https://orcid.org/0000-0002-0471-2035
Abstract
Background
Antiretroviral therapy (ART) is associated with gain in quantity of fat and muscle, but the impact on quality is less understood. The objective of this study was to compare fat and muscle density among people with HIV (PWH) on stable raltegravir (RAL), atazanavir with ritonavir (ATV/r), or darunavir with ritonavir (DRV/r), and explore implications on muscle function.
Methods
Participants from the Modena HIV Metabolic Clinic taking RAL, ATV/r, or DRV/r with at least 1 computed tomography (CT) scan were included. CT scans were reanalyzed for area and density of truncal fat and musculature. Multivariate models explored the effect of ART on fat and muscle density.
Results
One hundred six participants were receiving ATV/r, 48 DRV/r, and 141 RAL. In multivariate models (reference ATV/r), only DRV/r was associated with greater subcutaneous (SAT) and visceral adipose tissue (VAT) area, lower lateralis muscle density (more fat), and greater lateralis intermuscular fat area. Compared to ATV/r, RAL was independently associated with less psoas intermuscular fat area. Among all, greater paraspinal muscle density correlated with better physical function. No associations between ART group and physical function were seen among men; DRV/r was associated with stronger grip strength among women.
Conclusion
DRV/r was associated with greater fat area and lower density of both fat and muscle, and RAL with less intermuscular psoas fat. Higher density psoas and paraspinal musculature were associated with better physical function, suggesting potential clinical relevance of these findings.
Background
The contribution of antiretroviral therapy (ART) to body weight and composition has changed considerably over the last three decades. Initially, weight gain with ART was viewed as a “return to health” phenomenon, particularly among people with HIV (PWH) with advanced disease and wasting.Citation1,Citation2 Subsequent regimens were associated with localized deposition of body fat, or lipohypertrophy, primarily in the viscera, neck, or upper back.Citation3,Citation4 More recently, contemporary ART regimens including integrase strand transferase inhibitors (INSTIs) and tenofovir alafenamide (TAF) have been implicated in contributing to a greater than expected generalized weight gain, even among participants without preceding HIV-related weight loss. In the limited studies with body composition assessments through dual energy X-ray absorptiometry (DXA) or computed tomography (CT), this weight gain appears to be due to gains in subcutaneous adipose tissue (SAT), visceral adipose tissue (VAT), and lean body mass (LBM).Citation5–7 For example, in AIDS Clinical Trials Group (ACTG) study A5260s, among PWH randomized to atazanavir with ritonavir (ATV/r), darunavir with ritonavir (DRV/r), or raltegravir (RAL), each with tenofovir disoproxil fumarate and emtricitabine, similar significant increases were seen among all groups in CT-measured VAT and SAT and DXA-measured LBM.
While the quantity of fat and LBM are easily characterized and linked to health outcomes, the quality of fat and LBM may have an additive or unique impact on the development of comorbidities or the function of the organ (i.e. muscle strength), but have not been routinely measured. In a subsequent analysis of A5260s, we reanalyzed CT scans to measure the density and area of trunk fat and musculature. Over 96 weeks, SAT and VAT density decreased significantly regardless of ART regimen.Citation8 We found that all three regimens also had significant decreases in CT-measured skeletal muscle density, consistent with fatty infiltration and lower quality skeletal muscle. Furthermore, only the total muscle area, but not the lean area, increased with ART.Citation6 Together, these findings suggest that the DXA-measured LBM gains seen with ART initiation may actually be gains in poor quality, low-density, fatty muscle. Whether these changes are associated with clinical manifestations of impaired muscle function in this population are not known, as physical function was not ascertained in the study. However, other cohorts of older adults without HIV suggest that muscle density may be strongly linked to impairments in physical function and falls.Citation9–11 In a separate cohort of men with HIV, we found that the density of SAT was strongly associated with greater inflammation, independent of SAT area.Citation12 Together, these results suggest that contemporary ART is associated with increased quantity of fat and skeletal muscle (lean mass) and development of “fattier” fat and muscle, and that the density of these body composition parameters may provide further informative beyond the area or mass alone.
The objective of this study was to compare the fat and muscle density among PWH on stable therapy with RAL, ATV/r, and DRV/r, and explore potential implications on muscle function. We hypothesized that RAL may be associated with greater SAT or VAT, but would be associated with the least impairment in muscle and fat quality, and this would be associated with better physical performance.
