was measured with a chemiluminescence immunoassay (Immulite 2500, normal value: 5 to 12 μmol/L). All the laboratory measurements were conducted by investigators who were blinded to the characteristics of the patients.

In the present study, we assessed the possible association of homocysteine plasma level with blood and urine analyses results.

The project of the study was approved by the Clinical Research Committee of the Department of Health of Castellon, Spain. The study was carried out according to the good clinical practice principles set out in the Declaration of Helsinki in 1964 and subsequent updates. Participants’ identification details were substituted by codes, which were used throughout all phases of the study. On enrollment, patients gave informed consent to participate.

Statistics

Data collection and statistical analysis were performed with IBM SPSS Statistics, version 22 for Windows (SPSS, Inc. Chicago, Illinois, USA).

Continuous variables were summarized as mean and standard deviation, if normally distributed, or as median and interquartile range if not normally distributed. Discrete variables were summarized as percentages. Variables not showing a normal distribution were normalized by means of a log transformation when appropriate. In bivariate analyses, the Student t test, the Mann–Whitney U test or the χ² test were used as needed.

A multivariable analysis was carried out to try to find an association between homocysteine plasma level, as dependent variable, and the rest of the study variables, as independent variables. For that purpose, the automatic linear modeling, available in software IBM SPSS Statistics version 22, was used. The modeling default specifications were employed, including stepwise forward method, information criterion for entry/removal, inclusion with P-values less than 0.05, and removal with P-values greater than 0.1.

Results

Descriptive data and bivariable analyses

We excluded from the study two patients, a 33-year-old woman, because she was pregnant, and a 48-year-old man because he was taking vitamin supplements.

We included a total of 145 patients; 103 (71%) were male; mean and standard deviation of age was 41.1 and 8.1; 139 (96%) were natives of Spain, 3 (2%) were from other countries of Europe, 2 (1%) from Nigeria, and 1 (1%) from Brazil. All participants were white, except for the two Nigerians, who were black. No patient had any missing data.

Of the 145 participating patients, 117 (81%) were receiving antiretroviral medication. A total of three patients (2%) had presented a major cardiovascular event previously. Two of them had suffered a myocardial infarction; ages of both patients were 70 and 71 years, and they were the two oldest patients in the study. The other patient with the antecedent of cardiovascular disease had presented an extensive ischemic stroke; his age was 37. All 23 patients with the antecedent of dyslipidemia were on lipid-reducing medication. All 24 patients with the antecedent of diabetes mellitus were on treatment with insulin and/or oral antidiabetic drugs. And 15 of the 17 patients with the antecedent of hypertension were on blood pressure-lowering medication; the other two patients with hypertension were on treatment with low salt diet.

The overall mean and standard deviation homocysteine plasma level was 11.9 and 5.5 μmol/L (median: 10.6 μmol/L, interquartile range: 8.4–13.5 μmol/L, and minimum to maximum range: 4.1–39.9 μmol/L). A total of 54 patients (37%) presented homocysteine plasma levels higher than the considered upper limit of normal. A total of 31 patients (21%) were taking trimethoprim/sulfamethoxazole; the mean homocysteine serum level was 12.1 μmol/L in those patients, as compared to 11.8 in those who were not taking the medication (p = 0.793).

CD4 cell count was higher than 200 cells per mm³ in 114 participants (79%), and HIV RNA was lower than 1000 copies/ml (3.0 log₁₀) in 115 participants (79%). Prothrombin time was shortened in 36 patients (25%). Creatinine was higher than normal in 7 patients (5%). Two patients presented monoclonal gammapathy that had been previously diagnosed as monoclonal gammapathy of undetermined significance, and in the two cases the M component was of the IgG kappa class. Homocysteine plasma levels in these two patients were, respectively, 4.12 and 4.37 μmol/L, which were the first and third lowest results in the whole study.

In 42 patients (29%) a dipstick urine analysis disclosed at least one of the following abnormal results: high white blood cell count, or positive nitrate, protein, glucose, ketone bodies, urobilinogen, or red blood cells. Table 1 summarizes the rest of the study analyses results, as well as correlation of each one of the study variables with homocysteine plasma level.

