- Email: [email protected]
Increased risk of dysglycaemia in South Africans with HIV; especially those on protease inhibitors
Accepted Manuscript Increased risk of dysglycaemia in south africans with HIV; Especially those on protease inhibitors Naomi S Levitt, Nasheeta Peer, Krisela Steyn, Carl Lombard, Gary Maartens, Estelle V. Lambert, Joel A. Dave PII: DOI: Reference:
S0168-8227(16)30037-7 http://dx.doi.org/10.1016/j.diabres.2016.03.012 DIAB 6591
To appear in:
Diabetes Research and Clinical Practice
Received Date: Revised Date: Accepted Date:
29 April 2015 29 January 2016 19 March 2016
Please cite this article as: N.S. Levitt, N. Peer, K. Steyn, C. Lombard, G. Maartens, E.V. Lambert, J.A. Dave, Increased risk of dysglycaemia in south africans with HIV; Especially those on protease inhibitors, Diabetes Research and Clinical Practice (2016), doi: http://dx.doi.org/10.1016/j.diabres.2016.03.012
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
INCREASED RISK OF DYSGLYCAEMIA IN SOUTH AFRICANS WITH HIV; ESPECIALLY THOSE ON PROTEASE INHIBITORS Short running title: Dysglycaemia in South Africans with HIV Authors and affiliations: 1,2
Naomi S Levitt,
3,4
Nasheeta Peer, 2Krisela Steyn, 5Carl Lombard, 6Gary Maartens, 7Estelle
V Lambert, 1Joel A Dave 1. Division of Diabetic Medicine and Endocrinology, Department of Medicine, Faculty of Health Sciences, University of Cape Town (UCT), South Africa 2. Chronic Disease Initiative for Africa, Department of Medicine, Faculty of Health Sciences, UCT, South Africa 3. Non-communicable Diseases Research Unit, South African Medical Research Council (SAMRC), South Africa 4. Department of Medicine, Faculty of Health Sciences, UCT, South Africa 5. Biostatistics Unit, SAMRC, South Africa 6. Division of Clinical Pharmacology, Department of Medicine, Faculty of Health Sciences, UCT, South Africa 7. SAMRC/UCT Research Unit for Exercise Science and Sports Medicine, Department of Human Biology, Faculty of Health Sciences, UCT, South Africa
Corresponding author: Professor Naomi Levitt Address: Divisions of Diabetic Medicine and Endocrinology, Department of Medicine, University of Cape Town, Private Bag X3, Observatory 7925, South Africa Telephone: +27 21 404 5007, Fax: +27 21 406 6513 Email: [email protected]
Word count – abstract: 241 Word count – text: 2716 References: 30
ABSTRACT Aims To compare dysglycaemia prevalence (impaired fasting glucose (IFG), impaired glucose tolerance (IGT), or diabetes) in HIV-infected persons, stratified by antiretroviral therapy (ART), with a community-based survey (CBS) in Cape Town, South Africa. Methods Three groups of HIV-infected adults without known diabetes were conveniently sampled from community healthcare centres; ART-naïve, first-line ART (non-nucleoside reverse transcriptase inhibitor (NNRTI) plus dual NRTIs), and second-line ART (lopinavir/ritonavirboosted protease inhibitor plus dual NRTIs). The CBS recruited a representative crosssectional sample from urban townships. Participants reporting ART use or known diabetes were excluded. All participants underwent oral glucose tolerance testing. Multiple logistic regression determined independent associations with dysglycaemia. Results The samples comprised ART-naïve, first-line ART, second-line ART and CBS participants (n=393, 439, 108 and 880, respectively). Mean age was 34-40 years. Dysglycaemia prevalence was as follows: CBS 18.0%, ART-naïve 21.6%, first-line ART 26.0% and secondline ART 37.0%. Diabetes was similar across groups, but IGT was 3-4-fold higher in secondline ART and CBS compared with ART-naïve and first-line ART groups. In contrast, IFG was 14.3-21.2% across HIV groups but only 1.5% in the CBS. Increased risk of dysglycaemia was associated with older age, female gender, and HIV status (ART-naïve: OR 2.31, 95%CI 1.65-3.24; first-line ART: OR 2.47, 95%CI 1.80-3.38; second-line ART: OR 4.10, 95%CI 2.54-6.61). Diabetes family history and central obesity were not related to dysglycaemia. Conclusions In view of the increased risk of dysglycaemia in HIV-infected participants, screening for diabetes should be instituted in ART programmes.
Key words: South Africa, black, HIV, dysglycaemia, diabetes, ART
1. INTRODUCTION There is growing concern about the colliding epidemics of HIV/AIDS and non-communicable diseases (NCD) such as diabetes, particularly in sub-Saharan Africa (SSA)
1-3
. The region
has the largest number of people with HIV/AIDS globally and improved access to antiretroviral therapy (ART) has seen a substantial improvement in life expectancy in these patients
4, 5
. Simultaneously, increasing levels of urbanisation, ageing of the population and
the adoption of western lifestyles, with changes in eating patterns and rising levels of physical inactivity, are driving the NCD epidemic
2, 6, 7
.
With the ageing of the HIV population due to the reduction in premature mortality from ART, there is an increased risk of developing diabetes and other NCDs 8. While numerous reports describe an increased risk of diabetes in people receiving ART
9-11
, there are sparse data
from SSA where the majority of people on ART are still on first-line non-nucleoside reverse transcriptase inhibitor (NNRTI) regimens, and are predominantly younger women
12, 13
. By
contrast, in high income countries, protease inhibitor (PI)-based regimens are often used in first-line ART, and the majority of patients are men. We recently reported a high prevalence of dysglycaemia in a cross sectional study of HIV-infected participants from Cape Town (25.7% in ART-naive and 21.9% in those on first-line ART, especially efavirenz)
14
However,
newly diagnosed diabetes was uncommon. In this paper we extend our previous findings by: i) investigating the prevalence of dysglycaemia (defined as one or more of impaired fasting glucose (IFG), impaired glucose tolerance (IGT) or diabetes) in an additional sample of participants receiving second-line ART and ii) comparing this to HIV-infected cohorts who are ART-naïve and on first-line ART; as well as participants of a cross-sectional diabetes epidemiological study conducted concurrently in the same geographic area of Cape Town 15.
