Abstract
Study Objectives
The landscape of HIV infection has shifted dramatically over the last few decades. An extended lifespan has led to an increase in comorbidities, including disorders of sleep. While self-reported sleep disturbances in people living with HIV are common, differences in sleep architecture between those living with and without HIV have not been previously described.
Methods
Polysomnography data from the Multicenter AIDS Cohort Study were used to characterize differences in sleep architecture between men living with and without HIV. Parameters assessed included total sleep time, sleep stage distribution, arousal index, and frequency of sleep stage transitions. Multivariable regression was employed to adjust for demographic variables and explore effect modification by sleep-disordered breathing (SDB) severity.
Results
Compared to men without HIV (N = 349), men with HIV (N = 447) exhibited comparable total sleep time, but lower sleep efficiency and greater wake time after sleep onset. Independent of HIV status, SDB was associated with a greater percentage of N1 sleep and lower percentages of N2 and REM sleep. However, those with both HIV and severe SDB displayed the lowest sleep efficiency, the highest percentage of N1 sleep, and the lowest frequency of sleep stage transitions from nonrapid eye movement (non-REM)-to-REM sleep compared to all other HIV and SDB subgroups.
Conclusions
This study found an independent association between HIV, SDB, and altered sleep architecture, characterized by lower sleep efficiency, greater time in stage N1 sleep, and higher sleep stage instability. Further research is needed on the potential health implications of disrupted sleep in those with HIV and SDB.
Keywords: human immunodeficiency virus (HIV), sleep architecture, polysomnography, sleep-disordered breathing
Graphical Abstract
Graphical Abstract.
Statement of Significance This study addresses a critical gap in understanding sleep architecture in men living with HIV in the era of effective antiretroviral therapy. By leveraging objective sleep data from a large, well-characterized cohort, this work examines the independent and interactive effects of HIV and sleep-disordered breathing (SDB) on sleep structure. The results show that severe SDB, in combination with HIV, is associated with significant alterations in sleep architecture, including lower sleep efficiency, increased wake time, and a shift toward lighter sleep stages. These findings provide important insights that can guide future investigations into the clinical implications of altered sleep in people living with HIV, particularly given the growing recognition of sleep disorders in this population.
Introduction
In the era of combination antiretroviral therapy, the landscape of human immunodeficiency virus (HIV) infection has undergone a significant transformation. What was once a rapidly progressing disease has evolved into a chronic condition, with life expectancy now comparable to that of people without HIV [1]. This increase in life expectancy, however, has led to a notable rise in other chronic medical conditions, such as obstructive lung disease [2], type 2 diabetes mellitus [3], and cardiovascular disease [4]. Additionally, sleep disorders have emerged as a prevalent comorbidity, affecting approximately 60% of those living with HIV.[5] While self-reported sleep disturbances in HIV have received considerable attention [5,6], relatively little is known about the independent impact of HIV on sleep architecture. Initial reports from the 1980s, which included a small number of male patients, described various alterations in sleep architecture. These included a decrease in latency to sleep onset and the percentages of stage 2 and rapid eye movement (REM) sleep [7–9]. Other reported alterations include an increase in the arousal index, the frequency of sleep stage transitions, and the amount of slow wave sleep [7–9].
Subsequent to those early observations, which predate the advent of combined antiretroviral therapy, epidemiological and clinical studies have focused primarily on characterizing self-reported disturbance in sleep [10]. Sleep maintenance insomnia is a common problem in people living with HIV [11–19] along with habitually short sleep duration [20,21] and disordered circadian patterns [22,23]. While information on the burden of sleep disorders in people living with HIV has grown, there has been a relative lack of advancement in understanding how HIV affects sleep architecture, including the distribution of sleep stages. While sleep-disordered breathing (SDB) is more prevalent in HIV and is known to alter sleep architecture [24], the combined effects of HIV and SDB are also not known. Therefore, the overarching goal of this study was to compare sleep architecture between men living with and without HIV and determine whether SDB modifies the effects of HIV on sleep structure. To accomplish these objectives, polysomnographic data collected in the Multicenter AIDS Cohort Study (MACS) [25] were used to investigate the associations between HIV, SDB, and sleep architecture to obtain insights not available from self-reported measures.
Materials and Methods
Study population
The MACS, now integrated into the MACS-WIHS Combined Cohort Study, is a longitudinal cohort study on the natural history and progression of HIV disease among men who have sex with men [25]. Participant recruitment occurred across four enrollment waves (1984–1985, 1987–1991, 2001–2003, and 2010–2017) in Baltimore/Washington DC, Chicago, Pittsburgh/Columbus, and Los Angeles. In the original MACS enrollment (1984–1985), there was no test for HIV and HIV status was not known. When the HIV test became available, it was determined that about one-third of the men had HIV at enrollment. Many of the men initially without HIV acquired it in the first few years of the MACS, so that the cohort became roughly balanced between men with and without HIV, and this balance was maintained in later recruitment cycles. In these cycles, age was an enrollment criterion, with younger men generally being over-recruited to provide a balanced range of ages within the cohort. Before the advent of effective antiretroviral therapy in 1996, there were excess deaths among men with HIV due to AIDS, and this is the primary reason why in the MACS, overall, the men without HIV were older than the men with HIV. This difference is reflected in the men who enrolled in the MACS ancillary sleep assessment, which generally recruited from the overall cohort. Thus, in the overall and ancillary sleep study cohorts, men without HIV were not specifically matched to men living with HIV. [26]. Semiannual study visits included standardized interviews covering health behaviors, physical exams, anthropometric measurements, and assessments of T-cell subsets and plasma HIV RNA concentration. Between March 2018 and June 2019, an ancillary study was conducted within the MACS to characterize sleep using home polysomnography. All participants in the MACS were invited to enroll, and those who did provided informed consent. The study protocol approved by the Institutional Review Board at each clinical site [26].
