Introduction

By decreasing the immune system’s functional capacity, Hu man Immunodeficiency Virus (HIV) has become one of the greatest problems for the healthcare system, contributing significantly to the morbidity and mortality of HIV infected patients¹. It is estimated that around 36.9 million people were living with HIV globally in the year 2017².

In the general population, tobacco use is one of the most im portant factors that reduce life expectancy and one of the most modifiable³. There is high prevalence of tobacco use in HIV pa tients, leading to an increase in the production of inflammatory cells, a greater viral replication and, in general, contribute to al terations capable of affecting the health of these patients^4,5. Di verse problems exist, such as: a worse immunological response, as well as viral response; clinical manifestations induced by HIV; greater consumption of marijuana and alcohol; a decrease in quality of life; and greater risk of mortality⁴.

Antiretroviral therapy, highly active antiretroviral therapy (HA ART) has played a vital role in the reduction of viral replica tion⁶. The Dominican Republic is a country of great interest for future research. According to a progress report published by the World Health Organization (WHO), the island shared by the Dominican Republic and Haiti makes up about 70% of the cases of HIV in the Caribbean⁷. On its own, the Dominican Republic makes up 0.7% of the cases of HIV, covering 80% of the antire troviral treatment costs for their population⁸. This country has been one of the primary tobacco producers in Latin America⁹.

The objective of this investigation was to evaluate the rela tionship between the use of tobacco and the viral load and CD4+ T cell count in HIV patients in one of the “Unidad de Atención Integral (UAI)” sites, which are HIV health care cen ters, found in the Hospital Regional Universitario José María Cabral y Báez (the regional hospital of Santiago, Dom. Rep.). Patients were categorized as smokers, ex-smokers, and non-smokers. The second objective was to create a logistic re gression analysis to determine variables, such as tobacco use, age, sex, duration under antiretroviral treatment, length of time with diagnosis, depression and anxiety, illicit drug use, excessive alcohol consumption, sexually transmitted disease, and adverse reactions to treatment that may influence the effectiveness of the treatment

Methods

A descriptive observational study was implemented. The po pulation studied consisted of HIV patients 18 years old and above, on HAART antiretroviral therapy^3,10, who attended the UAI in the Hospital Regional Universitario José María Cabral y Báez in Santiago, Dominican Republic. The sample was se lected from the total population, obtained from the monthly reports to the “Dirección Provincial de Salud de Santiago” (Provincial Health Directorate of Santiago), that were actively enrolled and had records in this HIV health care center^11,12.

Using a confidence interval of 95%, the sample size was from the reference population using Raosoft^13,14,15. The sample size was 317 patients from a total of 1,789 patients enrolled at the UAI of which 172 patients met all of the following in clusion criteria: 18 years old or older; HIV diagnosis; on HA ART; treatment adherence; be enrolled and have records at the UAI; and completed the informed consent. We screened the adherence to treatment as an inclusion criteria using the following approach from Friedman et al¹¹: individuals were classified as “non-adherent” if they had missed taking at least one pill and “adherent” not having missed taking any pills during the two days prior to survey administration.

The study protocol and data accessed was approved by the Bioethics Committee, “Comité de Bioética de la Facultad de Ciencias de la Salud (COBE-FACS)”, from the Pontifícia Uni versidad Católica Madre y Maestra. We worked with the me dical doctors and personnel at the center, and as patients walked in, they would be approached and informed of the study and invited to participate voluntarily¹⁶. The CD4+ T cell count and viral load was obtained from their records and in corporated into the questionnaire^10,17.

The data was compiled using Microsoft Excel and analyzed using Statistical Package for the Social Sciences (SPSS) program¹⁸. The variables that were taken into account in this study were all con verted to qualitative dichotomous variables. The patients were categorized as smokers, ex-smokers, and never smokers using questions modified by the “Unidad Técnico Asesora de Investi gación” (Investigation Department) of the Pontificia Universidad Católica Madre y Maestra taken from “Encuesta Mundial de Ta baquismo en Adultos” (Global Questionnaire for Tobacco Use in Adults) ^19,20,21. The use of tobacco was measured as the use of any tobacco product at least once a week³.

Sex was either male or female²⁰. The age was reported by the patients and was later classified into two groups: < 45 years old and ≥ 45 years old^1,22. The greater the CD4+ T cell count (>250 cells/mm³) the greater the effectiveness of the treatment; likewise, the lower the viral load (<400 copies/ml), the more effective the antiretroviral treatment^6,23,24. Depression and anxie ty grouped into “Yes” (presence of a relevant clinical problem) and “No” ¹⁴. Illicit drug use into “Yes” and “No” using questions 11, 12 and 13 from the Spanish version of the questionnaire provided by the New York State Department of Health, Bureau of Communicable Disease Control¹³, specifically injected drugs, non-injected and/or inhaled drugs¹³, and a “Yes” to any one of these three questions classified the patient as a drug consumer.

Using the Center for Disease and Control’s (CDC) parameters, excessive alcohol consumption was either “Yes”, drinking 5 or more alcoholic beverages in one sitting for men and 4 or more alcoholic beverages for women in one sitting in the past 30 days²⁵, and “No”. Adverse reaction took into accou nt the most common adverse reaction(s) to the antiretroviral treatment reported in the pilot study, (dizziness, vomiting, night terrors and allergies).

The variables “duration under antiretroviral treatment” and “length of time with diagnosis” were reported as nominal polytomous variables and then converted into categorical dichotomous variables falling into the following: one year or more or less than one year²⁶. History of sexually transmit ted disease was divided into: having two or more sexually transmitted diseases, such as syphilis, chlamydia, gonorrhea, hepatitis A, B and/or C and the other was having only one sexually transmitted disease (i.e. HIV) ¹³.