2. MATERIAL AND METHODS 2.1. Study populations 2.1.1. HIV groups: HIV-infected adults aged ≥18 years were categorized into three groups: ART-naïve, first-line ART (stavudine (d4T), lamivudine (3TC), and efavirenz or nevirapine), and second-line ART (zidovudine (AZT), didanosine (ddI) and lopinavir-ritonavir (LPV-r)). Participants were conveniently sampled consecutively from a community healthcare centre in Cape Town 2008-2010. Second-line ART participants were recruited from a number of community healthcare centres in the same area, because there were smaller numbers on second-line at
the time. Participants were excluded if they had: 1) been on ART for less than six months, 2) a history of diabetes mellitus or IGT, 3) had an active, acute opportunistic infection, 4) severe diarrhea (six stools/day), 5) tuberculosis within one month of commencing treatment, 6) received glucocorticoid therapy within the past six months, or were 7) pregnant or 8) known to have renal failure. 2.1.2. Community-based survey: Residents living in the predominantly black African areas of Langa, Guguletu, Crossroads, Nyanga and Khayelitsha in Cape Town who were 25-74 years old were the target population. Based on an estimated diabetes prevalence of 8% with a precision of 1.5% twosided with 95% confidence, a sample size of 1000 was planned. A 3-stage cluster sampling stratified by area and housing type was undertaken in 2008/09. Individuals from households were selected using quotas for pre-specified age and gender categories. The exclusion criteria were: unable to give consent, on tuberculosis treatment, known to be on ART, treatment for cancer within the last year, bedridden, pregnant or lactating, or resident in Cape Town for less than 3 months. Replacements were allowed when confronted with individuals who met the exclusion criteria above, those who refused, or the randomly selected participant of the randomly selected household could not be contacted on the third attempt. For the purpose of this analysis, participants aged 25-60 years without a known diagnosis of diabetes were included. The decreased age criteria was to reduce the potential impact of an imbalance in the older ages between the CBS and the HIV groups, which had a small number of participants over the age of 60 years. 2.2. Data collection In both surveys fieldworkers administered questionnaires to obtain socio-demographic, selfreported medical and family history, tobacco (WHO STEP-wise surveillance questionnaire) 16 . Height, weight, and waist and hip circumferences were measured using standardised techniques. Blood samples, for glucose estimations, were drawn following an overnight fast of 10 hours. Thereafter, a standard oral glucose tolerance test (OGTT), using 75 grams of anhydrous glucose in 250ml of water, was administered, and blood samples taken 120 minutes later for glucose estimation
17
. Blood samples were kept on ice and transported to the laboratory
within six hours to be centrifuged and aliquoted. 2.3. Definitions
Diabetes, IGT and IFG were diagnosed according to the 1998 WHO definition 17. Body mass index (BMI) was calculated as the participant’s weight in kilograms divided by their height in metres squared (kg/m²). Obesity (≥30 kg/m²) and increased waist circumference (>94 cm in men and >80 cm and in women) were defined using standardised international criteria 18. 2.4. Data Analyses Data analyses were done using STATA 11 and SAS Version 9.2.3. The analysis of the combined data assumed that all groups were obtained through simple random sampling for inference purposes. Descriptive statistics, including crude prevalence and 95% confidence intervals (CI), were calculated. Univariate analyses (sociodemographic and anthropometric) are presented as mean values and standard deviations for continuous data, and as percentages for categorical data. Standardised prevalence for dysglyceamia and its components were calculated using the age and sex distribution of the 2009 South African Black population. In a multiple logistic regression model the outcome new onset dysglycaemia (IFG, IGT and diabetes combined) was modelled on age, BMI, sex, housing type, waist and hip circumferences, study group and family history of diabetes. The odds ratios (OR) and 95% CI are reported as the estimated risk measures. The University of Cape Town’s Research and Ethics Committee approved both studies. All participants signed informed consent.
3. RESULTS As seen in Table 1 the participants from the community-based sample were older than the HIV groups with a greater proportion of men, but women were in the majority in all groups. The highest proportion of participants living in built formal housing were those on second-line ART (39.8%) compared to the other groups (16.3-19.9%). A family history of diabetes was most common in the CBS (21.2%) compared to the other groups (15.6-16.4%). The median duration on first-line and second-line ART was 16 and 18 months, respectively. The mean BMI (Table 2) fell into the overweight range in all groups. The proportions of participants with obesity and increased waist circumference increased progressively from the ART-naive to the first-line ART, second-line ART and the CBS groups. Notably a much greater proportion of women than men had an increased waist circumference in each group. In contrast, the prevalence of dysglycaemia was lowest in the CBS group (18.0%), increased progressively in the ART-naïve (21.6%), the first-line ART (26.0%), and the second-line ART group (37.0%) (Table 3). The prevalence of diabetes was similar across the four groups. The prevalence of IGT was 3-4-fold higher in the second-line ART and CBS groups
compared to the ART-naïve and first-line ART groups. On the other hand, IFG was rare in the CBS, and the prevalence ranged between 14.3% and 21.2% across the HIV groups. In the multiple logistic regression analysis (Table 4) increasing age, female gender and HIV status were significantly associated with an increased risk of dysglycaemia. Compared to the community-based sample, the likelihood for dysglycaemia was increased (p<0.001) in the HIV infected participants and similar for ART-naïve (OR 2.31, 95% CI 1.653.24) and first-line ART (OR 2.47, 95% CI 1.80-3.38) groups while the prevalence for second-line ART was higher (OR 4.10, 95% CI 2.54-6.61). Waist circumference, quality of housing and a family history of diabetes were not associated with dysglycaemia. When the duration on ART was examined by splitting the two ART groups by their respective median times on ART, the risk for dysglycaemia was similar, when adjusted for the other covariates in the model (data not shown).
4. DISCUSSION The major finding of our study was the marked increase in the risk for dysglycaemia in HIVinfected participants compared to the community-based sample, particularly in the secondline ART group. It is of concern that the prevalence of dysglycaemia was 2-4-fold higher in HIV-infected groups compared to the CBS group, as diabetes in the community-based sample was already 1.5-fold greater than the prevalence found in this urban black population almost two decades ago 15. The increase in diabetes prevalence in the community-based sample was attributed to traditional risk factors such as obesity and urbanisation, however, the factors contributing to the higher dysglycaemia prevalence in HIV-infected individuals are likely to be more complex. Defronzo et al (2011) have suggested that the pathophysiology underlying IFG is moderate hepatic insulin resistance and impaired early but intact late insulin response19. In line with this, we previously reported that the HIV infected dysglycaemic subjects who were either ART naïve or receiving first line ART were more insulin resistant as measured by HOMA-IR and had lower beta-cell function (insulinogenic index and disposition index) than their normoglycaemic counterparts12.
Indeed the presence of HIV infection itself may
contribute to metabolic changes, both directly through immune activation and inflammation, and indirectly through immunodeficiency
10, 20, 21
. Increased pro-inflammatory markers such
as C-reactive protein (CRP), tumor necrosis factor-a (TNF-a), and interleukin-6 together with low levels of adiponectin characterise insulin resistance and type 2 diabetes in HIV infection 22-24
.
Furthermore, since there was a greater risk for dysglycaemia in participants on second-line ART compared to ART-naïve and first-line ART participants there appears to be an association between the type of ART regimen and dysglycaemia. The ART attributable risk for dysglycaemia may be through the effect of the drugs on insulin resistance and body fat distribution but also possibly through other pathways that are currently unclear 10.. Lopinavir-ritonavir, given to most participants on second-line ART, has been implicated in the development of insulin resistance and alterations in glucose metabolism in some but not all studies, and may possibly explain the higher risk for dysglycaemia associated with its use in our study
25, 26
. All HIV-infected individuals are at an increased risk for dysglycaemia,
however, participants on second-line ART appear to be more vulnerable and possibly require closer monitoring for metabolic complications. These findings highlight that the selection of ART regimen in HIV-infected participants should consider both the virological effectiveness and the metabolic effects of the drugs, particularly as antiretroviral drugs with more neutral metabolic profiles are available
27
. The
higher prevalence of dysglycaemia found in the HIV-infected participants, irrespective of ART status, compared to the community-based participants demonstrates the need for preventive screening for diabetes and behavioural counseling to promote lifestyle changes. Additionally, novel therapies should be considered to prevent or treat metabolic abnormalities in the HIV-infected patient such as the use of anti-inflammatory and immunemodulatory drugs 27. Nevertheless, despite the similarities in mean BMI between participants on both first- and second-line ART, the prevalence of obesity was much higher in those on second-line ART and may have, in addition to the diabetagenic effect of the ART, contributed to their greater dysglycaemia. Overweight/obesity is recognised as the environmental risk factor most strongly associated with the development of diabetes,
28
and obese HIV-infected individuals
should expect to accumulate the same metabolic abnormalities as the general population
21,
27
.
However, the relationship between traditional risk factors for diabetes, and HIV- and ARTinduced metabolic changes is complex and the latter may accelerate diabetogenesis in HIV infection in the absence of obesity 27. This may account for the finding that even though 80% of diabetic individuals are usually overweight/obese reported community-based study
29
, as was the case in our previously
15
, no significant associations between BMI, waist or hip
circumferences with dysglycaemia were present in the current study. A case-control study in Taiwan also did not show a significant relation between BMI and incident diabetes in HIVinfected patients 30.