Home polysomnography
The procedures for home polysomnography and participant guidance have been previously described [26]. Briefly, home polysomnography was conducted using a self-applied recorder (Nox A1, Nox Medical, Reykjavik, Iceland), which collected data on a frontal electroencephalogram (EEG) montage with the following derivations: AF4, AF3, AF7, and AF8. Additional physiological data recorded included the frontalis muscle electromyogram (EMG), the electrocardiogram, right and left anterior tibialis EMG, nasal airflow using a pressure transducer, pulse oximetry, and chest and abdominal respiratory movement. Digital data from the Nox A1 recorder were transmitted to a central reading facility for manual scoring of sleep stages and disordered breathing events. Sleep was categorized as wake, non-REM (NREM stage N1, N2, and N3) sleep, or REM sleep using 30-s epochs. Apneas were defined as an absence or near absence of airflow for at least 10 s. Hypopneas were defined as a reduction in airflow by 30% or more for at least 10 s, along with a decrease of ≥4% in oxygen saturation. The resulting apnea-hypopnea index (AHI4) was calculated as the number of apneas and hypopneas per hour detected during sleep. SDB was considered present if the AHI4 was ≥5 events/h. Severity of SDB was categorized as mild (5.0–14.9 events/h), moderate (15.0–29.9 events/h), or severe (≥30 events/h). Arousals were scored if there was an abrupt shift of EEG frequency including alpha, theta, and/or frequencies greater than 16 Hz lasting 3 s or more with at least 10 s of stable sleep preceding the change. The arousal frequency was defined as the number of arousals per hour of total sleep time.
Statistical analyses
Dependent variables derived from overnight polysomnography included metrics such as the total sleep time and the distribution of sleep stages (N1, N2, N3 [slow-wave], and REM sleep). Time to sleep onset (i.e. sleep latency) and time to the first episode of REM sleep were also assessed. In addition, indices reflecting sleep fragmentation were examined, including the overall arousal index (stratified by NREM and REM sleep) and the number of transitions between Wake, NREM, and REM sleep. Multivariable linear regression was utilized to model associations between HIV status and parameters of sleep architecture, including total sleep time, percentages of stages N1, N2, N3, and REM, as well as the arousal index. Poisson regression was employed to model the number of sleep stage transitions for NREM-to-wake, REM-to-wake, NREM-to-REM, and REM-to-NREM [27]. Adjusted estimates were derived to accommodate demographic and anthropometric (e.g. BMI) variations between men with and without HIV, utilizing multivariable models that incorporated age, race, and BMI as covariates. Race was categorized (Black, White, and Other), with White race serving as the reference group. Given the high prevalence of SDB in men with HIV [28] and its known effects on sleep quality, potential effect modification between HIV status and SDB severity (none, mild, moderate, or severe) was explored using interaction terms between the two variables. Because smoking status differed between men with and without HIV, sensitivity analyses were conducted including smoking status as a covariate. This inclusion did not change the point estimates in any of the models, and therefore, smoking status was not included for statistical parsimony. Additional analyses were also conducted only in men with HIV to explore the associations between antiretroviral therapy class (integrase inhibitor, nonnucleoside reverse transcriptase inhibitor, or protease inhibitor), CD4 cell count, viral load, and sleep parameters. Adjusted estimates and the corresponding 95% confidence intervals (CI) from the multivariable models are reported herein. All analyses were performed using Stata 17.0 software (Stata Inc., College Station, TX), with statistical significance set at a value of .05.
Results
The study sample included 447 men (56.2%) living with HIV and 349 men (43.8%) living without HIV. Compared to men without HIV, those with HIV had similar BMI values (Table 1), but were younger and more likely to be current smokers and of nonwhite race. The distribution of SDB severity, as assessed by the AHI4, was also similar between the two groups (Table 1). The vast majority of men with HIV (93.8%) were on antiretroviral therapy and were virologically suppressed (92.6%), with a median CD4 cell count of 700/μl. Among men with HIV, 46.7% were receiving integrase inhibitors, 24.4% were receiving nonnucleoside reverse transcriptase inhibitors, and 20.8% were receiving protease inhibitors.
Table 1.
Characteristics* of men with polysomnography by HIV status
| Men with HIV (N = 447) |
Men without HIV (N = 349) |
p † | |||
|---|---|---|---|---|---|
| Age, years | 56.0 | (49.0–63.0) | 63.0 | (56.0–68.0) | <.001 |
| Body mass index, kg/m2 | 26.6 | (23.5–30.3) | 27.0 | (24.1–30.5) | .31 |
| Race, % | |||||
| White | 252 | (56.4%) | 269 | (77.1%) | <.001 |
| Black | 146 | (32.6%) | 63 | (18.0%) | |
| Other | 49 | (11.0%) | 17 | (4.9%) | |
| Smoking status | |||||
| Never | 146 | (33.7%) | 122 | (35.5%) | <.001 |
| Former | 188 | (43.4%) | 180 | (52.3%) | |
| Current | 99 | (22.9%) | 42 | (12.2%) | |
| Apnea-hypopnea index, events/h | |||||
| <5.0 | 211 | (47.2%) | 160 | (45.8%) | .38 |
| 5.0–14.9 | 141 | (31.5%) | 104 | (29.8%) | |
| 15.0–29.9 | 60 | (13.4%) | 45 | (12.9%) | |
| ≥30.0 | 35 | (7.8%) | 40 | (11.5%) | |
| HIV viral load <200 copies/mL | 414 | (92.6%) | – | ||
| Antiretroviral therapy | 411 | (91.9%) | – | ||
| Integrase inhibitor | 209 | (46.7%) | – | ||
| NNRTI‡ | 109 | (24.4%) | – | ||
| Protease inhibitor | 93 | (20.8%) | – | ||
| CD4 cell count per μL | 700 | (517–900) | – | ||
| HIV viral load (copies/mL) | <20 | (1–20) | – | ||
Abbreviation: MACS, Multicenter AIDS Cohort Study.