Methods
This was a retrospective observational study of all patients attending the Modena HIV Metabolic Clinic who underwent CT scan while on at least six months of stable ARV therapy with RAL, ATV/r, or DRV/r. The Modena HIV Metabolic Clinic, at the University of Modena and Reggio Emilia, Italy, was initiated in 2004 to assess metabolic changes among people with HIV and has been described elsewhere.Citation13 In this tertiary care teaching Hospital in Northern Italy, all individuals are screened by a multidisciplinary team for immunometabolic disorders, HIV-associated comorbidities, geriatric syndromes and frailty (including grip strength measured by a hand held dynamometer). Physical function impairment is measured using a Short Physical Performance Battery (SPPB), including measures of time to complete a 4-meter walk, ability to balance on a tandem stand, and time to rise from a chair five times.Citation14 Patients also receive a thoracic and abdominal CT to detect coronary artery calcium, lung parenchyma abnormalities, visceral adipose tissue and liver steatosis.
All procedures followed were in accordance with the ethical standards of Regional Ethical Committee and with the Helsinki Declaration of 1975, as revised in 2000; all participants provided written, informed consent.
De-identified data were extracted from the Modena HIV Metabolic Clinic electronic database. Body weight was measured using a digital scale to the nearest 0.1 kg with subjects wearing light clothes without shoes. Height was measured using a wall-mounted stadiometer to the nearest 0.1 cm. Measurements for body weight and height were completed in triplicate and the mean recorded. Body mass index (BMI) was defined as weight in kilograms divided by height in meters squared.
Body composition measurement
CT scans were reanalyzed at the University of Colorado for area (cm2) and density (in Hounsfield Units, HU) of both truncal fat and muscles using a semiautomatic segmentation software based on an IDL platform (Exelis Visual Information Solutions, Boulder CO). The semiautomatic segmentation software takes advantage of the unique CT attenuation (HU) that each soft tissue gives off, and this segmentation technique has been used in similar studies.Citation9,Citation15,Citation16 Adipose tissues of interest were: 1) SAT (area and density), 2) VAT (area and density, and 3) intermuscular fat of the skeletal muscle groups (area). Skeletal muscle density, in addition to intermuscular fat, focused on the 1) rectus abdominis, 2) lateralis (consisting of external and internal oblique, and transversus abdominis), 3) psoas major, and 4) paraspinal muscles (erector spinae and transversospinalis). Any muscle component with a density greater than or equal to 30 HU was defined as lean muscle.Citation17 Intermuscular fat area was then defined as the total muscle area minus the lean muscle area. Goodpaster et al. have demonstrated that CT muscle attenuation correlates with muscle fat content obtained by muscle biopsy.Citation15 All analyses focused on a single CT slice between the third and fourth lumbar vertebrae.
Statistical analysis
SAS version 9.4 (SAS Institute, Cary, NC, USA) was used to conduct all statistical analyses in this study. Data distribution was examined for each continuous variable and no data transformation was conducted. Data errors (e.g. unreasonable values) were identified and verified with clinicians before they were removed from subsequent analyses. No outliers were detected. The significance level was set at .05.
The baseline characteristics were summarized and compared between the three treatment groups (ATV/r, DRV/r and RAL) using a χ2 test (). To address the issues of imbalanced and temporally irregular data when estimating the postintervention effects, the linear segmented regression autoregressive error model was employed.Citation18,Citation19 This approach utilized all available data, and did not require the same number of measurements from each subject or evenly spaced observations in time domain. The within-person correlation arising from longitudinal nature of data was accounted for by assuming the first-order Markov chain property. For the purpose of appropriate statistical modeling, Day 0 was defined as the start time of therapy, and all the measurements taken before Day 0 were included in our analysis to estimate the before-treatment baseline. We further assumed that for this population, the amount of body fat changed at a fixed rate over time for each individual during the study period. Relevant factors (e.g. sex, HIV duration, BMI) were examined and included in the model through stepwise selection with the inclusion and exclusion p value thresholds of .2 and .05, respectively. Due to differences in expected performance by sex, univariate linear models were constructed for separately for men and women for physical function analyses. Similar analyses were conducted to explore the relationship between physical functions and treatment groups.