HIV-specific influences may predispose to diabetes at lower levels of adiposity in the HIVinfected compared to the HIV-uninfected
27
population
. These effects are partly due to
inflammation and insulin resistance induced by lipodystrophy, co-morbidities and ART. Other risk factors may include high pre-ART viral loads and low baseline CD4 counts, which possibly increase the risk of insulin resistance and escalate the development of dysglycaemia independent of clinical adiposity 20. Nonetheless, traditional risk factors play an important role in the development of diabetes among HIV-infected cohorts
28
and were likely also influential in the progression to
dysglycaemia in this study. Age, a consistent predisposing risk factor for incident diabetes in HIV cohort studies
28
, was found to be significantly related to dysglycaemia. This suggests
that with increased longevity among HIV-infected patients on ART, a rise in dysglycaemia independent of HIV-related influences may occur with ageing of this population. The successful rollout of the ART campaign in South Africa has prolonged the life expectancy of HIV-infected patients such that diabetes and other cardiovascular diseases risk factors are likely to become a rising health burden. The absence of a significant association between a family history of diabetes and dysglycaemia may be related to a lack of knowledge about the disease rather than a true reflection of a negative family history. This suggestion is supported by a much lower prevalence of diabetes family history among unknown (22.8%) compared to known (46.2%) diabetic participants in our community-based study
15
. A family history of diabetes in the
latter study was significantly associated with diabetes in women (among whom unknown diabetes was 33.9%) but not in men (51.8% unknown diabetes). Nonetheless, further investigation is required to support this notion as not all HIV studies have shown a relation between family history and diabetes 28. The increased likelihood of dysglycaemia in women compared to men contrasts with the predominance of diabetes in the latter among HIV-infected individuals 20, 28. Mitigating factors may include ethnicity with few studies conducted in black populations to confirm the gender distribution or differences in HIV-related risk parameters. Our study had several limitations. First, the HIV status of the community participants was not known, although participants receiving ART were excluded. However, the HIV-infected proportion of the CBS, which we estimate to be about 10%, would have diluted any effect of HIV on dysglycaemia. Therefore the true risk of dysglycaemia in HIV-infected individuals is likely to be higher than we calculated. Second, we are unable to comment on the total diabetes prevalence in HIV-infected participants because those with known diabetes were excluded from the HIV study. Third, the small number of participants on second-line ART
was a consequence of the limited number of people receiving this therapy at the time of the study. Fourth, there were methodological differences in sampling design of the study groups. Also, considering the cross-sectional design of these studies, causality cannot be established. To facilitate any inference a basic sampling design were assumed p-values and confidence intervals must therefore be considered investigative. The major strength of the study is the use of an OGTT which has permitted an assessment of the prevalence of IGT in addition to IFG and the diagnosis of diabetes based on both fasting and 2-hour post OGTT measures. The findings of this study may be generalizable to other urban South African black populations.
5. CONCLUSION The increased likelihood of dysglycaemia with HIV infection highlights the need for concurrent management of cardio-metabolic abnormalities in the HIV-infected population. The rise in longevity on ART will inevitably be accompanied by a rise in diabetes-associated morbidity and mortality, with a greater burden imposed on the South African healthcare system. Although this study has demonstrated a higher prevalence of dysglycaemia in HIV-infected patients, especially those on second-line ART, the pathways contributing to the development and the differentiation of these diseases is currently poorly elucidated
27
. Further research is
needed to understand the pathophysiology of dysglycaemia in HIV-infected patients so as to optimise prevention and management strategies.
ACKNOWLEDGEMENTS: We would like to acknowledge the contributions of the study co-ordinators Carmen Delport and Serena van Haght and their respective teams of fieldworkers .GM acknowledges research support in part by the National Research Foundation (NRF) of South Africa (grant specific unique reference number 85810). The grant holder acknowledges that opinions, findings and conclusions or recommendations expressed in any publication generated by the NRF supported research are that of the author, and that the NRF accepts no liability whatsoever in this regard. Funding The Community-based survey was supported by an unrestricted grant from Servier Laboratories (South Africa); the South African Medical Research Council; the Initiative for Cardiovascular Health Research in Developing Countries (IC Health) Foundation Council;
and Brigham and Women’s Hospital, Harvard University. The study in the HIV cohort was funded by grants from the World Diabetes Foundation and the South African Department of Health. Conflict of interest Naomi Levitt has received travel support from Novo Nordisk, Eli Lilly Laboratories and Sanofi Aventis. All other authors report no potential conflicts of interest, including specific financial interests, relevant to the subject of this manuscript. Contributors
Table 1: Characteristics of the samples studied Communitybased
HIV Group
Survey st
ART naïve
1 line ART
2
nd
line ART
Number
880
393
439
108
Age in years, mean (SD)
39.9 (10.08)
33.6 (8.77)
36.1 (8.87)
36.6 (8.32)
Men
37.6
23.9
22.6
15.7
Women
62.4
76.1
77.5
84.3
Family history of diabetes, %
21.2
15.6
16.4
15.7
Alcohol use, %
44.7
52.0
42.6
39.8
Formal/built
18.9
19.9
16.3
39.8
Council/core
33.64
33.9
32.3
13.9
Shack/shelter/ hostel
47.5
44.6
48.5
40.7
3 (2-4)
4 (2-5)
3(2-5)
4 (3-5)
249 (154 –
315 (220 –
479 (261 -
434)
461)
624)
-
16 (10 – 25)
18 (9.5 - 26)
Gender, %:
Housing type, %:
Housing density Median CD4 count cells/µL (interquartile range)
Median duration ART (months) Number of participants on the following ART drugs:
-
-
Zidovudine (AZT)
-
-
145
82
Lamivudine (3TC)
-
-
439
36
Stavudine (D4T)
-
-
294
61
Efavirenz (EFV)
-
-
236
7
Lopinavir/Ritonavir
-
-
-
104
ART: antiretroviral therapy
Table 2: Anthropometry by group HIV Group
CommunityGroup
based Survey
st
ART Naive
1 line ART
N=393
N=439
N=880
nd
2
line ART N=108
Body mass index (BMI) Mean (SD), kg/m
2
29.3 (8.39)
25.9 (6.17)
26.8 (5.81)
27.5 (6.09)
41.4
21.9
23.7
30.6
Men
84.1 (12.99)
80.2 (8.01)
82.5 (9.45)
83.8 (9.97)
Women
96.2 (15.2)
84.1 (13.1)
87.8 (11.8)
90.8 (13.6)
Men (WC >94 cm)
18.7
6.4
14.1
17.7
Women (WC >80 cm)
85.6
53.5
70.9
75.8
2
Obesity (BMI ≥30 kg/m ),% Waist circumference (WC) Mean (SD), cm:
Raised, %:
ART: antiretroviral therapy
Table 3a: Crude prevalence (95% CI) of dysglycaemia (diabetes, IGT and IFG) by group HIV Group
CommunityGroup
Total dysglycaemia Diabetes
st
nd
based Survey
ART Naive
1 line ART
2
N=880
N=393
N=439
N=108
18.0 (15.4-20.5)
21.6 (17.5- 25.7)
26.0 (21.9-30.1)
37.0 (27.9-46.2)
line ART
4.9 (3.5-6.3)
3.1 (1.4-4.8)
2.3 (0.9-3.7)
5.6 (1.2-9.9)
IGT
11.6 (9.5-13.7)
4.3 (2.3-6.3)
2.5 (1.0-4.0)
12.0 (5.9-18.2)
IFG
1.5 (0.6-2.3)
14.3 (10.7-17.7)
21.2 (17.4-25.0)
19.4 (11.9-26.9)
ART: antiretroviral therapy; IGT: impaired glucose tolerance; IFG: impaired fasting glycaemia
Table 3b: Standardised* prevalence of dysglycaemia (diabetes, IGT and IFG) by group HIV Group
CommunityGroup
based Survey
ART Naive
1st line ART
2nd line ART
N=880
N=393
N=439
N=108
Total dysglycaemia
15.0
26.0
29.9
34.3
Diabetes
4.1
2.7
3.0
8.8
IGT
9.4
4.5
2.1
5.7
IFG
1.5
18.9
24.8
19.9
*standardized for age and sex using the 2009 South African midyear Black population
Table 4: Logistic regression analysis for independent factors associated with new onset dysglycaemia (IFG, IGT and diabetes combined) Odds Ratio
95% Confidence Interval Lower limit
P-value
Upper limit
<0.001
Age (years) 25-34
1.00
35-44
1.82
1.29
2.57
0.001
45-54
3.27
2.25
4.75
<0.001
55-64
4.75
3.09
7.32
<0.001
2
BMI (kg/m )
0.060 <25
25-30
1.34
.89
2.01
0.160
>30
1.92
1.11
3.30
0.019
Waist circumference (cm)
1.20
0.78
1.83
0.412
Hip circumference (cm)
1.00
0.99
1.01
0.975
Sex: women
2.17
1.54
3.03
<0.001
Housing
0.406 Formal
1.00
council
0.84
0.60
1.18
0.316
Shack/shelter/hostel
1.08
0.79
1.48
0.620
<0.001
Group
Community survey
1.00
ART naive
2.31
1.65
3.24
<0.001
st
2.47
1.80
3.38
<0.001
nd
2 line ART
4.10
2.54
6.61
<0.001
Family history of diabetes
1.11
0.82
1.50
0.482
1 line ART
ART: antiretroviral therapy; IGT: impaired glucose tolerance; IFG: impaired fasting glycaemia
REFERENCES 1. Levitt NS, Steyn K, Dave J and Bradshaw D. Chronic noncommunicable diseases and HIV-AIDS on a collision course: relevance for health care delivery, particularly in lowresource settings--insights from South Africa. The American journal of clinical nutrition. 2011; 94: 1690S-6S. 2. Mbanya JC, Motala AA, Sobngwi E, Assah FK and Enoru ST. Diabetes in subSaharan Africa. Lancet. 2010; 375: 2254-66. 3. Mayosi BM, Lawn JE, van Niekerk A, Bradshaw D, Abdool Karim SS and Coovadia HM. Health in South Africa: changes and challenges since 2009. Lancet. 2012; 380: 202943. 4. Hankins C. Overview of the Current State of the Epidemic. Current HIV/AIDS Reports. 2013: 1-11. 5. Bor J, Herbst AJ, Newell ML and Barnighausen T. Increases in adult life expectancy in rural South Africa: valuing the scale-up of HIV treatment. Science. 2013; 339: 961-5.