*Values shown are medians (interquartile range).
† p-value for comparing men with and without HIV.
‡NNRTI, nonnucleoside reverse transcriptase inhibitor.
Sleep architecture variables stratified by HIV status are shown in Table 2. Men living with HIV had lower values for time in bed (411.9 vs. 427.5 min, p = .02) and total sleep time (364.7 vs. 382.5 min; p = .006) than men without HIV. Unadjusted latencies to sleep onset (20.9 vs. 16.9 min; p = .03) and REM sleep (118.0 vs. 107.7 min; p = .07) were longer in men with HIV. No significant differences were observed between the two groups in the unadjusted values of sleep efficiency, wake time after sleep onset, percentages of NREM (i.e. N1, N2, N3) or REM sleep, and the arousal index. Sleep architecture variables were further stratified by HIV status and SDB severity (Table 3). Unadjusted values for time spent in bed and total sleep time were similar across categories of SDB severity and were not significantly different between men with and without HIV. Sleep efficiency was lower with increasing SDB severity, but only in men with HIV (p = .09 for linear trend across AHI categories). Additionally, time awake after sleep onset was higher with greater SDB severity only in men with HIV (p = .01 for linear trend across AHI categories). No differences were noted in either latency to sleep onset or to REM sleep across SDB severity irrespective of HIV status. Both groups of men displayed higher percentages of stage N1 sleep and a corresponding decrease in the percentage of stage N2 sleep with greater severity of SDB. Men with HIV and severe SDB exhibited the lowest percentages of stage N3 and REM sleep, whereas no distinct trend was observed for these sleep stages in men without HIV. Not surprisingly, the arousal index demonstrated an increase with AHI in men with and without HIV.
Table 2.
Unadjusted sleep architecture parameters* by HIV status
| Men with HIV (N = 447) |
Men without HIV (N = 349) |
p † | |||
|---|---|---|---|---|---|
| Time in bed, min | 411.9 | (95.4) | 427.5 | (86.6) | .02 |
| Total sleep time, min | 364.7 | (91.5) | 382.5 | (86.9) | .006 |
| Sleep Efficiency. % | 89.3 | (9.9) | 90.3 | (9.0) | .12 |
| Wake time after sleep onset, min | 47.1 | (45.7) | 45.2 | (41.8) | .53 |
| Latency to initial sleep onset, min | 20.9 | (29.1) | 16.9 | (21.9) | .03 |
| Latency to REM sleep onset, min | 118.0 | (86.1) | 107.7 | (74.0) | .07 |
| Stage N1 sleep, % | 19.4 | (11.3) | 19.8 | (10.7) | .58 |
| Stage N2, sleep, % | 62.5 | (12.2) | 62.6 | (11.5) | .96 |
| Stage N3 sleep, % | 4.0 | (6.7) | 3.7 | (6.3) | .61 |
| Stage REM sleep, % | 13.9 | (7.1) | 13.8 | (7.0) | .81 |
| Arousal index, events/h | 15.7 | (9.5) | 15.2 | (8.7) | .39 |
*Values reported are unadjusted means (SD).
† p-values for comparing men living with and without HIV.
Table 3.
Unadjusted sleep architecture parameters* by HIV status and AHI category
| Variable* | HIV status | Apnea-hypopnea index (events/h) | p † | |||||||
|---|---|---|---|---|---|---|---|---|---|---|
| <5.0 | 5.0–14.9 | 15.0–29.9 | ≥30.0 | |||||||
| Time in bed, min | + | 405.9 | (103.3) | 413.4 | (84.3) | 427.8 | (98.7) | 419.3 | (82.1) | .19 |
| − | 425.3 | (88.6) | 428.9 | (85.4) | 409.3 | (78.1) | 453.3 | (87.8) | .33 | |
| Total sleep time, min | + | 363.2 | (97.3) | 365.7 | (87.5) | 373.1 | (85.6) | 356.2 | (83.4) | .9 |
| − | 381.2 | (87.6) | 380.9 | (85.1) | 360.3 | (86.4) | 415.7 | (83.0) | .26 | |
| Sleep efficiency, % | + | 90.0 | (8.7) | 88.8 | (11.0) | 89.6 | (10.1) | 86.4 | (11.7) | .09 |
| − | 90.6 | (8.7) | 90 | (9.0) | 88.1 | (11.3) | 92.4 | (6.6) | .87 | |
| Wake after sleep onset, min | + | 42.8 | (37.0) | 47.8 | (42.9) | 51.6 | (69.5) | 63.1 | (51.1) | .01 |
| − | 44.2 | (40.7) | 48 | (44.5) | 49 | (42.3) | 37.6 | (30.7) | .73 | |
| Latency to initial sleep onset, min | + | 22.7 | (33.2) | 16.9 | (18.3) | 25.6 | (38.0) | 17.6 | (17.9) | .56 |
| − | 16.3 | (22.5) | 16.4 | (17.9) | 20 | (28.3) | 16.8 | (22.1) | .59 | |
| Latency to REM sleep onset, min | + | 114.0 | (84.4) | 119.8 | (78.8) | 111.6 | (94.6) | 145.4 | (106.2) | .17 |
| − | 101.5 | (75.6) | 111.1 | (67.7) | 105.9 | (67.8) | 125.9 | (87.9) | .09 | |
| Stage N1 sleep, % | + | 16.6 | (10.0 | 19.9 | (10.0) | 21 | (9.3) | 31.6 | (17.3) | <.001 |
| − | 17.4 | (9.3) | 19.2 | (8.2) | 25.5 | (13.3) | 24.9 | (14.4) | <.001 | |
| Stage N2, sleep, % | + | 63.4 | (12.1) | 62.9 | (11.2) | 62.4 | (11.8) | 55.9 | (15.2) | .006 |
| − | 64.2 | (10.5) | 63 | (10.0) | 58.4 | (12.3) | 59.5 | (14.7) | .001 | |
| Stage N3 sleep, % | + | 5.1 | (7.9) | 3.2 | (5.4) | 3.1 | (5.8) | 1.9 | (3.7) | .001 |
| − | 4.3 | (6.4) | 3.3 | (7.2) | 3.3 | (5.7) | 3.1 | (4.1) | .16 | |
| Stage REM sleep, % | + | 14.7 | (7.1) | 13.6 | (6.8) | 14 | (6.9) | 10.7 | (8.1) | .007 |
| − | 14.0 | (7.2) | 14.5 | (6.6) | 12.7 | (6.6) | 12.5 | (7.7) | .16 | |
| Arousal index, events/h | + | 12.4 | (7.3) | 16.3 | (8.4) | 20 | (9.9) | 25.8 | (13.3) | <.001 |
| − | 12.4 | (7.0) | 14.4 | (6.7) | 18.8 | (8.2) | 24.1 | (12.6) | <.001 | |
*Values reported are unadjusted means (SD).
