Skip Navigation LinksHome > September 1, 2005 - Volume 40 - Issue 1 > Increasing Risk of 5% or Greater Unintentional Weight Loss i...
JAIDS Journal of Acquired Immune Deficiency Syndromes:
Epidemiology and Social Science

Increasing Risk of 5% or Greater Unintentional Weight Loss in a Cohort of HIV-Infected Patients, 1995 to 2003

Tang, Alice M PhD*; Jacobson, Denise L PhD*; Spiegelman, Donna ScD†; Knox, Tamsin A MD‡; Wanke, Christine MD*

Free Access
Article Outline
Collapse Box

Author Information

From the *Division of Nutrition and Infection, Department of Family Medicine and Community Health, Tufts University School of Medicine, Boston, MA; †Departments of Epidemiology and Biostatistics, Harvard School of Public Health, Boston, MA; and ‡Division of Gastroenterology, New England Medical Center and Tufts University School of Medicine, Boston, MA.

Received for publication May 31, 2004; accepted February 4, 2005.

Supported by National Institute of Diabetes and Digestive and Kidney Diseases grant 1P01DK45734-06 and the General Clinical Research Center funded by the National Center for Research Resources of the National Institutes of Health under grant M01-RR00054.

Reprints: Alice M. Tang, Department of Public Health and Family Medicine, Tufts University School of Medicine, 136 Harrison Avenue, Posner 4, Boston, MA 02111 (e-mail: alice.tang@tufts.edu).

Collapse Box

Abstract

Although the incidence of most AIDS-defining opportunistic infections, including HIV wasting syndrome, has dramatically decreased since the introduction of highly active antiretroviral therapy (HAART), previous studies have shown that weight loss and wasting are still common in HIV-infected persons. We examined the 6-month risk and determinants of ≥5% weight loss during the period when the use of combination antiretroviral therapy and HAART was commonplace among 713 participants enrolled in the Nutrition for Healthy Living cohort in Boston, Massachusetts between 1995 and 2003. There was a significant 50% increase in the 6-month risk of ≥5% weight loss in the later HAART years (1998-2003) compared with the early HAART years (1995-1997) among most of the participants who reported they were not trying to lose weight (P = 0.002). In addition to calendar time, several other variables were significantly independently associated with risk of ≥5% weight loss, including use of injection drugs; living below the federal poverty level; higher body mass index (BMI; ≥25 kg/m2); lower CD4+ cell count; higher HIV viral load; and presence of diarrhea, nausea, or fever. The characteristics of weight loss in the later HAART years did not differ from the early HAART years with respect to initial body composition (eg, weight, BMI, triceps skinfold thickness) or changes in body composition during the periods of weight loss. In summary, we have found that the risk of ≥5% unintentional weight loss over 6-month intervals is on the rise in our cohort of HIV-infected participants, despite better control of HIV infection in recent years. Although we still do not know the exact cause of this increase, the fact that it exists indicates the need for clinicians who take care of HIV-infected patients to continue to pay attention to weight loss among particular segments of their patient population. This is particularly important because recent studies have shown that even a 5% weight loss in 6 months markedly increases the risk of death.

Before the advent of highly active antiretroviral therapy (HAART), severe weight loss and wasting were characteristic of AIDS patients.1-3 Since the introduction of HAART, the incidence of most AIDS-defining opportunistic infections has dramatically decreased.4 Although some studies demonstrate that the incidence of HIV-related wasting syndrome has also declined in the HAART era, data from the Nutrition for Healthy Living (NFHL) cohort show that weight loss and wasting are still common in HIV-infected persons and that even a 5% weight loss in 6 months markedly increases the risk of death.5,6

The studies that have reported declines in the incidence of HIV-related wasting syndrome have based their analyses on the Centers for Disease Control and Prevention (CDC) surveillance definition of wasting: “profound involuntary weight loss of >10% of baseline body weight plus either chronic diarrhea (at least two loose stools per day for ≥30 days) or chronic weakness and documented fever (for ≥30 days, intermittent or constant) in the absence of a concurrent illness or condition other than HIV infection that could explain the findings (eg, cancer, tuberculosis, cryptosporidiosis, or other specific enteritis)”.7 Smit et al8 found that the incidence of physician-diagnosed HIV wasting syndrome consistently increased from approximately 7 cases per 1000 person-years in 1988 through 1990 to around 22 cases per 1000 person-years in 1994 through 1995 among approximately 2000 participants in the Multicenter AIDS Cohort Study (MACS). The incidence subsequently dropped to 13.4 cases per 1000 person-years during the period from 1996 through 1999 when HIV treatment was characterized as predominantly HAART.

Dworkin et al9 also examined the incidence of physician-diagnosed HIV wasting syndrome among 46,678 HIV-infected patients followed from 1992 to 1999 in the Adult and Adolescent Spectrum of HIV Disease (ASD) Project, a national surveillance project of the CDC. In this study, the incidence of wasting syndrome decreased from 30.2 cases per 1000 person-years in 1992 to 11.9 cases per 1000 person-years in 1999, with the most significant rate of decline occurring after 1995.