6. Kengne AP, Echouffo-Tcheugui J-B, Sobngwi E and Mbanya J-C. New insights on diabetes mellitus and obesity in Africa–Part 1: prevalence, pathogenesis and comorbidities. Heart. 2013. 7. Mensah GA. Descriptive Epidemiology of Cardiovascular Risk Factors and Diabetes in Sub-Saharan Africa. Progress in Cardiovascular Diseases. 2013; 56: 240-50. 8. McDonald CL and Kaltman JR. Cardiovascular disease in adult and pediatric HIV/AIDS. Journal of the American College of Cardiology. 2009; 54: 1185-8. 9. Barbaro G. Metabolic and cardiovascular complications of highly active antiretroviral therapy for HIV infection. Current HIV research. 2006; 4: 79-85. 10. Martinez E, Larrousse M and Gatell JM. Cardiovascular disease and HIV infection: host, virus, or drugs? Current opinion in infectious diseases. 2009; 22: 28-34. 11. Nigatu T, Oldenburg B, Elliott J, Setswe G and Woldegiorgis MA. The incidence of cardiovascular disease, cancer and type 2 diabetes comorbidities in HIV infection: A systematic review. Journal of Nursing Education and Practice. 2013; 3: 58-67. 12. Dave JA, Lambert EV, Badri M, West S, Maartens G and Levitt NS. Effect of nonnucleoside reverse transcriptase inhibitor-based antiretroviral therapy on dysglycemia and insulin sensitivity in South African HIV-infected patients. Journal of Acquired Immune Deficiency Syndromes. 2011; 57: 284-9. 13. Dillon DG, Gurdasani D, Riha J, et al. Association of HIV and ART with cardiometabolic traits in sub-Saharan Africa: a systematic review and meta-analysis. International journal of epidemiology. 2013; 42: 1754-71. 14. Dave JA, Lambert EV, Badri M, West S, Maartens G and Levitt NS. Effect of nonnucleoside reverse transcriptase inhibitor-based antiretroviral therapy on dysglycemia and insulin sensitivity in South African HIV-infected patients. Journal of acquired immune deficiency syndromes. 2011; 57: 284-9. 15. Peer N, Steyn K, Lombard C, Lambert EV, Vythilingum B and Levitt NS. Rising diabetes prevalence among urban-dwelling black South Africans. PloS one. 2012; 7: e43336. 16. Bonita R, deCourten M, Dwyer T, Jamrozik K and Winkelmann R. Surveillance of risk factors for non-communicable disease: The WHO STEP-wise Approach Geneva: World Health Org., 2002. 17. World Health Organization. Definition, Diagnosis and Classification of Diabetes Mellitus and its Complications: Report of a WHO Consultation. Geneva: World Health Org, 1999. 18. World Health Organization. Obesity: Preventing and Managing the Global Epidemic: Report of a WHO Consultation. Geneva: World Health Org, 2000. 19. DeFronzo RA and Abdul-Ghani MA. Preservation of beta-cell function: the key to diabetes prevention. J Clin Endocrinol Metab. 2011; 96: 2354-66. 20. Kalra S and Agrawal N. Diabetes and HIV: Current Understanding and Future Perspectives. Current diabetes reports. 2013; 13: 419-27. 21. Aberg JA. Cardiovascular complications in HIV management: past, present, and future. Journal of Acquired Immune Deficiency Syndromes. 2009; 50: 54-64. 22. Samaras K. Prevalence and pathogenesis of diabetes mellitus in HIV-1 infection treated with combined antiretroviral therapy. Journal of Acquired Immune Deficiency Syndromes. 2009; 50: 499-505. 23. Neuhaus J, Jacobs DR, Jr., Baker JV, et al. Markers of inflammation, coagulation, and renal function are elevated in adults with HIV infection. The Journal of infectious diseases. 2010; 201: 1788-95. 24. Maartens G, Celum C and Lewin SR. HIV infection: epidemiology, pathogenesis, treatment, and prevention. Lancet. 2014; 384: 258-71. 25. Paula AA, Falcao MC and Pacheco AG. Metabolic syndrome in HIV-infected individuals: underlying mechanisms and epidemiological aspects. AIDS research and therapy. 2013; 10: 32. 26. Dube MP, Shen C, Greenwald M and Mather KJ. No impairment of endothelial function or insulin sensitivity with 4 weeks of the HIV protease inhibitors atazanavir or
lopinavir-ritonavir in healthy subjects without HIV infection: a placebo-controlled trial. Clinical infectious diseases : an official publication of the Infectious Diseases Society of America. 2008; 47: 567-74. 27. Lake JE and Currier JS. Metabolic disease in HIV infection. The Lancet infectious diseases. 2013; 13: 964-75. 28. Samaras K. The burden of diabetes and hyperlipidemia in treated HIV infection and approaches for cardiometabolic care. Curr HIV/AIDS Rep. 2012; 9: 206-17. 29. Bruno G and Landi A. Epidemiology and costs of diabetes. Transplantation proceedings. 2011; 43: 327-9. 30. Lo YC, Chen MY, Sheng WH, et al. Risk factors for incident diabetes mellitus among HIV-infected patients receiving combination antiretroviral therapy in Taiwan: a case-control study. HIV medicine. 2009; 10: 302-9.
Highlights • The prevalence of diabetes was similar across in people who were HIV-infected antiretroviral therapy (ART)-naïve, on first-line ART: non-nucleoside reverse transcriptase inhibitor (NNRTI) plus dual NRTIs, on second-line ART: lopinavir/ritonavir plus dual NRTIs) and particiants from a community based prevalence survey. • Impaired glucose tolerance (IGT) and impaired fasting glycaemia (IFG) were differentially prevalent by HIV status and treatment regimens, which may be suggestive of different pathways leading to the development of diabetes. • Participants on lopinavir/ritonavir-based regimens were four-times and almost two-times more likely to have dysglycaemia than HIV-uninfected and ART naïve HIV-infected participants, respectively, highlighting the need to consider the metabolic effects of antiretrovirals when selecting ART regimens.