† p-values for linear trend across categories of the AHI in men living with and without HIV.
Given the differences in demographic variables (e.g. age, race) between men with and without HIV, multivariable regression analyses were used to quantify the adjusted differences between the two groups. After including age, race, and BMI in multivariable models, time spent in bed (Figure 1A) was similar between men with and without HIV across all four categories of SDB severity. For men without SDB, or those with either mild or moderate SDB, total sleep time (Figure 1B), sleep efficiency (Figure 1C), and wake after sleep onset (Figure 1D) did not differ significantly by HIV status. However, among those with severe SDB, total sleep time and sleep efficiency were significantly lower, and wake after sleep onset was significantly higher, in men with HIV than without HIV. The adjusted latency to initial sleep onset (Figure 2A) was not statistically different across HIV status or SDB categories. While latency to REM sleep onset was also similar in men with and without HIV, an increasing trend was observed with SDB severity (Figure 2B; p = .03 for linear trend across AHI categories).
Figure 1.
Adjusted mean values and 95% CIs for time in bed (A), total sleep time (B), sleep efficiency (C), and wake after sleep onset (D) for men with and without HIV across categories of the AHI. Values are adjusted for age, race, and body mass index. *p-values for comparing men with and without HIV who had an AHI ≥ 30 events/h.
Figure 2.
Adjusted mean values and 95% CIs for latencies to initial sleep onset (A) and first episode of REM sleep (B) for men with and without HIV across categories of the AHI. Values are adjusted for age, race, and body mass index with no differences between men with and without HIV.
Figure 3 shows the adjusted percentages of sleep stages (i.e. N1, N2, N3, and REM) across HIV status and SDB categories. With increasing SDB severity, the percentage of stage N1 sleep increased (p < .001 for linear trend across AHI categories; Figure 3A), whereas percentages of stage N2 (Figure 3B) and REM sleep (Figure 3D) decreased (both with p < .001 for linear trend across AHI categories). No trend in stage N3 sleep was noted as a function of SDB severity (Figure 3C). Moreover, the percentage of stage N1 sleep was significantly higher in men with HIV than without HIV and concurrent severe SDB. No other significant differences were noted across HIV status in percentage of stage N2, N3, and REM sleep. In the overall study population, the age-, race-, and BMI-adjusted arousal indices across the four SDB categories were 12.5 (95% CI = 11.5%–13.5%), 15.6 (95% CI = 14.5%–16.8%), 19.7 (95% CI = 18.0%–21.3%), and 25.2 (95% CI = 23.2%–27.2%), respectively. No differences in the arousal index by HIV status were observed, regardless of SDB severity (data not shown).
Figure 3.
Adjusted mean values and 95% CIs for percentage of stage N1 (A), N2 (B), N3 (C), and REM (D) sleep for men with and without HIV across categories of the AHI. Values are adjusted for age, race, and body mass index. *p-values for comparing men with and without HIV who had an AHI ≥ 30 events/h.
Transition frequencies between NREM, REM, and wakefulness were subsequently examined to characterize sleep microarchitecture. Unadjusted transition frequencies from NREM-to-wake and from REM-to-wake, as well as between NREM and REM, are shown in Table 4. With increasing SDB severity, there was a notable increase in the NREM-to-wake transition frequency (p < .001 for linear trend across AHI categories), with no evidence of heterogeneity between men with and without HIV (Figure 4A). The REM-to-wake transition frequency showed no trend with increasing SDB severity and was similar between men with and without HIV (Figure 4B). After adjusting for age, race, and BMI, men with HIV and severe SDB exhibited a lower frequency of NREM-to-REM and REM-to-NREM transitions compared to the other groups (Figure 4C and D). Analyses restricted to men with HIV did not reveal any significant associations between any of the sleep architecture parameters and HIV viral load, CD4 cell count, or type of antiretroviral therapy (data not shown).
Table 4.
