During acute-phase responses resulting from infection or trauma, there are wide ranges of pathophysiological responses including the reduction in serum albumin levels. Reductions in serum albumin levels are attributed to increasing levels of inflammatory biomarkers, such as tumor necrosis factor (TNF) and interleukin-6 (IL-6), which are responsible for reductions in hepatic albumin synthesis and increases in albumin leakage to the extravascular space, while enhancing the degradation of albumin1,2.
Several studies involving patients infected with human immunodeficiency virus (HIV) have determined that hypoalbuminemia (defined as serum albumin concentrations <3.5g/dL) is associated with more rapid progression to acquired immune deficiency syndrome (AIDS) and mortality3–5.
The aims of this cross-sectional study were to determine the prevalence and investigate the risk factors associated with hypoalbuminemia in outpatients with HIV/AIDS.
This transversal study was performed on adult outpatients infected with HIV in the Eduardo de Menezes Hospital, Hospital Foundation of Minas Gerais, Belo Horizonte, Minas Gerais, Brazil. The sample size was calculated based on the total number of outpatients infected with HIV (n=1,400) who regularly attended the Infectious Diseases Service of the Hospital Eduardo de Menezes-HEM da Fundação Hospitalar do Estado de Minas Gerais-FHEMIG. The prevalence of hypoalbuminemia in patients infected with HIV ranged from 5.9% to 39.5%6–8. The sample number was calculated based on a hypoalbuminemia prevalence of 50%, with a range of 8% at a 95% confidence level. A minimum sample of 136 was obtained. The study included patients infected with HIV who were older than 18 years of age and agreed to participate in the research through the signing of a free and informed consent agreement. Pregnant women with HIV were excluded. The outpatients with HIV in this study were selected after drawing among those who were booked on the day for an appointment at the HEM. All the selected subjects agreed to participate and had their data collected.
A single researcher performed data collection. Atenrollment, clinical information, and demographic and anthropometric characteristics were obtained. Laboratory test results [cluster of differentiation (CD) 4 cell counts, viral load, albumin, and total protein] were collected from the medical records. Blood samples were collected after undergoing a 12-hour fast.
Counting of CD4 cells was performed by the Laboratory of the Fundação Ezequiel Dias, Belo Horizonte, Minas Gerais; other biochemical tests were performed by the Clinical Analysis Laboratory of the HEM/FHEMIG. CD4+ T-cell counts were obtained using flow cytometry. Viral load was determined using the Versant® HIV 1 RNA 3.0 (bDNA) (Bayer®, Tarrytown, NY, USA) test, with a detection limit threshold of 50 copies/mL of plasma. Plasma levels of total protein and albumin were analyzed enzymatically using a colorimetric assay (Vitros Chemistry Products, Johnson & Johnson Clinical Diagnostics®, Rochester, USA). Hypoalbuminemia was defined by plasma levels of albumin <3.5g/dL7.
All procedures for obtaining anthropometric measurements were carried out in accordance with standardized criteria9. Height was measured using a wall-mounted wooden stadiometer. Weight was measured using a mechanical weighing scale (Filizola®, Filizola S.A. Pesagem e Automação, São Paulo, SP, Brazil), with a maximum capacity of 150kg and precision of 100g. Body mass index (BMI) was calculated by dividing the current body weight (kg) by the square of the height (m2)9. According to BMI, patients were classified as underweight (BMI <18.5kg/m2), normal weight (BMI between 18.5 and 24.99 kg/m2), or overweight (BMI ≥25 kg/m2)9.
Data analysis was carried out with Stata software, version 10.0, at a 5% significance level. Continuous variables with temporal information on infection diagnosis, antiretroviral therapy exposure and age were categorized using the median as the cutoff point. Categorical variables were described through frequencies and compared using the Chi-square test or Fisher’s exact test.