S0168-8227(16)30037-7 http://dx.doi.org/10.1016/j.diabres.2016.03.012 DIAB 6591
To appear in:
Diabetes Research and Clinical Practice
Received Date: Revised Date: Accepted Date:
29 April 2015 29 January 2016 19 March 2016
Please cite this article as: N.S. Levitt, N. Peer, K. Steyn, C. Lombard, G. Maartens, E.V. Lambert, J.A. Dave, Increased risk of dysglycaemia in south africans with HIV; Especially those on protease inhibitors, Diabetes Research and Clinical Practice (2016), doi: http://dx.doi.org/10.1016/j.diabres.2016.03.012
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
INCREASED RISK OF DYSGLYCAEMIA IN SOUTH AFRICANS WITH HIV; ESPECIALLY THOSE ON PROTEASE INHIBITORS Short running title: Dysglycaemia in South Africans with HIV Authors and affiliations: 1,2
Naomi S Levitt,
3,4
Nasheeta Peer, 2Krisela Steyn, 5Carl Lombard, 6Gary Maartens, 7Estelle
V Lambert, 1Joel A Dave 1. Division of Diabetic Medicine and Endocrinology, Department of Medicine, Faculty of Health Sciences, University of Cape Town (UCT), South Africa 2. Chronic Disease Initiative for Africa, Department of Medicine, Faculty of Health Sciences, UCT, South Africa 3. Non-communicable Diseases Research Unit, South African Medical Research Council (SAMRC), South Africa 4. Department of Medicine, Faculty of Health Sciences, UCT, South Africa 5. Biostatistics Unit, SAMRC, South Africa 6. Division of Clinical Pharmacology, Department of Medicine, Faculty of Health Sciences, UCT, South Africa 7. SAMRC/UCT Research Unit for Exercise Science and Sports Medicine, Department of Human Biology, Faculty of Health Sciences, UCT, South Africa
Corresponding author: Professor Naomi Levitt Address: Divisions of Diabetic Medicine and Endocrinology, Department of Medicine, University of Cape Town, Private Bag X3, Observatory 7925, South Africa Telephone: +27 21 404 5007, Fax: +27 21 406 6513 Email: [email protected]
Word count – abstract: 241 Word count – text: 2716 References: 30
ABSTRACT Aims To compare dysglycaemia prevalence (impaired fasting glucose (IFG), impaired glucose tolerance (IGT), or diabetes) in HIV-infected persons, stratified by antiretroviral therapy (ART), with a community-based survey (CBS) in Cape Town, South Africa. Methods Three groups of HIV-infected adults without known diabetes were conveniently sampled from community healthcare centres; ART-naïve, first-line ART (non-nucleoside reverse transcriptase inhibitor (NNRTI) plus dual NRTIs), and second-line ART (lopinavir/ritonavirboosted protease inhibitor plus dual NRTIs). The CBS recruited a representative crosssectional sample from urban townships. Participants reporting ART use or known diabetes were excluded. All participants underwent oral glucose tolerance testing. Multiple logistic regression determined independent associations with dysglycaemia. Results The samples comprised ART-naïve, first-line ART, second-line ART and CBS participants (n=393, 439, 108 and 880, respectively). Mean age was 34-40 years. Dysglycaemia prevalence was as follows: CBS 18.0%, ART-naïve 21.6%, first-line ART 26.0% and secondline ART 37.0%. Diabetes was similar across groups, but IGT was 3-4-fold higher in secondline ART and CBS compared with ART-naïve and first-line ART groups. In contrast, IFG was 14.3-21.2% across HIV groups but only 1.5% in the CBS. Increased risk of dysglycaemia was associated with older age, female gender, and HIV status (ART-naïve: OR 2.31, 95%CI 1.65-3.24; first-line ART: OR 2.47, 95%CI 1.80-3.38; second-line ART: OR 4.10, 95%CI 2.54-6.61). Diabetes family history and central obesity were not related to dysglycaemia. Conclusions In view of the increased risk of dysglycaemia in HIV-infected participants, screening for diabetes should be instituted in ART programmes.
Key words: South Africa, black, HIV, dysglycaemia, diabetes, ART
1. INTRODUCTION There is growing concern about the colliding epidemics of HIV/AIDS and non-communicable diseases (NCD) such as diabetes, particularly in sub-Saharan Africa (SSA)
1-3
. The region
has the largest number of people with HIV/AIDS globally and improved access to antiretroviral therapy (ART) has seen a substantial improvement in life expectancy in these patients
4, 5
. Simultaneously, increasing levels of urbanisation, ageing of the population and
the adoption of western lifestyles, with changes in eating patterns and rising levels of physical inactivity, are driving the NCD epidemic
2, 6, 7
.
With the ageing of the HIV population due to the reduction in premature mortality from ART, there is an increased risk of developing diabetes and other NCDs 8. While numerous reports describe an increased risk of diabetes in people receiving ART
9-11
, there are sparse data
from SSA where the majority of people on ART are still on first-line non-nucleoside reverse transcriptase inhibitor (NNRTI) regimens, and are predominantly younger women
12, 13
. By
contrast, in high income countries, protease inhibitor (PI)-based regimens are often used in first-line ART, and the majority of patients are men. We recently reported a high prevalence of dysglycaemia in a cross sectional study of HIV-infected participants from Cape Town (25.7% in ART-naive and 21.9% in those on first-line ART, especially efavirenz)
14
However,
newly diagnosed diabetes was uncommon. In this paper we extend our previous findings by: i) investigating the prevalence of dysglycaemia (defined as one or more of impaired fasting glucose (IFG), impaired glucose tolerance (IGT) or diabetes) in an additional sample of participants receiving second-line ART and ii) comparing this to HIV-infected cohorts who are ART-naïve and on first-line ART; as well as participants of a cross-sectional diabetes epidemiological study conducted concurrently in the same geographic area of Cape Town 15.
2. MATERIAL AND METHODS 2.1. Study populations 2.1.1. HIV groups: HIV-infected adults aged ≥18 years were categorized into three groups: ART-naïve, first-line ART (stavudine (d4T), lamivudine (3TC), and efavirenz or nevirapine), and second-line ART (zidovudine (AZT), didanosine (ddI) and lopinavir-ritonavir (LPV-r)). Participants were conveniently sampled consecutively from a community healthcare centre in Cape Town 2008-2010. Second-line ART participants were recruited from a number of community healthcare centres in the same area, because there were smaller numbers on second-line at
the time. Participants were excluded if they had: 1) been on ART for less than six months, 2) a history of diabetes mellitus or IGT, 3) had an active, acute opportunistic infection, 4) severe diarrhea (six stools/day), 5) tuberculosis within one month of commencing treatment, 6) received glucocorticoid therapy within the past six months, or were 7) pregnant or 8) known to have renal failure. 2.1.2. Community-based survey: Residents living in the predominantly black African areas of Langa, Guguletu, Crossroads, Nyanga and Khayelitsha in Cape Town who were 25-74 years old were the target population. Based on an estimated diabetes prevalence of 8% with a precision of 1.5% twosided with 95% confidence, a sample size of 1000 was planned. A 3-stage cluster sampling stratified by area and housing type was undertaken in 2008/09. Individuals from households were selected using quotas for pre-specified age and gender categories. The exclusion criteria were: unable to give consent, on tuberculosis treatment, known to be on ART, treatment for cancer within the last year, bedridden, pregnant or lactating, or resident in Cape Town for less than 3 months. Replacements were allowed when confronted with individuals who met the exclusion criteria above, those who refused, or the randomly selected participant of the randomly selected household could not be contacted on the third attempt. For the purpose of this analysis, participants aged 25-60 years without a known diagnosis of diabetes were included. The decreased age criteria was to reduce the potential impact of an imbalance in the older ages between the CBS and the HIV groups, which had a small number of participants over the age of 60 years. 2.2. Data collection In both surveys fieldworkers administered questionnaires to obtain socio-demographic, selfreported medical and family history, tobacco (WHO STEP-wise surveillance questionnaire) 16 . Height, weight, and waist and hip circumferences were measured using standardised techniques. Blood samples, for glucose estimations, were drawn following an overnight fast of 10 hours. Thereafter, a standard oral glucose tolerance test (OGTT), using 75 grams of anhydrous glucose in 250ml of water, was administered, and blood samples taken 120 minutes later for glucose estimation
17
. Blood samples were kept on ice and transported to the laboratory
within six hours to be centrifuged and aliquoted. 2.3. Definitions
Diabetes, IGT and IFG were diagnosed according to the 1998 WHO definition 17. Body mass index (BMI) was calculated as the participant’s weight in kilograms divided by their height in metres squared (kg/m²). Obesity (≥30 kg/m²) and increased waist circumference (>94 cm in men and >80 cm and in women) were defined using standardised international criteria 18. 2.4. Data Analyses Data analyses were done using STATA 11 and SAS Version 9.2.3. The analysis of the combined data assumed that all groups were obtained through simple random sampling for inference purposes. Descriptive statistics, including crude prevalence and 95% confidence intervals (CI), were calculated. Univariate analyses (sociodemographic and anthropometric) are presented as mean values and standard deviations for continuous data, and as percentages for categorical data. Standardised prevalence for dysglyceamia and its components were calculated using the age and sex distribution of the 2009 South African Black population. In a multiple logistic regression model the outcome new onset dysglycaemia (IFG, IGT and diabetes combined) was modelled on age, BMI, sex, housing type, waist and hip circumferences, study group and family history of diabetes. The odds ratios (OR) and 95% CI are reported as the estimated risk measures. The University of Cape Town’s Research and Ethics Committee approved both studies. All participants signed informed consent.
