It is aging and not antiretroviral therapy the strongest risk factor for chronic pain in HIV-positive population ?

Background. Chronic pain in HIV-positive patients is a serious health problem that limits patients’ normal functioning both somatically and psychologically. The current state of knowledge on the topic is insufficient, with the underlying causes of this pain unexplained. We have attempted to answer the question if aging is more stronger risk factor for chronic pain in HIV-infected patients, than antiretroviral therapy (ART). Methods. This study was prospective, observational, cross-sectional study, including consecutive HIV-infected patients under specialist care. During their routine visit all patients reporting any pain were asked to fill in the Brief Pain Inventory (BPI) form and were subject to a brief examination performed by a physician who afterwards completed a Douleur Neuropathique en 4 Questions form (DN4). Logistic regression models were used to identify factors associated with chronic pain occurrence. Results. A total of 196 HIV-positive subjects, 96 (48.9% of the study group) of them reporting pain within the week prior to enrollment. The reported pain was mostly (75%) limited to a single area of the body (most commonly to the lower limbs). Pain duration was reported to be >6 months previous to study enrollment by 57 subjects (59.4% of those reporting pain). The patients with and without pain differed significantly in terms of age at study inclusion (with the median age of 45.3 years in the pain group vs. 39.6 years in the no pain group; p=0.0002); median duration of specialist care (10.8 years vs. 4.9 years, respectively; p=0.0008), median nadir CD4+ cell counts (168 cells/mcL vs. 253 cells/mcL), median duration of ART (8.5 years vs. 3.4 years; p=0.0046), viral rebound after complete suppression (5.1% vs. 38.3%; p=0.018), as well as previous treatment with zidovudine (44.6% vs. 30.5%; p=0.063) and ‘D’ drugs (33.9% vs. 11%; p=0.0004). Conclusions. The prevalence of chronic pain in the studied population of HIV-positive Polish patients

the study group was age, which poses an important clinical and epidemiological problem due to the aging of the HIV-positive population.

Background
As a result of introducing combination antiretroviral therapy (cART) for HIV-1-positive patients their survival has been improved to the level of that in the general population [1,2]. One of the effects of this phenomenon is the increased prevalence and early onset of non-infectious co-morbitidies (particularly cardiovascular disease, chronic kidney disease, type 2 diabetes mellitus, and cancer) in HIV-positive individuals [3][4][5][6]. At the same time, we are observing accentuated aging in HIV-positive people, which is associated mainly with increased immune activation, impaired regulatory functions, as well as direct effects of HIV replication in tissues and organs. One important question that has not been answered yet is whether or not osteoarthritis is more common and/or occurs at a younger age in this population in comparison with the general population. If so, this may have a significant impact on the overall benefit of cART [7,8,9]. Accentuated aging and early onset of selected non-infectious, non-communicable diseases in HIV-1-positive adults result in an increasing prevalence of pain (of various nature) in this population. It is chronic pain that constitutes a particularly big challenge for the primary healthcare team, as it is believed to detrimentally affect the patients' personality, their mental balance, and their ability to perform their social and professional roles [10,11].
Chronic pain is a common occurrence in HIV-infected patients and has an impact on quality of life and antiretroviral adherence. The current state of scientific research in this field is insufficient, and the causes of pain phenomena remain largely unexplained.
However, there is scientific evidence that chronic pain is associated with an increased incidence and level of depression, which may be an additional reason hindering the treatment of HIV-infected people, especially in the aspect of therapeutic adherence.
Emotions, combined with a pain that commands the patient's attention, make the experience even harder to bear. Not uncommonly, the situation is made worser by problems at work due to an inability to perform the required tasks, which leads either to financial problems or experiencing the lack of professional fulfillment [12,13]. Therefore, the purpose of this study was to determine the prevalence of chronic pain in the HIVpositive population remaining under continuous specialist care, as well as the underlying causative factors of such pain.