Unadjusted average (SD) frequency of sleep stage transitions per hour of sleep by AHI category and HIV status
| Transition type | HIV status | Apnea-hypopnea index (events/h) | p * | |||||||
|---|---|---|---|---|---|---|---|---|---|---|
| <5.0 | 5.0–14.9 | 15.0–29.9 | ≥30.0 | |||||||
| NREM-to-Wake | + | 17.5 | (10.2) | 20.3 | (12.0) | 20.4 | (10.5) | 26.1 | (19.2) | <.001 |
| − | 17.6 | (9.9) | 20.0 | (11.3) | 19.6 | (11.5) | 24.4 | (19.9) | <.001 | |
| REM-to-Wake | + | 2.5 | (2.7) | 2.4 | (2.0) | 2.4 | (2.0) | 2.2 | (2.6) | .2 |
| − | 2.5 | (2.3) | 2.4 | (2.9) | 2.7 | (3.2 | 2.8 | (4.0) | .17 | |
| NREM-to-REM | + | 8.1 | (5.2) | 8.4 | (5.3 | 9.3 | (6.9) | 5.6 | (4.3 | .09 |
| − | 8.5 | (5.6) | 9.2 | (5.1) | 9 | (6.1) | 10.4 | (7.4) | .001 | |
| REM-to-NREM | + | 5.4 | (4.5) | 6.0 | (4.4) | 7.2 | (6.7) | 3.7 | (3.2 | .84 |
| − | 5.9 | (5.1) | 6.2 | (4.4) | 6.3 | (5.5) | 7.7 | (7.0) | <.001 | |
* p-values for linear trend across categories of the AHI in men living with and without HIV.
Figure 4.
Adjusted frequency of transitions and 95% CIs for NREM-to-wake (A), REM-to-wake (B), NREM-to-REM (C), and REM-to-NREM (D) transitions for men with and without HIV across categories of the AHI. Values are adjusted for age, race, and body mass index. *p-values for comparing men with and without HIV who had an AHI ≥ 30 events/h.
Discussion
Although a few studies have examined the association between HIV and sleep architecture [7–9], none have been conducted in the era of effective antiretroviral therapy. Furthermore, no studies have examined the independent and interactive associations between HIV, SDB, and sleep structure. The present study is by far the largest study to characterize sleep architecture in men with and without HIV and assess the combined effects of HIV and SDB. The primary finding of this study is that the combination of HIV and severe SDB was significantly associated with lower sleep efficiency, greater wake time after sleep onset, and higher proportion of time spent in sleep stage N1. Interestingly, this group also had fewer NREM-to-REM sleep and REM-to-NREM sleep transitions. Moreover, the expected decrease in stage N2 and REM sleep associated with SDB was observed in both men with and without HIV, and was not affected by HIV. Fundamentally, this study shows differences in sleep architecture between men with and without HIV that indicate impaired sleep quality in the former but only with the co-occurrence of severe SDB.
In clinical studies of people without HIV, frequent arousals resulting from recurrent apneas and hypopneas with moderate to severe SDB instigates sleep stage transitions that result in: (1) a shift toward lighter stages of sleep (i.e. higher amounts of N1 and lower amounts of REM sleep) [24], (2) poor sleep efficiency, and (3) greater wake time after sleep onset. In the current study, some of these trends were greater in men with than without HIV. Such alterations, which constitute fragmented sleep, have been shown to impair daytime alertness and cognitive function [29,30]. Sleep fragmentation in healthy volunteers can acutely increase blood pressure in proportion to the intensity of the arousal [31]. Furthermore, murine models confirm that sleep fragmentation can not only increase blood pressure, but also impair vascular endothelial function [32]. Similarly, a number of clinical studies have also found an independent association between arousals and prevalent hypertension. Taken together, these data support the notion that disrupted sleep can negatively influence cardiovascular health [33–37]. Given the clinical sequelae associated with sleep fragmentation, additional research is needed to define the potential effects of altered sleep architecture on health outcomes in people living with HIV.
The underlying etiology of altered sleep architecture in people living with HIV remains unclear. Most likely, immune activation and the chronic inflammatory state resulting from HIV infection are contributing factors [38]. Additionally, the effects of ART must also be considered given that sleep disturbances persist in people with HIV regardless of treatment status [39]. However, in the current study, none of the HIV-related markers, such as ART use, CD4 cell count, or HIV viral load, were associated with altered sleep architecture. These findings suggest that, despite adherence to ART, the persistent neuropathology documented with HIV infection [40,41], even with undetectable viral load, may continue to disturb sleep continuity particularly in the presence of severe SDB. Recent findings suggest that HIV infection can disrupt circadian rhythms, as evidenced by delayed dim light melatonin onset and a shorter phase angle of entrainment. These disruptions could contribute to sleep disturbances independent of ART use or immune status. Thus, persistent neuropathology and other HIV-related mechanisms, such as circadian misalignment, may play a role in the disruption of sleep architecture in this population [23]. Emerging evidence also indicates that people living with HIV and SDB may experience more severe symptoms due to the combined effects of HIV-related inflammation and SDB-associated intermittent hypoxemia [42]. These conditions can exacerbate cardiovascular, metabolic, and neurocognitive comorbidities, contributing to increased fatigue, systemic inflammation, and reduced quality of life. Importantly, people living with HIV and SDB often present with sleep-related symptoms despite the absence of traditional SDB risk factors [43]. This atypical presentation may delay the timely diagnosis of SDB, underscoring the need for routine screening in people with HIV, particularly when symptoms of sleepiness [44] or fatigue persist despite effective suppression of HIV replication through ART. Translating the findings of the current study into potential interventions is a crucial next step. Given the sleep fragmentation observed with SDB, positive airway pressure therapy holds promise for improving sleep continuity and mitigating associated cardiovascular risk in people living with HIV. Additionally, cognitive-behavioral therapy for insomnia may address co-occurring insomnia and SDB in this population, potentially enhancing both sleep quality and overall quality of life, as observed in individuals without HIV. However, dedicated interventional studies in people living with HIV are necessary to determine whether these therapies can improve their sleep, quality of life, and cardiovascular health.