3. RESULTS As seen in Table 1 the participants from the community-based sample were older than the HIV groups with a greater proportion of men, but women were in the majority in all groups. The highest proportion of participants living in built formal housing were those on second-line ART (39.8%) compared to the other groups (16.3-19.9%). A family history of diabetes was most common in the CBS (21.2%) compared to the other groups (15.6-16.4%). The median duration on first-line and second-line ART was 16 and 18 months, respectively. The mean BMI (Table 2) fell into the overweight range in all groups. The proportions of participants with obesity and increased waist circumference increased progressively from the ART-naive to the first-line ART, second-line ART and the CBS groups. Notably a much greater proportion of women than men had an increased waist circumference in each group. In contrast, the prevalence of dysglycaemia was lowest in the CBS group (18.0%), increased progressively in the ART-naïve (21.6%), the first-line ART (26.0%), and the second-line ART group (37.0%) (Table 3). The prevalence of diabetes was similar across the four groups. The prevalence of IGT was 3-4-fold higher in the second-line ART and CBS groups
compared to the ART-naïve and first-line ART groups. On the other hand, IFG was rare in the CBS, and the prevalence ranged between 14.3% and 21.2% across the HIV groups. In the multiple logistic regression analysis (Table 4) increasing age, female gender and HIV status were significantly associated with an increased risk of dysglycaemia. Compared to the community-based sample, the likelihood for dysglycaemia was increased (p<0.001) in the HIV infected participants and similar for ART-naïve (OR 2.31, 95% CI 1.653.24) and first-line ART (OR 2.47, 95% CI 1.80-3.38) groups while the prevalence for second-line ART was higher (OR 4.10, 95% CI 2.54-6.61). Waist circumference, quality of housing and a family history of diabetes were not associated with dysglycaemia. When the duration on ART was examined by splitting the two ART groups by their respective median times on ART, the risk for dysglycaemia was similar, when adjusted for the other covariates in the model (data not shown).
4. DISCUSSION The major finding of our study was the marked increase in the risk for dysglycaemia in HIVinfected participants compared to the community-based sample, particularly in the secondline ART group. It is of concern that the prevalence of dysglycaemia was 2-4-fold higher in HIV-infected groups compared to the CBS group, as diabetes in the community-based sample was already 1.5-fold greater than the prevalence found in this urban black population almost two decades ago 15. The increase in diabetes prevalence in the community-based sample was attributed to traditional risk factors such as obesity and urbanisation, however, the factors contributing to the higher dysglycaemia prevalence in HIV-infected individuals are likely to be more complex. Defronzo et al (2011) have suggested that the pathophysiology underlying IFG is moderate hepatic insulin resistance and impaired early but intact late insulin response19. In line with this, we previously reported that the HIV infected dysglycaemic subjects who were either ART naïve or receiving first line ART were more insulin resistant as measured by HOMA-IR and had lower beta-cell function (insulinogenic index and disposition index) than their normoglycaemic counterparts12.
Indeed the presence of HIV infection itself may
contribute to metabolic changes, both directly through immune activation and inflammation, and indirectly through immunodeficiency
10, 20, 21
. Increased pro-inflammatory markers such
as C-reactive protein (CRP), tumor necrosis factor-a (TNF-a), and interleukin-6 together with low levels of adiponectin characterise insulin resistance and type 2 diabetes in HIV infection 22-24
.
Furthermore, since there was a greater risk for dysglycaemia in participants on second-line ART compared to ART-naïve and first-line ART participants there appears to be an association between the type of ART regimen and dysglycaemia. The ART attributable risk for dysglycaemia may be through the effect of the drugs on insulin resistance and body fat distribution but also possibly through other pathways that are currently unclear 10.. Lopinavir-ritonavir, given to most participants on second-line ART, has been implicated in the development of insulin resistance and alterations in glucose metabolism in some but not all studies, and may possibly explain the higher risk for dysglycaemia associated with its use in our study
25, 26
. All HIV-infected individuals are at an increased risk for dysglycaemia,
however, participants on second-line ART appear to be more vulnerable and possibly require closer monitoring for metabolic complications. These findings highlight that the selection of ART regimen in HIV-infected participants should consider both the virological effectiveness and the metabolic effects of the drugs, particularly as antiretroviral drugs with more neutral metabolic profiles are available
27
. The
higher prevalence of dysglycaemia found in the HIV-infected participants, irrespective of ART status, compared to the community-based participants demonstrates the need for preventive screening for diabetes and behavioural counseling to promote lifestyle changes. Additionally, novel therapies should be considered to prevent or treat metabolic abnormalities in the HIV-infected patient such as the use of anti-inflammatory and immunemodulatory drugs 27. Nevertheless, despite the similarities in mean BMI between participants on both first- and second-line ART, the prevalence of obesity was much higher in those on second-line ART and may have, in addition to the diabetagenic effect of the ART, contributed to their greater dysglycaemia. Overweight/obesity is recognised as the environmental risk factor most strongly associated with the development of diabetes,
28
and obese HIV-infected individuals
should expect to accumulate the same metabolic abnormalities as the general population
21,
27
.
However, the relationship between traditional risk factors for diabetes, and HIV- and ARTinduced metabolic changes is complex and the latter may accelerate diabetogenesis in HIV infection in the absence of obesity 27. This may account for the finding that even though 80% of diabetic individuals are usually overweight/obese reported community-based study
29
, as was the case in our previously
15
, no significant associations between BMI, waist or hip
circumferences with dysglycaemia were present in the current study. A case-control study in Taiwan also did not show a significant relation between BMI and incident diabetes in HIVinfected patients 30.