Table 1 Baseline characteristics
Results
Study population
Of the 295 participants with available CT images, 106 participants were receiving ATV/r, 48 DRV/r, and 141 RAL. The baseline characteristics of each group are shown in . The DRV/r group tended to be older and had a longer duration of HIV compared to the ATV/r or RAL groups. The median BMI, SAT area, and VAT area were also greater in the DRV/r group as compared to the ATV/r or RAL groups.
Effects of ART on fat density and area
In multivariate models adjusting for age, sex, BMI, and/or duration of ART (), compared to ATV/r, DRV/r use was independently associated with significantly greater VAT (26.2 cm2) and a trend towards greater SAT area (18.7 cm2), after adjusting for duration of ART, BMI, and sex. Compared to ATV/r, there was no significant effect of RAL on SAT or VAT area. Although DRV/r tended to have a lower SAT density among all participants (–2.46 HU), differences did not reach statistical significance (p = .054). Associations with age, sex, BMI, and duration of ART are shown in .
Table 2 Multivariate model results for subcutaneous and visceral fat area and density (reference group ATV/r)
Effects of ART on muscle density and intermuscular fat area
Compared to ATV/r, DRV/r use was independently associated with lower lateralis muscle density (–2.43 HU) and greater lateralis intermuscular fat area (1.64 cm2; and , respectively), after adjusting for age, sex, and BMI. Compared to ATV/r, RAL was independently associated with less psoas intermuscular fat area (–0.48 cm2). Associations with age, sex, BMI, and duration of HIV or ART are shown in and Citation4.
Table 3 Multivariate model results for skeletal muscle density (reference group = ATV/r)
Table 4 Multivariate model results for intermuscular fat area (reference group = ATV/r)
Correlations between fat, muscle, and ART with measures of physical function
Among men, greater psoas and paraspinal muscle density (less fat) correlated with stronger grip strength and higher (better) SPPB performance (). No associations between ART group and physical function were seen among men. Among females, greater paraspinal muscle density and less intermuscular paraspinal fat were associated with higher SPPB performance, and DRV/r use was associated with stronger grip strength.
Table 5 Correlations between measures of physical function and fat and muscle density or area by sex
Discussion
In contrast to our hypothesis, we did not find that RAL was associated with consistently higher quality muscle and fat, or with better physical function compared to ATV/r. Although fewer participants were prescribed DRV/r compared to ATV/r and RAL, we found that DRV/r use was generally associated with greater fat area and lower density of both fat and muscle. We did find that RAL was associated with less intermuscular psoas fat, and, consistent with our second hypothesis, higher density psoas and paraspinal musculature were associated with better physical function (grip strength and SPPB), suggesting potential clinical relevance of these findings.
The protease inhibitors, including DRV/r, have long been implicated in fat accumulation,Citation20,Citation21 however, A5260s, a blinded, randomized controlled trial of tenofovir disoproxil fumarate and emtricitabine plus DRV/r, ATV/r, or RAL, observed similar gains in VAT, SAT, and DXA-measured total body fat in all three arms.Citation5 In the larger parent study, A5257, which included only body weight, BMI, and waist circumference measurements, RAL was associated with greater gain in waist circumference than DRV/r in the overall cohort, and women experienced greater increases in waist circumference on RAL compared to ATV/r.Citation22 Initiation of PI versus INSTI or switch from PI to INSTI have shown variable results regarding gain in weight, total fat, or fat depots with either PI or INSTIs.Citation23–26 As our DRV/r participants tended to have a longer duration of HIV, we may have also detected the legacy effects of exposure to older, more toxic ART rather than effects of DRV/r itself.