Results

Approximately 60% of the participants were less than 45 years old; around 58.7% of the patients were female. 45.3% of the patients were categorized as current smokers. Tho se that were not classified as current smokers were further classified as either ex-smokers or never smokers, of which 17% were ex-smokers and 83% were never smokers. 21% of the patients analyzed had a viral load greater than or equal to 400 copies/ml; 13% had a CD4+ T cell count less than or equal to 250cells/mm³.

Table 1 shows that 77.3% of patients with a CD4+ T cell cou nt ≤ 250 cells/mm3 were smokers. In regards to non-current smokers, 59.3% of patients with a CD4+ T cell count ≥ 251 cells/mm³ were of this category (P = 0.001). In Table 2, 75% of patients with a viral load ≥ 400 copies/ml were smokers. 76 out of the 85 non-current smokers had a viral load ≤ 400 copies/ml (P < 0.001). No relationship was found between viral load and whether a patient was an ex-smoker or never smoker (P = 0.703). Figure 1 demonstrates the logistic re gression analysis that was done with the viral load and all the variables included in the study. Controlling for the confoun ding variables, it was observed that the variables that de monstrated a significant association with the viral load were smoking and age. Smokers had 6.285 times the probability of having a viral load ≥ 400 copies/ml (P < 0.001). Having an age of less than 45 years carried a 3.313 times the probability of having a viral load ≥ 400 copies/ml (P = 0.024).

Table 1 Relationship of CD4+ T cells in smokers, ex-smokers and never smokers 

^Source: Data collection instrument of the relationship of smoking with the viral load and CD4+ T cell count in HIV patients, 2015.

Table 2 Relationship of viral load in smokers, ex-smokers, and never smokers 

^Source: Data collection instrument of the relationship of smoking with the viral load and CD4+ T cell count in HIV patients, 2015

Discussion

It is anticipated that for the year 2020, tobacco use will be come the top issue in global health⁵, highlighting the impor tance of research in this area. Smoking showed significant associations with one of the parameters: viral load. Patients who were smokers had 6.285 times the probability of having an increase in viral load compared to non-smokers.

In the study by Valiathan et al¹⁷, controlling for viral load it was observed that there existed a characteristically persistent acti vation of the immune system and an inflammatory response associated with HIV infection that accelerated the decrease in function of the immune system and can then increase the risks of further infections. It has been reported that the use of to bacco is associated with a decrease in the immune response, an increase in the inflammatory response, increase in oxidati ve stress, opportunistic infections, and possibly, an increase in the replication of HIV-1, a possible decrease in antiretroviral medication effectiveness, and a progressive increase in deve loping AIDS, concluding that HIV smokers lose more life years to smoking than to the actual HIV infection itself^5,17.

Being that tobacco use is more prevalent in the HIV popu lation4, this group runs a significant risk. In Table 1, 77.3% of patients with a CD4+ T cell count ≤ 250 cells/mm3 were smokers. In addition, smokers were 75% of those with a viral load ≥ 400 copies/ml. These results support that the para meters used to measure the effectiveness of the treatment, viral load and CD4+ T cell count, were directly affected by smoking (Table 1 and 2). Taking into account these results, we can affirm that HIV positive patients have a decreased immune system response, leading to an inadequate response to their treatment. We observed that 82.9% of non-current smokers who presented with a viral load < 400 copies/ml were never smokers, which further supports that HIV patients who do not smoke, have a lower amount of the virus in their system (Table 2).

The logistic regression analysis observed in Fig. 1, when com pared to the patients 45 years or older, patients younger than 45 years old had a 3.13 times greater probability of having a viral load ≥ 400 copies/ml (P = 0.024). Villante et al²² reported that younger patients were more prone to be smokers (P < 0.05), also supporting the results found in our study. It is im portant to consider age as a confounding factor as there was a higher proportion of participants under the age of 45 years old enrolled in the study, and age may have had an effect between the relationship of smoking and viral load.

Source: Data collection instrument of the relationship of smoking with the viral load and CD4+ T cell count in HIV patients, 2015

Figure 1 Logistic regression of the viral load and the following variables: smoking, sex, illicit drug use, age, length of time with diagnosis, duration under antiretroviral treatment, adverse reaction, sexually transmitted disease, depression and anxiety, and excessive alcohol consumption. 

The literature reports that in most cases the patients who become infected with HIV are mainly from low income resou rces and have low education levels²⁷, therefore lack of aware ness could delay their search for medical attention, and as a result, delay their treatment process, leading to an increase in a poor response to the antiretroviral treatment. In agree ment with the literature²⁸, the logistic regression analysis in our study (Fig. 1) showed that there did not exist a significant association between time of diagnosis and the duration un der antiretroviral treatment. Likewise, the following variables did not show a significant association with the viral load: sex, depression and anxiety, adverse reaction, a history of sexually transmitted disease, excessive consumption of alcohol, nor illicit drugs use.

In order to incorporate variables valuable to our study, it was of great importance to perform an exhaustive research of the literature^{3,6,12,22,29,30,31}. To our knowledge, our study brings new information to Latin American countries, specifically the Dominican Republic. Further strengthening our research, we utilized the two parameters (viral load and CD4+ T cell count) that have been used to determine effectiveness of antiretro viral treatment, however, being that the study was a cross-sectional transverse design, we were not able to determine cause and effect relationships between the variables. In the case of the use of tobacco, we used a questionnaire already validated with which we were able to obtain the necessary information that helped in the design of our study.