Patients And Methods
We performed prospective, observational, cross-sectional study, including consecutive The inclusion criteria for this study were the age of 18 years or older and a documented HIV-1 infection, as well as a written informed consent. Any mental condition, diagnosed based on the available clinical data (history-taking or the available medical records), was an exclusion criterion. Chronic pain was defined as pain lasting a minimum of 6 months.
The data on the subject's sex, age, route of HIV infection, age at the time of registration at a specialist clinic for HIV/AIDS-positive patients, laboratory test results (CD4+ cell count, HIV viral load), duration of specialist care (in years), and history of antiretroviral therapy (ART) (yes or no) were obtained from the clinic's electronic database (table 1).
All patients reporting any pain were additionally asked to fill in the Brief Pain Inventory (BPI) form and were subject to a brief examination performed by a physician who afterwards completed a DN4 Douleur Neuropathique en 4 Questions (DN4) form.
The general information form contained questions on the history of any pain over the previous week, mean duration, date of onset, frequency, and any help by physicians other than an infectious disease specialist (particularly those specializing in pain therapy) in dealing with the pain. Subsequent questions addressed the use of any psychoactive drugs or other drugs that could affect the perception of pain, as well as cART adherence (including the frequency of missing a dose).
The subjects who reported pain in the initial questionnaires went on to complete the BPI -Short Form [14]. Some of the questions in this 9-item questionnaire are about the intensity of pain felt at the moment and within the previous 24 hours, quantified in an 11-point numerical rating scale, ranging from 0 to 10, where 0 indicates no pain and 10 -the worst pain imaginable. In addition, all respondents were asked to mark the approximate location of their pain on a diagram of the human body. The questionnaire also asks about treatments or medications used for the pain and asks the respondent to rate the effectiveness of these treatments or medications on a scale from 0% to 100% (in 10% increments), where 0% indicates no relief and 100% -complete pain relief. The questionnaire also assesses the extent to which the pain interferes with selected aspects of everyday life (normal work, general activity, mood, walking ability, relations with other people, sleep, and enjoyment of life) with the answer choices ranging from zero (no interference) to 10 (complete interference).

Statistical analyses
In order to compare the study groups the Chi-squared and Kruskal-Wallis tests were used in statistical analysis. The potential prognostic factors for the study endpoint were identified via uni-and multivariate logistic regression analyses. All statistical analyses were conducted with the use of SAS version 9.3 (SAS Institute, Cary, NC).

Results
A total of 196 subjects aged from 34. 6  Experiencing pain in the week prior to study inclusion was reported by 96 subjects (48.9% of the study group). The mean duration of pain was characterized as "several seconds" by 3 subjects (3.1%), "several minutes" by 28 subjects (28.9%), "several hours" by 51 subjects (52.6%), whereas "continuous pain" was reported by 13 subjects (13.4%). Out of the subjects reporting pain, 57 (59.4%) identified the onset of symptoms as over 6 months before study enrollment.
Generalized pain was reported by 3 subjects (4.3%). Forty-three subjects (75.4%) reported pain limited to a single region of the body, 12 subjects (21%) reported pain limited to two regions, and 2 subjects (3.5%) reported pain limited to 4 regions.

Pain intensity score in a Numerical Rating Scale
The subjects who declared chronic pain (n=57) rated pain intensity in a numerical rating scale (NRS). According to the established standard for this scale, the score of 1-4 points indicated mild pain, 5-6 points indicated moderate pain, and 7-10 points indicated severe pain. Assessed over the period of the previous 24 hours, the intensity of pain was rated as mild, moderate, or severe by 19 (33.3%), 24 (42.1%), and 14 (24.6%) subjects, respectively, when assessing their pain at its worst; by 47 (82.5%), 7 (12.3%), and 3 (4.2%) subjects, respectively, when assessing their pain at its least; and by 44 (77.2%), 7 (12.3%), and 6 (10.5%) subjects, respectively, when assessing their pain on the average. At the time of completing the questionnaire, 48 subjects (85.7%) rated their pain as mild, 4 subjects (7.1%) -as moderate, and 4 subjects (7.1%) -as severe.

Comparison of patient characteristics between the chronic pain group (n=57) and no chronic pain group (n=139).
Both study groups were statistically comparable with respect to sex (22.8% of women in the chronic pain group vs. 15 The median lymphocyte counts at the time when the subjects started to receive specialist care and at the study inclusion were also comparable between the two groups. In the group of those reporting pain, the median CD4+ cell count at the time when they started receiving specialist care was 348 cells/mcL (vs. 350 cells/mcL in the no-pain group; p=0.761), with 25% of the pain-reporting subjects having CD4+ counts lower than 189 cells/mcL (vs. <196 cells/mcL in the no-pain group). The most recent (prior to study inclusion) median CD4+ cell counts were higher at 516 cells/mcL and 563 cells/mcL, respectively (p=0.256).
At study inclusion, 184 subjects were receiving ART (55 in the pain group and 127 in the no pain group). The two groups were comparable in terms of ART rates (98.2% vs. 92.1%, respectively; p=0.185) and initial viral suppression (98.2% vs. 99.2%; p=0.517). Poor cART adherence (based on an initial questionnaire) was reported by 38.2% of subjects with chronic pain and by 30% of those without chronic pain (p=0.306).
The two groups also differed significantly in terms of the route of HIV infection. The chronic pain group was characterized by higher rates of individuals who contracted HIV through intravenous drug use (28.1% vs. 15.1%), heterosexual contact (21.0% vs. 14.8%), and in other ways (8.8% vs. 2.2%); whereas the no chronic pain group was characterized by higher rates of individuals who contracted HIV through homosexual contact (36.7% vs. 31.6%). The route of infection was characterized as unknown by comparable proportions of subjects from both groups . Summing up, the univariate analysis variables that showed statistical significance (p<0.01) and were evaluated in multivariate analysis were: the route of HIV infection, , age at study inclusion, nadir CD4+ cell count, ART duration (years), detectable HIV RNA levels, and previous ART with AZT or 'D' drugs.