Table 1 Blood analysis results of the study patients, quantitative data, and bivariable correlations between study variables

Test	Normal values	Mean or median*	Standard Deviation or interquartile range*	Pearson or Spearman* correlation with homocysteine	P of correlation with homocysteine
Homocysteine, μmol/L	5–12	11.6	*8.4 − 13.5	N. A.	N. A.
Glucose, mg/dL	70–111	105	32	−0.008	0.920
Blood urea nitrogen (BUN), mg/dL	4.0–24.0	13.6	4.0	0.170	0.041
Creatinine, mg/dL	0.50–1.20	0.85	*0.76 − 0.98	0.381	< 0.001
Uric acid, mg/dL	2.40–7.00	4.99	1.52	0.258	0.002
Total bilirubin, mg/dL	0.50–1.00	0.76	0.74	−0.063	0.451
Albumin, g/dL	3.50–5.50	3.83	0.52	0.152	0.069
Gamma–globulins, g/dL	0.70–1.70	1.50	0.60	−0.118	0.159
IgG, mg/dL	800–1700	1871	804	−0.090	0.279
Calcium, mg/dL	8.50–10.50	9.05	0.39	0.072	0.387
Phosphorus, mg/dL	2.50–4.50	3.05	0.51	0.048	0.565
Total cholesterol, mg/dL	140–240	174	47	0.109	0.154
Triglycerides, mg/dL	50–165	154	112	0.063	0.452
HDL–cholesterol, mg/dL	45.0–65.0	41.2	13.1	−0.003	0.969
Folic acid, μg/L	4.00–16.00	4.4	*2.9 − 6.8	−0.477	<0.001
Vitamin B12, ng/L	280–990	328	*258 − 459	−0.223	0.007
Glycosylated hemoglobin, %	4.00–6.00	5.66	0.89	−0.006	0.204
White blood cells, x 10^3/mm^3	4.80–10.80	6.30	2.14	−0.028	0.742
Hemoglobin, g/dL	12.0–18.0	13.8	1.9	0.235	0.004
Hematocrit, %	37.0–52.0	40.8	5.1	0.210	0.011
Mean corpuscular volume, fL	80–94	103	14	0.050	0.552
Platelet count, × 10^3/mm^3	130–400	212	82	0.064	0.446
Prothrombin time, seconds	13.0–17.0	13.9	2.5	−0.225	0.007
Partial thromboplastin time, activated, seconds	20.0–40.0	28.8	4.65	−0.039	0.643
Fibrinogen, mg/dL	200–400	369	113	−0.140	0.092
Erythrocyte sedimentation rate, mm per hour	0.0–10.0	18	*6 − 36	−0.134	0.109
T lym7phocyte count, per mm^3	900–5200	1890	*1329 − 2446	−0.064	0.447
CD4 T lymphocyte count, per mm^3	410–1590	510	380	−0.083	0.319

* Note: Medias, interquartile range and Spearman test in variables with skewness >1; N. A., not applicable.

Multivariable analysis

An automatic linear modeling was performed with homocysteine plasma level as the dependent variable and the rest of the study variables as independent variables. The analysis gave as a result a model significantly different from zero (F = 10,577, p < 0.001), where one-third of the variability of homocysteine plasma level was explained by the rest of the study variables (adjusted R² 0.330).

Five variables were included in the model: prothrombin time and folic acid that showed a mild negative association with homocysteine plasma level, creatinine serum level that showed a mild positive association with homocysteine plasma level, vitamin B₁₂ serum level that showed a very mild negative association with homocysteine plasma level, and erythrocyte sedimentation rate that showed no reliable association with homocysteine plasma level. Table 2 shows the numeric details of those results.