This study has several notable strengths. First, the use of home polysomnography provided an objective assessment of sleep, minimizing the biases often associated with self-reported data. Second, the MACS provided a well-characterized cohort of men with and without HIV, enabling a comprehensive assessment of associations between HIV, SDB, and sleep architecture. Third, the large sample size permitted rigorous analysis across several measures of sleep architecture. By evaluating total sleep time, sleep stage distribution, arousal index, and transitions between sleep stages, the current study provides a detailed understanding of the associations between HIV, SDB, and sleep quality. Despite these strengths, the current study also has some limitations. Because it included only men, generalizability of the findings is limited. Furthermore, its cross-sectional design precludes establishing causal associations between HIV status and sleep quality. While the use of home polysomnography is a strength, as it allows for assessments of sleep in the participant’s natural environment, it could influence certain measurements, such as sleep latency and total sleep time. However, any such influence is likely to have been nondifferential with respect to HIV status, given that all participants underwent identical procedures. As a result, the observed differences in sleep parameters between men with and without HIV are unlikely to be biased. Another limitation is that the data were collected more than 5 years ago, during which antiretroviral regimens have evolved. These changes could potentially affect HIV-related outcomes and comorbidities, such as sleep disturbances, and may limit the generalizability to current clinical populations. However, current HIV treatments are largely consistent with those used 5 years ago, with integrase inhibitors remaining the most common components of ART regimens. The primary difference is the availability of long-acting cabotegravir, though its uptake has been limited. Future research using more recent cohorts and incorporating evolving treatment protocols will be essential to validate and expand upon the findings reported herein. Finally, while this study focused on characterizing the associations between HIV and sleep quality, it did not examine potential links between sleep disturbances (e.g. sleep stage distribution, transitions) and cardiometabolic or other health outcomes. Establishing these associations was a necessary first step before investigating sleep disturbances as causal intermediaries for health outcomes. Ongoing research aims to address these critical questions comprehensively, and future studies should prioritize elucidating the causal pathways between sleep disturbances and cardiometabolic health in this population.
Acknowledgments
The authors gratefully acknowledge the contributions of the study participants and dedication of the staff at the MACS sites.
Contributor Information
Naresh M Punjabi, Department of Medicine, Miller School of Medicine, University of Miami, Miami, FL, USA.
Todd T Brown, Department of Medicine, School of Medicine, Johns Hopkins University, Baltimore, MD, USA.
Darko Stefanovski, Clinical Studies-New Bolton Center, University of Pennsylvania, Philadelphia, PA, USA.
Rashmi Nisha Aurora, Department of Medicine, Grossman School of Medicine, NYU, New York City, NY, USA.
Sanjay R Patel, Department of Medicine, School of Medicine, University of Pittsburgh, Pittsburgh, PA, USA.
Valentina Stosor, Department of Medicine, Northwestern University Feinberg School of Medicine, Chicago, IL, USA.
Joshua Hyong-Jin Cho, Department of Psychiatry, University of California (Los Angeles, David Geffen School of Medicine), Los Angeles, CA, USA.
Gypsyamber D’Souza, Molecular Microbiology and Immunology, Johns Hopkins Bloomberg School of Public Health, Baltimore, MD, USA.
Joseph B Margolick, Molecular Microbiology and Immunology, Johns Hopkins Bloomberg School of Public Health, Baltimore, MD, USA.
Disclosure statement
Financial disclosure: Dr. Punjabi received support from grants HL117167, HL146709, and HL118414 from the National Heart, Lung, and Blood Institute. The MACS Combined Cohort Study (MWCCS) was supported by the following grants: Baltimore CRS (Todd Brown and Joseph Margolick), U01-HL146201; Data Analysis and Coordination Center (Gypsyamber D’Souza, Stephen Gange, and Elizabeth Topper), U01-HL146193; Chicago-Northwestern CRS (Steven Wolinsky, Frank Palella, and Valentina Stosor), U01-HL146242; Los Angeles CRS (Roger Detels and Matthew Mimiaga), U01-HL146333; Pittsburgh CRS (Jeremy Martinson and Charles Rinaldo). The MWCCS is funded primarily by the National Heart, Lung, and Blood Institute (NHLBI), with additional co-funding from the Eunice Kennedy Shriver National Institute of Child Health & Human Development (NICHD), National Institute on Aging (NIA), National Institute of Dental & Craniofacial Research (NIDCR), National Institute of Allergy and Infectious Diseases (NIAID), National Institute of Neurological Disorders and Stroke (NINDS), National Institute of Mental Health (NIMH), National Institute on Drug Abuse (NIDA), National Institute of Nursing Research (NINR), National Cancer Institute (NCI), National Institute on Alcohol Abuse and Alcoholism (NIAAA), National Institute on Deafness and Other Communication Disorders (NIDCD), National Institute of Diabetes and Digestive and Kidney Diseases (NIDDK), National Institute on Minority Health and Health Disparities (NIMHD), and in coordination and alignment with the research priorities of the National Institutes of Health, Office of AIDS Research (OAR). MACS data collection was also supported by UL1-TR003098 (JHU ICTR), UL1-TR001881 (UCLA CTSI).
Non-financial disclosure: None.