Relatively little is known about how contemporary ART regimens impact muscle function. Prior cohort studies have observed a greater frequency of myalgias or weakness (by self-report) and laboratory evidence of muscle toxicity (by CK elevation) among some participants taking raltegravir compared to other ART regimens.Citation27,Citation28 In contrast, in an observational ACTG study of PWH aged 40 years or older on ART, persons prescribed RAL had a significantly lower odds of having slow gait speed, suggesting a potentially protective effect of RAL.Citation29 Other studies in PWH have found an association between efavirenz use and greater frailty,Citation29 a phenotype incorporating functional measurements of grip strength and gait speed. Efavirenz has also been associated with increased risk for falling, while we have previously shown that protease inhibitors were associated with lower risk of falls.Citation30
What significance might the density of these specific muscle groups have on physical function? Here, the density of the psoas and the paraspinal muscles were associated with better physical function. The rectus abdominis and lateralis muscle groups create the anterolateral abdominal wall, while psoas major and paraspinal muscle groups create the posterior abdominal wall. The psoas major plays an important role in flexing the thigh, and the paraspinal muscle group acts as the trunk extensor. Working together as a system, these trunk muscles have a key role in spinal stability and fall prevention.Citation31,Citation32 Thus, not surprisingly, low muscle density has been associated poorer physical function in multiple cohorts.Citation10,Citation33 Anderson et al. found both men and women with a higher muscle density (less fat) to have reduced postural sway, and among women only, a higher SPPB score.Citation11
Our study population was small (N = 295) compared to other large cohorts, and study arms were not randomized. In the present study, provider prescribing or participant preference may have influenced the use of DRV/r, and these unmeasured factors may not be completely accounted for in our adjusted analyses. However, the proportion of female participants (N = 93), CT scans over multiple time points, and physical function data are strengths of our analysis. In addition, no individuals received tenofovir alafenamide during the study period, thus differences between arms were largely driven by the third agent rather than differences in the backbone.
In summary, we found that DRV/r use was associated with greater fat area and lower density of both fat and muscle, and RAL with less intermuscular psoas fat. Although no ART arms correlated with impairment in grip or SPPB performance, higher density psoas and paraspinal musculature were associated with better physical function (grip strength and SPPB), suggesting potential clinical relevance of these findings. Further studies and larger cohorts are needed to understand the longer-term/longitudinal implications of changes in muscle fat and density due to ART in PWH populations.
Disclosure statement
JEL has received research grants and consulting fees from Gilead Sciences. KME has received research grants (to the institution) from Gilead Sciences and consulting fees (to the institution) from ViiV and Gilead Sciences. GG received research funding from Merck Sharp and Dohme Corp for this work.
Additional information
Funding
Notes on contributors
Stefan Adrian
Stefan Adrian, MS, is a research assistant at the School of Medicine, University of Colorado Anschutz Medical Campus, Aurora, CO, USA. He has extensive training in anatomy and imaging techniques.
Hongyu Miao
Hongyu Miao, PhD, is an Associate Professor at School of Public Health, University of Texas Health Science Center at Houston, Houston, TX, USA. He is an experienced statistician and data scientist exploring both infectious diseases and neural disorder diseases.
Han Feng
Han Feng is a PhD student at Biostatistics, University of Texas Health Science Center at Houston, USA. His interests are in high dimensional data analysis, network data analysis, and electronic health records.
Ann Scherzinger
Ann Scherzinger, PhD, is a Professor of Radiology- Radiologic Sciences at the University of Colorado Anschutz Medical Campus, Aurora, CO, USA. Her interests are in imaging techniques for body composition assessments.
Giulia Nardini
Giulia Nardini, MBiol, is responsible for the coordination of the research projects, patients' recruitment, characterization and research laboratory activities at Modena HIV Metabolic Clinic at University of Modena and Reggio Emilia, Modena, Italy.
Barbara Beghetto
Barbara Beghetto, MBiol, is responsible for the coordination of the research projects, patients' recruitment, characterization and research laboratory activities at Modena HIV Metabolic Clinic at University of Modena and Reggio Emilia, Modena, Italy.
Enrica Roncaglia
Enrica Roncaglia, MBiol, is responsible for the coordination of the research projects, patients' recruitment, characterization and research laboratory activities at Modena HIV Metabolic Clinic at University of Modena and Reggio Emilia, Modena, Italy.
Guido Ligabue
Prof. Guido Ligabue, MD, is a radiologist and associate professor at University of Modena and Reggio Emilia. Prof. Ligabue has been collaborating with Modena HIV Metabolic Clinic for years. In particular, he evaluates thoracic and abdominal CT scans and has long-term experience in this research field.
Jovana Milic
Jovana Milic, MD, is a junior research fellow and PhD student at the Modena HIV Metabolic Clinic, University of Modena and Reggio Emilia, Italy. Her particular focus is the non-invasive diagnostics of non-alcoholic fatty liver disease and other metabolic alterations in people living with HIV.