Table 2 Information regarding coefficients of independent variables included in the multiple regression model

Variable			95% confidence interval
	β coefficient	P	Lower	Upper
Prothrombin time*	−1.286	<0.001	−1.854	−0.754
Folic acid*	−0.331	<0.001	−0.483	−0.187
Creatinine*	9.926	<0.001	6.351	15.246
Vitamin B12*	−0.007	0.005	−0.018	−0.001
Erythrocyte sedimentation rate*	−1.601	0.096	−3.985	0.453

* Note: Variables logarithmically transformed by automatic linear modeling.

Discussion

Homocysteine plasma level was higher in our patients than in most other populations of HIV-infected people described in the literature.^22–26 Although results in this field are varied, and a recent study has found a very high prevalence of hyperhomocysteinemia among these patients.²⁷

The most interesting finding in our study was the negative association of homocysteine plasma level with prothrombin time, i.e. patients with increased homocysteine plasma level had a shortened prothrombin time. The results agree with the increased risk of thrombosis found in the HIV-infected population,^28,29 in patients with hyperhomocysteinemia^30–32 or in subjects with both conditions.^27,33 The mechanism of hypercoagulability in HIV infection and also in hyperhomocysteinemia is poorly understood, and presumably several factors are involved in both cases. Our study shows that activation of the extrinsic pathway of the coagulation system is common in patients with both conditions and presumably the abnormality plays a significant role in the prothrombotic state associated with HIV infection and hyperhomocysteinemia. No association was found of homocysteine plasma level with other coagulation tests, which suggests that homocysteine influences only the extrinsic pathway of the coagulation cascade.

We also documented an association between high homocysteine plasma level and high serum creatinine level. Several other studies had found such relationship previously.^34,35 That association may be explained, at least in part, by the diminished glomerular filtration rate of homocysteine and the important role of the kidney in the removal of the amino acid.³⁶ According to the opinion of some authors, renal dysfunction, a well recognized cardiovascular risk factor,³⁷ could be the authentic responsible of the increased cardiovascular morbidity and mortality associated with increased homocysteine plasma level.³⁸ This would explain the demonstrated lack of beneficial effect of homocysteine-lowering strategies in the prevention of cardiovascular disease.^13–16

We also found in our study the well-known associations of high homocysteine plasma level with both low folic acid and low vitamin B₁₂ serum level. In the case of vitamin B₁₂, the association we found was mild, in contrast with the stronger associations found in other studies.³¹ In contrast with the results of other studies, we did not find an association of homocysteine plasma level with serum HDL–cholesterol,³⁹ glycosylated hemoglobin⁴⁰ or other analyses results.

One curious finding in this study was the negative association of homocysteine plasma level with the presence of monoclonal gammopathy, i.e. patients with monoclonal gammopathy had lower homocysteine plasma level than those without the abnormal protein. This finding must be interpreted with great care, as only two patients in our study presented a monoclonal gammopathy. In a review of the literature, by means of a Medline search using “(monoclonal gammapathy OR monoclonal gammopathy) AND homocysteine” as search strategy keys, we found no studies assessing such possible association. Nevertheless, in contrast to our findings, a report exists of a patient with Schnitzler’s syndrome, a rare condition characterized by chronic urticaria, intermittent fever, bone pain, and monoclonal IgM gammopathy, who presented elevated homocysteine plasma level.⁴¹

The multivariable model in this study only explained one third of the variability of homocysteine plasma level. This means that knowledge of the determinants of the amino acid plasma level is still limited, and therefore more studies are needed to clarify the authentic role of homocysteine in the pathogenesis of cardiovascular disease.

The main limitations of the present study are the cross-sectional design and the relatively small population analyzed. Failure to include other variables that may influence homocysteine plasma level¹⁷ is another potential problem. Despite those inconveniencies, our study provides some new relevant information regarding the determinants of homocysteine plasma level in the HIV-infected population.

Disclaimer statements

Contributors Bernardino Roca, MD, PhD, is an associate professor of Medicine and research interests include cardiovascular risk factors in HIV-infected patients. Manuel Roca, MD, is a collaborative professor of Medicine and research interests include eye infections. Guillermo Girones, MD, is a collaborative professor of Medicine and research interests include tropical infections.

Conflicts of interest The authors of this work report no conflicts of interest.