In an earlier study, Mocroft et al10 described temporal changes in the incidence of individual AIDS-defining illnesses over a 10-year period (1987-1997) in a London clinic. Similar to the overall trend of AIDS-defining illnesses, the incidence of HIV wasting syndrome itself declined dramatically from a relatively steady average of 26.2 cases per 1000 person-years in the years up to 1996 to 4 cases per 1000 person-years in 1997. In all these studies, the decline in incidence of HIV wasting syndrome observed in the late 1990s was attributed to the introduction of HAART.

Although the clinical diagnosis of HIV wasting syndrome, as defined by the CDC, seems to be on the decline, Wanke et al5 found that weight loss and wasting are still common in HIV-infected patients in the post-HAART era. In this study, the definition of wasting was based on weights and heights measured every 6 months in a research setting rather than on physician diagnoses. Wasting was defined as (1) >10% loss of baseline weight; (2) >5% weight loss between study visits, with the loss sustained for at least 1 year; or (3) body mass index (BMI) falling to <20 kg/m2. Symptoms were not included in this definition of wasting. Of 466 individuals who were enrolled in the study at the time and did not self-report wasting at the time of their enrollment, 34% met at least 1 of these definitions of wasting since their initial study visit. Of the 64 participants who were not on HAART at enrollment and who met at least 1 of the wasting definitions during their follow-up, 31 (48%) met that definition after HAART therapy was begun. In addition, 58% of the cohort lost more than 1.5 kg of weight between any 2 consecutive study visits.

Earlier reports have shown that weight loss of as little as 5% was predictive of death.11 We recently found that this remains true in the era of HAART.6 Therefore, it is important to document the continued occurrence of weight loss in the current era so that physicians may be aware of the possible risks to their patients. To further the findings of Wanke et al5 in our cohort, we examined changes in the risk of weight loss during the period spanning combination antiretroviral therapy (ART) and HAART. To our knowledge, there have been no other reports in the literature on trends in actual weight loss after the introduction of HAART.

Back to Top | Article Outline

METHODS

Study Population

All participants are enrolled in the Tufts University NFHL study, an ongoing cohort study of the nutritional and metabolic consequences of HIV infection. Recruitment for this study has been ongoing since 1995. To date, a total of 881 participants have been enrolled in the study. Eligible participants include HIV-positive adults (aged 18 years or older) living in the greater Boston area or Rhode Island. Individuals are excluded if they have any of the following conditions at enrollment: pregnancy, diabetes, thyroid disease, or any malignancies other than Kaposi sarcoma. Participants are also excluded if they are not fluent in English. Further details of this study have been reported elsewhere.12-14 Briefly, all participants are seen at semiannual visits. Data collected at each study visit include weight, height, skinfold (triceps, suprailiac, and subscapular) measurements, bioelectrical impedance analysis (BIA; BIA-101A; RJL Systems, Clinton Twp, MI), resting energy expenditure (REE), and dietary intake from 3-day food records. Fat-free mass (FFM) is derived from BIA resistance and reactance measurements using the sex- and weight-specific equations of Lukaski.15 Fat mass is obtained by subtracting FFM from total body weight. Body cell mass (BCM) is calculated from BIA resistance and reactance measurements using the equations of Kotler et al.16 At each visit, participants are also administered a detailed questionnaire eliciting information on sociodemographic characteristics, clinical status, health-related quality of life, use of recreational drugs, use of HIV-related medications, and whether or not they had intentionally tried to lose weight since their previous visit. Blood is collected and stored at each visit for immunologic, biochemical, and nutritional testing.

To be eligible for this analysis, participants had to have at least 2 study visits for the calculation of weight change. Only person-intervals (a person-interval being bounded by 2 visits) of 4 to 10 months were considered in this analysis to standardize the rate of weight loss among participants. Of the 881 participants enrolled in the NFHL study since 1995, 168 (19%) were excluded from this analysis because they did not have at least 2 study visits within 4 to 10 months apart. Twenty-six percent of those who had only 1 study visit died after their baseline visit. The remaining 713 participants contributed a total of 4237 person-intervals to the analysis. The median number of intervals contributed by each participant was 5 (interquartile range [IQR]: 2-9 intervals), with participants contributing a minimum of 1 interval and a maximum of 17 intervals. The median interval length was 6.2 months between study visits (IQR: 5.9-6.7 months).

This study has been approved by the Institutional Review Boards of Tufts-New England Medical Center and Miriam Hospital in Providence, Rhode Island.

Back to Top | Article Outline
Definition of HIV-Related Weight Loss

At each study visit, participants were weighed (to the nearest 0.1 kg) without shoes and in light clothing using a calibrated standing beam balance scale. Change in weight was calculated by subtracting weight at the previous visit from weight at the current visit. Percent changes were then calculated based on weights at the start of each interval. Negative changes of 5% or more were considered significant weight losses.