Giovanni Guaraldi
Prof. Giovanni Guaraldi, MD, is an associate professor of infectious diseases at University of Modena and Reggio Emilia, Italy. Since 2002, he has been running Modena HIV Metabolic Clinic (MHMC). This tertiary level referral center offers a multidisciplinary team consultant service consisting of infectious disease physicians, nutritionists, personal trainers for physical activities, psychologists, cardiologist, nephrologists, endocrinologists and plastic surgeons for diagnosis and treatment of non-infectious co-morbidities. More than 4000 people are in charge at MHMC. He is a panel member of the European AIDS Clinical Society (EACS) guidelines on prevention and management of HIV associated co-morbidities. He has published widely on clinical aspects of HIV treatment and care.
Jordan E. Lake
Jordan E. Lake, MD, is an Associate Professor of Medicine at University of Texas Health Science Center at Houston, Houston, TX, USA. Her research and clinical interests are on adipose tissue in adults with HIV.
Kristine M. Erlandson
Kristine M. Erlandson, MD, is an Associate Professor of Medicine at the University of Colorado Anschutz Medical Campus, Aurora, CO, USA. Her research and clinical interests are on the older adults with HIV.
References
- Kumar S, Samaras K. The impact of weight gain during HIV treatment on risk of pre-diabetes, diabetes mellitus, cardiovascular disease, and mortality. Front Endocrinol (Lausanne). 2018; 9:705.
- Mankal PK, Kotler DP. From wasting to obesity, changes in nutritional concerns in HIV/AIDS. Endocrinol Metab Clin North Am. 2014; 43(3):647–663.
- Bogner JR, Vielhauer V, Beckmann RA, et al. Stavudine versus zidovudine and the development of lipodystrophy. J Acquir Immune Defic Syndr. 2001; 27(3):237–244.
- Falutz J. Therapy insight: body-shape changes and metabolic complications associated with HIV and highly active antiretroviral therapy. Nat Clin Pract Endocrinol Metab. 2007; 3(9):651–661.
- McComsey GA, Moser C, Currier J, et al. Body composition changes after initiation of raltegravir or protease inhibitors: ACTG A5260s. Clin Infect Dis. 2016; 62(7):853–862.
- Erlandson KM, Fiorillo S, Masawi F, et al. Antiretroviral initiation is associated with increased skeletal muscle area and fat content. Aids 2017; 31(13):1831–1838.
- Mallon PW, Miller J, Cooper DA, Carr A. Prospective evaluation of the effects of antiretroviral therapy on body composition in HIV-1-infected men starting therapy. Aids 2003; 17(7):971–979.
- Debroy P, Lake JE, Moser C, et al. Antiretroviral therapy initiation is associated with decreased visceral and subcutaneous adipose tissue density in people living with HIV. Clin Infect Dis [published ahead of print]. 2020. DOI:10.1093/cid/ciaa196.
- Goodpaster BH, Carlson CL, Visser M, et al. Attenuation of skeletal muscle and strength in the elderly: The Health ABC Study. J Appl Physiol (Bethesda, MD 1985). 2001; 90(6):2157–2165.
- Scott D, Shore-Lorenti C, McMillan LB, et al. Calf muscle density is independently associated with physical function in overweight and obese older adults. J Musculoskelet Neuronal Interact. 2018; 18(1):9–17.
- Anderson DE, Quinn E, Parker E, et al. Associations of computed tomography-based trunk muscle size and density with balance and falls in older adults. J Gerontol A Biol Sci Med Sci. 2016; 71(6):811–816.
- Lake JE, Debroy P, Ng D, et al. Associations between subcutaneous fat density and systemic inflammation differ by HIV serostatus and are independent of fat quantity. Eur J Endocrinol/Eur Federation Endocr Soc. 2019; 181(4):451–459.
- Guaraldi G, Orlando G, Squillace N, et al. Multidisciplinary approach to the treatment of metabolic and morphologic alterations of HIV-related lipodystrophy. HIV Clin Trials 2006; 7(3):97–106.
- Guralnik JM, Simonsick EM, Ferrucci L, et al. A short physical performance battery assessing lower extremity function: association with self-reported disability and prediction of mortality and nursing home admission. J Gerontol. 1994; 49(2):M85–M94.