References
- 1. Samji H, Cescon A, Hogg RS, et al. Closing the gap: increases in life expectancy among treated HIV-positive individuals in the United States and Canada. PLoS One. 2013;8(12):e81355. 10.1371/journal.pone.0081355 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2. Drummond MB, Kunisaki KM, Huang L. Obstructive lung diseases in HIV: a clinical review and identification of key future research needs. Semin Respir Crit Care Med. 2016;37(02):277–288. 10.1055/s-0036-1578801 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3. Hadigan C, Kattakuzhy S. Diabetes mellitus type 2 and abnormal glucose metabolism in the setting of human immunodeficiency virus. Endocrinol Metab Clin North Am. 2014;43(3):685–696. 10.1016/j.ecl.2014.05.003 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4. Ballocca F, D'Ascenzo F, Gili S, Grosso MW, Gaita F. Cardiovascular disease in patients with HIV. Trends Cardiovasc Med. 2017;27(8):558–563. 10.1016/j.tcm.2017.06.005 [DOI] [PubMed] [Google Scholar]
- 5. Wu J, Wu H, Lu C, Guo L, Li P. Self-reported sleep disturbances in HIV-infected people: a meta-analysis of prevalence and moderators. Sleep Med. 2015;16(8):901–907. 10.1016/j.sleep.2015.03.027 [DOI] [PubMed] [Google Scholar]
- 6. Lee S, Oh JW, Park KM, Ahn JY, Lee S, Lee E. The prevalence and moderating factors of sleep disturbances in people living with HIV: a systematic review and meta-analysis. Sci Rep. 2024;14(1):14817. 10.1038/s41598-024-65713-x [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7. Norman SE, Resnick L, Cohn MA, Duara R, Herbst J, Berger JR. Sleep disturbances in HIV-seropositive patients. JAMA. 1988;260(7):922. 10.1001/jama.1988.03410070046015 [DOI] [PubMed] [Google Scholar]
- 8. Norman SE, Chediak AD, Kiel M, Cohn MA. Sleep disturbances in HIV-infected homosexual men. AIDS. 1990;4(8):775–782. 10.1097/00002030-199008000-00009 [DOI] [PubMed] [Google Scholar]
- 9. Norman SE, Chediak AD, Freeman C, et al. Sleep disturbances in men with asymptomatic human immunodeficiency (HIV) infection. Sleep. 1992;15(2):150–155. 10.1093/sleep/15.2.150 [DOI] [PubMed] [Google Scholar]
- 10. Kunisaki KM, De Francesco D, Sabin CA, et al. Sleep disorders in human immunodeficiency virus: a substudy of the pharmacokinetics and clinical observations in people over fifty (POPPY) study. Open Forum Infect Dis. 2021;8(1):ofaa561. 10.1093/ofid/ofaa561 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11. Rubinstein ML, Selwyn PA. High prevalence of insomnia in an outpatient population with HIV infection. J Acquir Immune Defic Syndr Hum Retrovirol. 1998;19(3):260–265. 10.1097/00042560-199811010-00008 [DOI] [PubMed] [Google Scholar]
- 12. Phillips KD, Sowell RL, Rojas M, Tavakoli A, Fulk LJ, Hand GA. Physiological and psychological correlates of fatigue in HIV disease. Biol Res Nurs. 2004;6(1):59–74. 10.1177/1099800404264846 [DOI] [PubMed] [Google Scholar]
- 13. Phillips KD, Sowell RL, Boyd M, Dudgeon WD, Hand GA, Mind-Body Research Group. Sleep quality and health-related quality of life in HIV-infected African-American women of childbearing age. Qual Life Res. 2005;14(4):959–970. 10.1007/s11136-004-2574-0 [DOI] [PubMed] [Google Scholar]
- 14. Reid S, Dwyer J. Insomnia in HIV infection: a systematic review of prevalence, correlates, and management. Psychosom Med. 2005;67(2):260–269. 10.1097/01.psy.0000151771.46127.df [DOI] [PubMed] [Google Scholar]
- 15. Hudson AL, Portillo CJ, Lee KA. Sleep disturbances in women with HIV or AIDS: efficacy of a tailored sleep promotion intervention. Nurs Res. 2008;57(5):360–366. 10.1097/01.NNR.0000313501.84604.2c [DOI] [PubMed] [Google Scholar]
- 16. Jean-Louis G, Weber KM, Aouizerat BE, et al. Insomnia symptoms and HIV infection among participants in the Women's Interagency HIV Study. Sleep. 2012;35(1):131–137. 10.5665/sleep.1602 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17. Gamaldo CE, Spira AP, Hock RS, et al. Sleep, function and HIV: a multi-method assessment. AIDS Behav. 2013;17(8):2808–2815. 10.1007/s10461-012-0401-0 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18. Allavena C, Guimard T, Billaud E, et al. Prevalence and risk factors of sleep disturbance in a large HIV-infected adult population. AIDS Behav. 2016;20(2):339–344. 10.1007/s10461-015-1160-5 [DOI] [PubMed] [Google Scholar]
- 19. Duncan DT, Goedel WC, Mayer KH, et al. Poor sleep health and its association with mental health, substance use, and condomless anal intercourse among gay, bisexual, and other men who have sex with men. Sleep Health. 2016;2(4):316–321. 10.1016/j.sleh.2016.07.003 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20. Lee KA, Gay C, Portillo CJ, et al. Types of sleep problems in adults living with HIV/AIDS. J Clin Sleep Med. 2012;8(1):67–75. 10.5664/jcsm.1666 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21. Gamaldo CE, Gamaldo A, Creighton J, et al. Evaluating sleep and cognition in HIV. J Acquir Immune Defic Syndr. 2013;63(5):609–616. 10.1097/QAI.0b013e31829d63ab [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22. Taibi DM, Price C, Voss J. A pilot study of sleep quality and rest-activity patterns in persons living with HIV. J Assoc Nurses AIDS Care. 2013;24(5):411–421. 10.1016/j.jana.2012.08.001 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23. Redman KN, O'Brien KE, Ruiz FS, et al. Delayed circadian rhythms in older Africans living with human immunodeficiency virus (HIV). J Pineal Res. 2023;74(1):e12838. 10.1111/jpi.12838 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24. Fietze I, Quispe-Bravo S, Hansch T, Rottig J, Baumann G, Witt C. Arousals and sleep stages in patients with obstructive sleep apnoea syndrome: changes under nCPAP treatment. J Sleep Res. 1997;6(2):128–133. 10.1046/j.1365-2869.1997.00029.x [DOI] [PubMed] [Google Scholar]