Multivariate logistic regression analysis
Ultimately, 184 subjects were analyzed with the multivariate regression model. Those subjects who were not receiving any ART at the time of inclusion into the study were excluded from this analysis due to the fact that various aspects of ART were to be analyzed in this model. A multivariate regression analysis of all the parameters that showed statistical significance in univariate regression models yielded only one parameter that was still statistically significant. It was age at study inclusion (Table 3) In comparison with the studies quoted above, our study demonstrated comparable rates of reported pain, with somewhat lower rates of chronic pain. We would like to emphasize that comparing different studies in terms of the rates of pain is very difficult, due to the considerable variations in the study populations. For instance, a population of Thai patients practically cannot be compared with any European patient population, due to their substantial differences, both genetic and cultural. Even comparisons between European and American studies are difficult, as American studies are frequently conducted in very specific populations, e.g. in a population of social outcasts. Another obstacle in comparing the results of different pain-related studies is the lack of a universally accepted definition of chronic pain [21][22][23][24][25]. For example, in an American study by Miaskowski et al., as well as in our study, chronic pain was defined as pain lasting over 6 months. However, the proportion of the study population reporting pain in that American study (in contrast to that in our study) was very high at 90%. In light of the fact that the population analyzed in the American study had been recruited from the REACH cohort (constituting exclusively the homeless), any reliable comparison with our study is impossible.
The fact that the prevalence of chronic pain in our study was lower than that in the study by Lawson et al. [18] is most likely due to their defining chronic pain as pain lasting over 3 months.
Comparing the results of our study with the studies on chronic pain conducted in HIVnegative adults, chronic pain was considerably more common in the HIV-positive population. The 2006 study by Breivik et al. has the most similar study design to that of our study, as those authors had adopted the same definition of chronic pain as that in our study, and the study populations were comparable. Breivik's study involved several tens of thousands subjects from about a dozen European countries and Israel and showed that 19% of them suffered chronic pain (defined as pain lasting ≥6 months) [26]. A 2014 study by Kennedy et al. also observed chronic pain (defined as pain lasting >3 months) in 19% of adults from a large sample of the general population in the United States [27].
However, a recent, 2016, literature review by Fayaz, regarding the general population of the United Kingdom showed higher rates of chronic pain ranging from 35% to 51% [28].
Like the study by Kennedy et al., the review by Fayaz defined chronic pain as pain lasting 3 months or more.
Another important aspect of our study involved determining the severity of chronic pain.
In our study population the average pain intensity over the previous 24 hours was rated as moderate or severe by one in five subjects (22.8%). This is a much lower proportion than that reported in other studies, already mentioned above. For instance, the study by Miaskowski conducted in a cohort of the homeless in San Francisco, showed 92% of the chronic pain cases to be of moderate to severe intensity [29]. Most of the 18 papers on the severity of chronic pain evaluated as part of the already mentioned systematic review by Parker et al. showed moderate to severe pain [15]. In comparison, in the study by Uebelacker et al. the vast majority (81%) of patients with chronic pain rated its average intensity over the previous week as moderate or severe [17]. In Lawson's study, subjects rated the pain felt at that moment in an 11-point Visual Analogue Scale (VAS), and the median pain severity was 5 (i.e. moderate pain) [18]. The severity of chronic pain in HIVnegative populations can be compared, as before, with that in the 2006 study by Breivik et al., where most subjects rated the last pain they felt as moderate or severe (NRS was used, as in our study).
Comparing the individual studies in terms of pain severity also raises many doubts due to the differences in study methodologies. Comparison difficulties may be due to the differences in the questions on pain intensity, on the period when the pain was present ("right now", "within the last 24 hours", "within the last week"), and on the pain intensity "on the average", "at its worst", "at its least" in the given period. HIV-positive individuals in Poland may be subject to a greater stigmatization, particularly self-stigmatization. This, in turn, may lead to lowered expectations in terms of their quality of life and, in consequence, to underrating their pain intensity. The Stigma Index study clearly showed that in the populations of some Eastern European countries (Poland, Estonia, Moldova, Turkey, and Ukraine) stigmatization of HIV-positive individuals is greater than that in Western Europe [30]. In our opinion, future studies aiming to compare the prevalence of pain in HIV-positive populations in various countries should put more emphasis on evaluating the factors that affect the subjective assessment of pain as well as the unquantifiable causative factors responsible for the severity and nature of pain [30]. The study protocol had been approved by the Bioethics Committee (KB/292/2013). All participants gave written informed consent prior to their inclusion in the study.

Consent for publication
The datasets used and/or analysed during the current study are available from the corresponding author on reasonable request.   Chronic pain severity (n=57) rated with Brief Pain Inventory (BPI) Short Form