The Shapiro-Wilk test was used to evaluate the normality of continuous variables. Age was also presented as means and standard deviations (SD). To compare age by gender, the Student’s t test was used. A logistic regression model was used to evaluate the association between hypoalbuminemia and other investigated characteristics. The strength of the association was measured through the odds ratio (OR) with 95% confidence intervals (CIs). The variables that presented p-values <0.25 in the univariate logistic regression analysis were selected for final model construction. Goodness-of-fit was verified using the Hosmer-Lemeshow statistics method.
Among the 196 outpatients infected with HIV, the male gender comprised 67.9%. The mean ages of men and women infected with HIV were 41.1 (7.6) years and 42.4 (9.4) years, respectively (p-value=0.31). The proportion of individuals exposed to antiretroviral therapy was higher among men, as shown in Table 1. The overall prevalence of hypoalbuminemia was 11.7%. Table 2 shows the results of ORs for characteristics associated with hypoalbuminemia. The variables selected for final logistic regression model construction for hypoalbuminemia were duration of antiretroviral therapy, current antiretroviral therapy, and BMI. In the final logistic regression model, the only risk factor associated with hypoalbuminemia was current antiretroviral therapy (no exposure: OR=3.46, 95% CI=1.20-10.02).
|n (%)||n (%)||n (%)||p-value|
|Age (years) (n=196)|
|<40||91 (46.4)||66 (49.6)||25 (39.7)|
|≥40||105 (53.6)||67 (50.4)||38 (60.3)||0.19|
|BMI (kg/m2) (n=194)|
|<18.5||16 (8.2)||9 (6.8)||7 (11.3)|
|18.5-24.9||115 (59.3)||84 (63.6)||31 (50.0)|
|≥25||63 (32.5)||39 (29.6)||24 (38.7)||0.18|
|Duration of HIV infection (years) (n=196)|
|<6||103 (52.6)||71 (53.4)||32 (50.8)|
|≥6||93 (47.4)||62 (46.6)||31 (49.2)||0.74|
|Current antiretroviral therapy (n=196)|
|no||22 (11.2)||9 (6.8)||13 (20.6)|
|yes||174 (88.8)||124 (93.2)||48 (79.4)||<0.05|
|Duration of antiretroviral therapy (years) (n=174)|
|<5||85 (48.8)||63 (50.8)||22 (44.0)|
|≥5||89 (51.2)||61 (49.2)||28 (56.0)||0.42|
|CD4 cell count (cells/mm3) (n=196)|
|<200||21 (10.7)||13 (9.7)||8 (12.7)|
|200-499||94 (48.0)||69 (51.9)||25 (39.7)|
|≥500||81 (41.3)||51 (38.4)||30 (47.6)||0.28|
|Albumin (g/dL) (n=196)|
|<3.5||23 (11.7)||14 (10.5)||9 (14.3)|
|≥3.5||173 (88.3)||119 (84.5)||54 (85.7)||0.44|
|N||no||yes||Odds ratio (95% CI)|
|Duration of HIV infection (years)|
|Current antiretroviral therapy|
|Duration of antiretroviral therapy (years) (n=174)|
|Without antiretroviral therapy||22||16||6||2.81 (0.89-8.86)|
|CD4 cell count (cells/mm3)|
In this study, we noted that the prevalence of hypoalbuminemia among outpatients infected with HIV was 11.7%. The factor associated with hypoalbuminemia was current exposure to antiretroviral therapy. The risk for hypoalbuminemia was greater among patients not exposed to antiretroviral therapy.
In previously reported studies, the prevalence of hypoalbuminemia varied from 5.9% to 39.5%6–8,10. This variation may be linked to factors that could influence the serum albumin levels, such as nutritional status, inflammatory response, renal function, hepatic function, and enteropathy11. The prevalence of hypoalbuminemia observed in this study (11.7%) can be explained by the fact that we evaluated albumin levels in a group of patients extensively exposed to antiretroviral therapy; therefore, these patients presented better immunological conditions than non-exposed patients.
The main limitations of this cross-sectional study on identifying risk factors are the temporal bias, because the time sequence cannot be established, and the survival bias, both of which may affect prevalence.
This study shows that the monitoring of plasma albumin is key to determine when antiretroviral therapy should be initiated in individuals who are not exposed to antiretroviral medicines.