- 25. Kaslow RA, Ostrow DG, Detels R, Phair JP, Polk BF, Rinaldo CR Jr. The Multicenter AIDS Cohort Study: rationale, organization, and selected characteristics of the participants. Am J Epidemiol. 1987;126(2):310–318. 10.1093/aje/126.2.310 [DOI] [PubMed] [Google Scholar]
- 26. Punjabi NM, Brown T, Aurora RN, et al. Methods for home-based self-applied polysomnography: the Multicenter AIDS Cohort Study. Sleep Adv. 2022;3(1):zpac011. 10.1093/sleepadvances/zpac011 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27. Cameron AC, Trivedi PK. Regression Analysis of Count Data. 2nd ed. Cambridge; New York, NY: Cambridge University Press; 2013. [Google Scholar]
- 28. Punjabi NM, Brown TT, Aurora RN, et al. Prevalence and predictors of sleep-disordered breathing in men participating in the Multicenter AIDS cohort study. Chest. 2023;163(3):687–696. 10.1016/j.chest.2022.10.030 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29. Stepanski EJ. The effect of sleep fragmentation on daytime function. Sleep. 2002;25(3):268–276. 10.1093/sleep/25.3.268 [DOI] [PubMed] [Google Scholar]
- 30. Nair D, Zhang SX, Ramesh V, et al. Sleep fragmentation induces cognitive deficits via nicotinamide adenine dinucleotide phosphate oxidase-dependent pathways in mouse. Am J Respir Crit Care Med. 2011;184(11):1305–1312. 10.1164/rccm.201107-1173OC [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31. Davies RJ, Belt PJ, Roberts SJ, Ali NJ, Stradling JR. Arterial blood pressure responses to graded transient arousal from sleep in normal humans. J Appl Physiol. 1993;74(3):1123–1130. 10.1152/jappl.1993.74.3.1123 [DOI] [PubMed] [Google Scholar]
- 32. Carreras A, Zhang SX, Peris E, et al. Chronic sleep fragmentation induces endothelial dysfunction and structural vascular changes in mice. Sleep. 2014;37(11):1817–1824. 10.5665/sleep.4178 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33. Ringler J, Basner RC, Shannon R, et al. Hypoxemia alone does not explain blood pressure elevations after obstructive apneas. J Appl Physiol. 1990;69(6):2143–2148. 10.1152/jappl.1990.69.6.2143 [DOI] [PubMed] [Google Scholar]
- 34. Sulit L, Storfer-Isser A, Kirchner HL, Redline S. Differences in polysomnography predictors for hypertension and impaired glucose tolerance. Sleep. 2006;29(6):777–783. 10.1093/sleep/29.6.777 [DOI] [PubMed] [Google Scholar]
- 35. Kuwabara M, Tomitani N, Shiga T, Kario K. Polysomnography-derived sleep parameters as a determinant of nocturnal blood pressure profile in patients with obstructive sleep apnea. J Clin Hypertens (Greenwich). 2018;20(6):1039–1048. 10.1111/jch.13308 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 36. Ren R, Zhang Y, Yang L, Somers VK, Covassin N, Tang X. Association between arousals during sleep and hypertension among patients with obstructive sleep apnea. J Am Heart Assoc. 2022;11(1):e022141. 10.1161/JAHA.121.022141 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 37. Morrell MJ, Finn L, Kim H, Peppard PE, Badr MS, Young T. Sleep fragmentation, awake blood pressure, and sleep-disordered breathing in a population-based study. Am J Respir Crit Care Med. 2000;162(6):2091–2096. 10.1164/ajrccm.162.6.9904008 [DOI] [PubMed] [Google Scholar]
- 38. O'Brien KE, Riddell NE, Gomez-Olive FX, Rae DE, Scheuermaier K, von Schantz M. Sleep disturbances in HIV infection and their biological basis. Sleep Med Rev. 2022;65:101571. 10.1016/j.smrv.2021.101571 [DOI] [PubMed] [Google Scholar]
- 39. Redman KN, Karstaedt AS, Scheuermaier K. Increased CD4 counts, pain and depression are correlates of lower sleep quality in treated HIV positive patients with low baseline CD4 counts. Brain Behav Immun. 2018;69:548–555. 10.1016/j.bbi.2018.02.002 [DOI] [PubMed] [Google Scholar]
- 40. Kumar AM, Ownby RL, Waldrop-Valverde D, Fernandez B, Kumar M. Human immunodeficiency virus infection in the CNS and decreased dopamine availability: relationship with neuropsychological performance. J Neurovirol. 2011;17(1):26–40. 10.1007/s13365-010-0003-4 [DOI] [PubMed] [Google Scholar]
- 41. Obermann M, Kuper M, Kastrup O, et al. Substantia nigra hyperechogenicity and CSF dopamine depletion in HIV. J Neurol. 2009;256(6):948–953. 10.1007/s00415-009-5052-3. [DOI] [PubMed] [Google Scholar]
- 42. Owens RL, Hicks CB. A wake-up call for human immunodeficiency virus (HIV) providers: obstructive sleep apnea in people living with HIV. Clin Infect Dis. 2018;67(3):472–476. 10.1093/cid/ciy217 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 43. Patil SP, Brown TT, Jacobson LP, et al. Sleep disordered breathing, fatigue, and sleepiness in HIV-infected and -uninfected men. PloS One. 2014;9(7):e99258. 10.1371/journal.pone.0099258 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 44. Orr JE, Velazquez J, Schmickl CN, et al. Sleep, sleep apnea, and fatigue in people living with HIV. J Acquir Immune Defic Syndr. 2024;97(2):192–201. 10.1097/QAI.0000000000003481 [DOI] [PMC free article] [PubMed] [Google Scholar]