HIV-specific influences may predispose to diabetes at lower levels of adiposity in the HIVinfected compared to the HIV-uninfected
27
population
. These effects are partly due to
inflammation and insulin resistance induced by lipodystrophy, co-morbidities and ART. Other risk factors may include high pre-ART viral loads and low baseline CD4 counts, which possibly increase the risk of insulin resistance and escalate the development of dysglycaemia independent of clinical adiposity 20. Nonetheless, traditional risk factors play an important role in the development of diabetes among HIV-infected cohorts
28
and were likely also influential in the progression to
dysglycaemia in this study. Age, a consistent predisposing risk factor for incident diabetes in HIV cohort studies
28
, was found to be significantly related to dysglycaemia. This suggests
that with increased longevity among HIV-infected patients on ART, a rise in dysglycaemia independent of HIV-related influences may occur with ageing of this population. The successful rollout of the ART campaign in South Africa has prolonged the life expectancy of HIV-infected patients such that diabetes and other cardiovascular diseases risk factors are likely to become a rising health burden. The absence of a significant association between a family history of diabetes and dysglycaemia may be related to a lack of knowledge about the disease rather than a true reflection of a negative family history. This suggestion is supported by a much lower prevalence of diabetes family history among unknown (22.8%) compared to known (46.2%) diabetic participants in our community-based study
15
. A family history of diabetes in the
latter study was significantly associated with diabetes in women (among whom unknown diabetes was 33.9%) but not in men (51.8% unknown diabetes). Nonetheless, further investigation is required to support this notion as not all HIV studies have shown a relation between family history and diabetes 28. The increased likelihood of dysglycaemia in women compared to men contrasts with the predominance of diabetes in the latter among HIV-infected individuals 20, 28. Mitigating factors may include ethnicity with few studies conducted in black populations to confirm the gender distribution or differences in HIV-related risk parameters. Our study had several limitations. First, the HIV status of the community participants was not known, although participants receiving ART were excluded. However, the HIV-infected proportion of the CBS, which we estimate to be about 10%, would have diluted any effect of HIV on dysglycaemia. Therefore the true risk of dysglycaemia in HIV-infected individuals is likely to be higher than we calculated. Second, we are unable to comment on the total diabetes prevalence in HIV-infected participants because those with known diabetes were excluded from the HIV study. Third, the small number of participants on second-line ART
was a consequence of the limited number of people receiving this therapy at the time of the study. Fourth, there were methodological differences in sampling design of the study groups. Also, considering the cross-sectional design of these studies, causality cannot be established. To facilitate any inference a basic sampling design were assumed p-values and confidence intervals must therefore be considered investigative. The major strength of the study is the use of an OGTT which has permitted an assessment of the prevalence of IGT in addition to IFG and the diagnosis of diabetes based on both fasting and 2-hour post OGTT measures. The findings of this study may be generalizable to other urban South African black populations.
5. CONCLUSION The increased likelihood of dysglycaemia with HIV infection highlights the need for concurrent management of cardio-metabolic abnormalities in the HIV-infected population. The rise in longevity on ART will inevitably be accompanied by a rise in diabetes-associated morbidity and mortality, with a greater burden imposed on the South African healthcare system. Although this study has demonstrated a higher prevalence of dysglycaemia in HIV-infected patients, especially those on second-line ART, the pathways contributing to the development and the differentiation of these diseases is currently poorly elucidated
27
. Further research is
needed to understand the pathophysiology of dysglycaemia in HIV-infected patients so as to optimise prevention and management strategies.
ACKNOWLEDGEMENTS: We would like to acknowledge the contributions of the study co-ordinators Carmen Delport and Serena van Haght and their respective teams of fieldworkers .GM acknowledges research support in part by the National Research Foundation (NRF) of South Africa (grant specific unique reference number 85810). The grant holder acknowledges that opinions, findings and conclusions or recommendations expressed in any publication generated by the NRF supported research are that of the author, and that the NRF accepts no liability whatsoever in this regard. Funding The Community-based survey was supported by an unrestricted grant from Servier Laboratories (South Africa); the South African Medical Research Council; the Initiative for Cardiovascular Health Research in Developing Countries (IC Health) Foundation Council;
and Brigham and Women’s Hospital, Harvard University. The study in the HIV cohort was funded by grants from the World Diabetes Foundation and the South African Department of Health. Conflict of interest Naomi Levitt has received travel support from Novo Nordisk, Eli Lilly Laboratories and Sanofi Aventis. All other authors report no potential conflicts of interest, including specific financial interests, relevant to the subject of this manuscript. Contributors
Table 1: Characteristics of the samples studied Communitybased
HIV Group
Survey st
ART naïve
1 line ART
2
nd
line ART
Number
880
393
439
108
Age in years, mean (SD)
39.9 (10.08)
33.6 (8.77)
36.1 (8.87)
36.6 (8.32)
Men
37.6
23.9
22.6
15.7
Women
62.4
76.1
77.5
84.3
Family history of diabetes, %
21.2
15.6
16.4
15.7
Alcohol use, %
44.7
52.0
42.6
39.8
Formal/built
18.9
19.9
16.3
39.8
Council/core
33.64
33.9
32.3
13.9
Shack/shelter/ hostel
47.5
44.6
48.5
40.7
3 (2-4)
4 (2-5)
3(2-5)
4 (3-5)
249 (154 –
315 (220 –
479 (261 -
434)
461)
624)
-
16 (10 – 25)
18 (9.5 - 26)
Gender, %:
Housing type, %:
Housing density Median CD4 count cells/µL (interquartile range)
Median duration ART (months) Number of participants on the following ART drugs:
-
-
Zidovudine (AZT)
-
-
145
82
Lamivudine (3TC)
-
-
439
36
Stavudine (D4T)
-
-
294
61
Efavirenz (EFV)
-
-
236
7
Lopinavir/Ritonavir
-
-
-
104
ART: antiretroviral therapy
Table 2: Anthropometry by group HIV Group
CommunityGroup
based Survey
st
ART Naive
1 line ART
N=393
N=439
N=880
nd
2
line ART N=108
Body mass index (BMI) Mean (SD), kg/m
2
29.3 (8.39)
25.9 (6.17)
26.8 (5.81)
27.5 (6.09)
41.4
21.9
23.7
30.6
Men
84.1 (12.99)
80.2 (8.01)
82.5 (9.45)
83.8 (9.97)
Women
96.2 (15.2)
84.1 (13.1)
87.8 (11.8)
90.8 (13.6)
Men (WC >94 cm)
18.7
6.4
14.1
17.7
Women (WC >80 cm)
85.6
53.5
70.9
75.8
2
Obesity (BMI ≥30 kg/m ),% Waist circumference (WC) Mean (SD), cm:
Raised, %:
ART: antiretroviral therapy
Table 3a: Crude prevalence (95% CI) of dysglycaemia (diabetes, IGT and IFG) by group HIV Group
CommunityGroup
Total dysglycaemia Diabetes
st
nd
based Survey
ART Naive
1 line ART
2
N=880
N=393
N=439
N=108
18.0 (15.4-20.5)
21.6 (17.5- 25.7)
26.0 (21.9-30.1)
37.0 (27.9-46.2)
line ART
4.9 (3.5-6.3)
3.1 (1.4-4.8)
2.3 (0.9-3.7)
5.6 (1.2-9.9)
IGT
11.6 (9.5-13.7)
4.3 (2.3-6.3)
2.5 (1.0-4.0)
12.0 (5.9-18.2)
IFG
1.5 (0.6-2.3)
14.3 (10.7-17.7)
21.2 (17.4-25.0)
19.4 (11.9-26.9)
ART: antiretroviral therapy; IGT: impaired glucose tolerance; IFG: impaired fasting glycaemia
Table 3b: Standardised* prevalence of dysglycaemia (diabetes, IGT and IFG) by group HIV Group
CommunityGroup
based Survey
ART Naive
1st line ART
2nd line ART
N=880
N=393
N=439
N=108
Total dysglycaemia
15.0
26.0
29.9
34.3
Diabetes
4.1
2.7
3.0
8.8
IGT
9.4
4.5
2.1
5.7
IFG
1.5
18.9
24.8
19.9
*standardized for age and sex using the 2009 South African midyear Black population
Table 4: Logistic regression analysis for independent factors associated with new onset dysglycaemia (IFG, IGT and diabetes combined) Odds Ratio
95% Confidence Interval Lower limit
P-value
Upper limit
<0.001
Age (years) 25-34
1.00
35-44
1.82
1.29
2.57
0.001
45-54
3.27
2.25
4.75
<0.001
55-64
4.75
3.09
7.32
<0.001
2
BMI (kg/m )
0.060 <25
25-30
1.34
.89
2.01
0.160
>30
1.92
1.11
3.30
0.019
Waist circumference (cm)
1.20
0.78
1.83
0.412
Hip circumference (cm)
1.00
0.99
1.01
0.975
Sex: women
2.17
1.54
3.03
<0.001
Housing
0.406 Formal
1.00
council
0.84
0.60
1.18
0.316
Shack/shelter/hostel
1.08
0.79
1.48
0.620
<0.001
Group
Community survey
1.00
ART naive
2.31
1.65
3.24
<0.001
st
2.47
1.80
3.38
<0.001
nd
2 line ART
4.10
2.54
6.61
<0.001
Family history of diabetes
1.11
0.82
1.50
0.482
1 line ART
ART: antiretroviral therapy; IGT: impaired glucose tolerance; IFG: impaired fasting glycaemia
REFERENCES 1. Levitt NS, Steyn K, Dave J and Bradshaw D. Chronic noncommunicable diseases and HIV-AIDS on a collision course: relevance for health care delivery, particularly in lowresource settings--insights from South Africa. The American journal of clinical nutrition. 2011; 94: 1690S-6S. 2. Mbanya JC, Motala AA, Sobngwi E, Assah FK and Enoru ST. Diabetes in subSaharan Africa. Lancet. 2010; 375: 2254-66. 3. Mayosi BM, Lawn JE, van Niekerk A, Bradshaw D, Abdool Karim SS and Coovadia HM. Health in South Africa: changes and challenges since 2009. Lancet. 2012; 380: 202943. 4. Hankins C. Overview of the Current State of the Epidemic. Current HIV/AIDS Reports. 2013: 1-11. 5. Bor J, Herbst AJ, Newell ML and Barnighausen T. Increases in adult life expectancy in rural South Africa: valuing the scale-up of HIV treatment. Science. 2013; 339: 961-5.