- Goodpaster BH, Kelley DE, Thaete FL, He J, Ross R. Skeletal muscle attenuation determined by computed tomography is associated with skeletal muscle lipid content. J Appl Physiol (Bethesda, MD 1985). 2000; 89(1):104–110.
- Schafer AL, Vittinghoff E, Lang TF, et al. Fat infiltration of muscle, diabetes, and clinical fracture risk in older adults. J Clin Endocrinol Metab. 2010; 95(11):E368–E372.
- Goodpaster BH, Thaete FL, Kelley DE. Composition of skeletal muscle evaluated with computed tomography. Ann NY Acad Sci. 2000; 904:18–24.
- Zhang F, Wagner AK, Soumerai SB, Ross-Degnan D. Methods for estimating confidence intervals in interrupted time series analyses of health interventions. J Clin Epidemiol. 2009; 62(2):143–148.
- Wagner AK, Soumerai SB, Zhang F, Ross-Degnan D. Segmented regression analysis of interrupted time series studies in medication use research. J Clin Pharm Ther. 2002; 27(4):299–309.
- Srinivasa S, Grinspoon SK. Metabolic and body composition effects of newer antiretrovirals in HIV-infected patients. Eur J Endocrinol. 2014; 170(5):R185–R202.
- Safrin S, Grunfeld C. Fat distribution and metabolic changes in patients with HIV infection. Aids 1999; 13(18):2493–2505.
- Bhagwat P, Ofotokun I, McComsey GA, et al. Changes in waist circumference in HIV-infected individuals initiating a raltegravir or protease inhibitor regimen: effects of sex and race. Open Forum Infect Dis. 2018; 5(11):ofy201.
- Lake JE, McComsey GA, Hulgan T, et al. Switch to raltegravir from protease inhibitor or nonnucleoside reverse-transcriptase inhibitor does not reduce visceral fat in human immunodeficiency virus-infected women with central adiposity. Open Forum Infect Dis. 2015; 2(2):ofv059.
- Sax PE, Erlandson KM, Lake JE, et al. Weight gain following initiation of antiretroviral therapy: risk factors in randomized comparative clinical trials. Clin Infect Dis [published ahead of print]. 2019. DOI:10.1093/cid/ciz999.
- Norwood J, Turner M, Bofill C, et al. Brief report: weight gain in persons with HIV switched from efavirenz-based to integrase strand transfer inhibitor-based regimens. J Acquir Immune Defic Syndr. 2017; 76(5):527–531.
- Kerchberger AM, Sheth AN, Angert CD, et al. Weight gain associated with integrase stand transfer inhibitor use in women. Clin Infect Dis. 2020; 71(3):593–600.
- Calza L, Danese I, Colangeli V, et al. Skeletal muscle toxicity in HIV-1-infected patients treated with a raltegravir-containing antiretroviral therapy: a cohort study. AIDS Res Hum Retroviruses 2014; 30(12):1162–1169.
- Lee FJ, Amin J, Bloch M, Pett SL, Marriott D, Carr A. Skeletal muscle toxicity associated with raltegravir-based combination antiretroviral therapy in HIV-infected adults. J Acquir Immune Defic Syndr. 2013; 62(5):525–533.
- Erlandson KM, Wu K, Koletar SL, et al. Association between frailty and components of the frailty phenotype with modifiable risk factors and antiretroviral therapy. J Infect Dis. 2017; 215(6):933–937.
- Erlandson KM, Plankey MW, Springer G, et al. Fall frequency and associated factors among men and women with or at risk for HIV infection. HIV Med. 2016; 17(10):740–748.
- Stokes IA, Gardner-Morse MG, Henry SM. Abdominal muscle activation increases lumbar spinal stability: analysis of contributions of different muscle groups. Clin Biomech (Bristol, Avon). 2011; 26(8):797–803.
- Panjabi MM. The stabilizing system of the spine. Part I. Function, dysfunction, adaptation, and enhancement. J Spinal Disord. 1992; 5(4):383–389; discussion 97.
- Cleary LC, Crofford LJ, Long D, et al. Does computed tomography-based muscle density predict muscle function and health-related quality of life in patients with idiopathic inflammatory myopathies? Arthritis Care Res. 2015; 67(7):1031–1040.