6. Kengne AP, Echouffo-Tcheugui J-B, Sobngwi E and Mbanya J-C. New insights on diabetes mellitus and obesity in Africa–Part 1: prevalence, pathogenesis and comorbidities. Heart. 2013. 7. Mensah GA. Descriptive Epidemiology of Cardiovascular Risk Factors and Diabetes in Sub-Saharan Africa. Progress in Cardiovascular Diseases. 2013; 56: 240-50. 8. McDonald CL and Kaltman JR. Cardiovascular disease in adult and pediatric HIV/AIDS. Journal of the American College of Cardiology. 2009; 54: 1185-8. 9. Barbaro G. Metabolic and cardiovascular complications of highly active antiretroviral therapy for HIV infection. Current HIV research. 2006; 4: 79-85. 10. Martinez E, Larrousse M and Gatell JM. Cardiovascular disease and HIV infection: host, virus, or drugs? Current opinion in infectious diseases. 2009; 22: 28-34. 11. Nigatu T, Oldenburg B, Elliott J, Setswe G and Woldegiorgis MA. The incidence of cardiovascular disease, cancer and type 2 diabetes comorbidities in HIV infection: A systematic review. Journal of Nursing Education and Practice. 2013; 3: 58-67. 12. Dave JA, Lambert EV, Badri M, West S, Maartens G and Levitt NS. Effect of nonnucleoside reverse transcriptase inhibitor-based antiretroviral therapy on dysglycemia and insulin sensitivity in South African HIV-infected patients. Journal of Acquired Immune Deficiency Syndromes. 2011; 57: 284-9. 13. Dillon DG, Gurdasani D, Riha J, et al. Association of HIV and ART with cardiometabolic traits in sub-Saharan Africa: a systematic review and meta-analysis. International journal of epidemiology. 2013; 42: 1754-71. 14. Dave JA, Lambert EV, Badri M, West S, Maartens G and Levitt NS. Effect of nonnucleoside reverse transcriptase inhibitor-based antiretroviral therapy on dysglycemia and insulin sensitivity in South African HIV-infected patients. Journal of acquired immune deficiency syndromes. 2011; 57: 284-9. 15. Peer N, Steyn K, Lombard C, Lambert EV, Vythilingum B and Levitt NS. Rising diabetes prevalence among urban-dwelling black South Africans. PloS one. 2012; 7: e43336. 16. Bonita R, deCourten M, Dwyer T, Jamrozik K and Winkelmann R. Surveillance of risk factors for non-communicable disease: The WHO STEP-wise Approach Geneva: World Health Org., 2002. 17. World Health Organization. Definition, Diagnosis and Classification of Diabetes Mellitus and its Complications: Report of a WHO Consultation. Geneva: World Health Org, 1999. 18. World Health Organization. Obesity: Preventing and Managing the Global Epidemic: Report of a WHO Consultation. Geneva: World Health Org, 2000. 19. DeFronzo RA and Abdul-Ghani MA. Preservation of beta-cell function: the key to diabetes prevention. J Clin Endocrinol Metab. 2011; 96: 2354-66. 20. Kalra S and Agrawal N. Diabetes and HIV: Current Understanding and Future Perspectives. Current diabetes reports. 2013; 13: 419-27. 21. Aberg JA. Cardiovascular complications in HIV management: past, present, and future. Journal of Acquired Immune Deficiency Syndromes. 2009; 50: 54-64. 22. Samaras K. Prevalence and pathogenesis of diabetes mellitus in HIV-1 infection treated with combined antiretroviral therapy. Journal of Acquired Immune Deficiency Syndromes. 2009; 50: 499-505. 23. Neuhaus J, Jacobs DR, Jr., Baker JV, et al. Markers of inflammation, coagulation, and renal function are elevated in adults with HIV infection. The Journal of infectious diseases. 2010; 201: 1788-95. 24. Maartens G, Celum C and Lewin SR. HIV infection: epidemiology, pathogenesis, treatment, and prevention. Lancet. 2014; 384: 258-71. 25. Paula AA, Falcao MC and Pacheco AG. Metabolic syndrome in HIV-infected individuals: underlying mechanisms and epidemiological aspects. AIDS research and therapy. 2013; 10: 32. 26. Dube MP, Shen C, Greenwald M and Mather KJ. No impairment of endothelial function or insulin sensitivity with 4 weeks of the HIV protease inhibitors atazanavir or
lopinavir-ritonavir in healthy subjects without HIV infection: a placebo-controlled trial. Clinical infectious diseases : an official publication of the Infectious Diseases Society of America. 2008; 47: 567-74. 27. Lake JE and Currier JS. Metabolic disease in HIV infection. The Lancet infectious diseases. 2013; 13: 964-75. 28. Samaras K. The burden of diabetes and hyperlipidemia in treated HIV infection and approaches for cardiometabolic care. Curr HIV/AIDS Rep. 2012; 9: 206-17. 29. Bruno G and Landi A. Epidemiology and costs of diabetes. Transplantation proceedings. 2011; 43: 327-9. 30. Lo YC, Chen MY, Sheng WH, et al. Risk factors for incident diabetes mellitus among HIV-infected patients receiving combination antiretroviral therapy in Taiwan: a case-control study. HIV medicine. 2009; 10: 302-9.
Highlights • The prevalence of diabetes was similar across in people who were HIV-infected antiretroviral therapy (ART)-naïve, on first-line ART: non-nucleoside reverse transcriptase inhibitor (NNRTI) plus dual NRTIs, on second-line ART: lopinavir/ritonavir plus dual NRTIs) and particiants from a community based prevalence survey. • Impaired glucose tolerance (IGT) and impaired fasting glycaemia (IFG) were differentially prevalent by HIV status and treatment regimens, which may be suggestive of different pathways leading to the development of diabetes. • Participants on lopinavir/ritonavir-based regimens were four-times and almost two-times more likely to have dysglycaemia than HIV-uninfected and ART naïve HIV-infected participants, respectively, highlighting the need to consider the metabolic effects of antiretrovirals when selecting ART regimens.