Back to Top | Article Outline
Statistical Analysis

This repeated-measures analysis includes data collected from study visits between February 1995 and March 2004. The unit of analysis was person-intervals. The outcome of interest was loss of 5% weight or more from the previous study visit (yes or no). Because repeated measures on the same individuals were allowed, the generalized estimating equation (GEE) approach for clustered data was used to adjust for within-individual correlations. Because recurrent episodes of ≥5% weight loss within a person were treated as separate episodes, our measure of risk is referred to as a “6-month risk” (ie, weight loss over an average period of 6 months) rather than an incidence or prevalence. To compare the 6-month risk of ≥5% weight loss according to various characteristics, GEE models using the binomial distribution and log link were fit. This type of model yields risk ratios (RRs), which, in this case, are interpreted as the 6-month risk of ≥5% weight loss among those with a certain characteristic compared with the 6-month risk of ≥5% weight loss among those without that characteristic. Restricted cubic spline functions were fit to the data to evaluate evidence of nonlinear associations between ≥5% weight loss and the independent variables.17

Our primary independent variable of interest was calendar time, which we examined as a continuous variable and as a binary variable. The midpoint of the visit dates that bounded each weight loss interval was used as the calendar time of that interval. The binary variable (labeled “post-1997”) was used to compare the 6-month risk of ≥5% weight loss among the later HAART years (1998-2004) with the 6-month risk of ≥5% weight loss in the pre- and/or early HAART years (1995-1997).

We considered several other determinants of weight loss as potential confounders in the relation between calendar time and weight loss and as independent determinants of weight loss. These included the following: gender, race and/or ethnicity (white vs. nonwhite), age, injection drug use (IDU; ever vs. never), insecure housing (yes vs. no), poverty (yes vs. no), presence of symptoms over the interval (eg, nausea, diarrhea, thrush, painful gums, fever, cough), BMI (<20, 20 to <25, 25 to <30, and ≥30 kg/m2), HIV viral load (<104, 104-105, and >105 copies/mL), CD4+ cell counts <200 cells/μL, serum albumin level (<3.4 g/dL), any AIDS-defining opportunistic infections over the interval, REE per kilogram of FFM in quartiles, energy intake per kilogram of weight (quartiles), duration of HIV infection, year of enrollment, HAART use (yes vs. no), duration of HAART use (in months), any ART use (yes vs. no), nucleoside analogue reverse transcriptase inhibitor (NRTI) use (yes vs. no), and nonnucleoside reverse transcriptase inhibitor (NNRTI) use (yes vs. no). Age in years was categorized as <35, 35 to 45, and >45 years based on spline curves indicating that the association between risk of ≥5% weight loss and age might be nonlinear. Poverty was defined according to the 1998 US federal poverty level definition (ie, annual income less than $8050 for an individual income plus allowances of $2800 for each additional household member). A participant was classified as impoverished if their self-reported household income range fell within the poverty level. Insecure housing was defined as currently living in a shelter, rooming or boarding house, residential drug or alcohol treatment facility, or on the streets. Secure housing included living in one's own home, a parent's home, or a friend's home. HAART regimens included the following: 2 protease inhibitors (PIs), 2 NRTIs with 1 PI, 2 NRTIs with 1 NNRTI, or 3 NRTIs. In the models, values were taken at the beginning of the intervals for BMI, viral load, CD4+ cell counts, serum albumin level, REE per kilogram of FFM, and energy intake per kilogram of weight. Self-reported ART, symptom variables, duration of HAART use, poverty, and insecure housing were taken at the end of the interval to reflect what occurred during the interval. The missing indicator method18 was used to deal with 2 variables (REE per kilogram of FFM and insecure housing) that had 307 (7%) and 123 (3%) missing intervals of observations, respectively. This method was not used for variables with fewer missing values.

Multivariate models were fit, including all variables that were associated at the level of P < 0.20 in the bivariate analyses and forcing in calendar time. The models were fit using a data set that had complete observations for all included variables. A step-down approach was then used to determine the best-fitting model, including all variables that were independently associated with ≥5% weight loss or were confounders in the association between calendar time and ≥5% weight loss.

To determine whether the effect of calendar period differed among various subgroups of the population, each hypothesized determinant of weight loss was entered into separate models that included an interaction term with a binary indicator set to 1 if post-1997 and 0 otherwise. The generalized score statistic was used to assess the significance of interaction terms (P < 0.05).

All statistical analyses were carried out using the SAS statistical software package (version 9.0; SAS Institute, Cary, NC). GEE analyses were performed using PROC GENMOD and PROC MIXED in SAS.

Back to Top | Article Outline

RESULTS

Table 1 shows the baseline characteristics of participants who were included in the analysis. Participants included in the analysis differed somewhat from those who were excluded. Included participants were slightly older (mean age: 41 vs. 39 years) and more likely to be gay men (52% vs. 43%). The analysis group also tended to include more whites (58% vs. 47%) and fewer African-Americans (29% vs. 34%). Significantly fewer of the included participants had ever injected drugs (35% vs. 45%; P = 0.02). Overall, these differences are likely a result of the exclusion of particular segments of the study population who were more difficult to follow-up on a regular 6-month study visit schedule.

Table 1
Table 1
Image Tools

Three hundred seventy seven (53%) of 713 participants contributed a total of 589 (14%) of 4237 intervals during which there was 5% or more weight loss from the previous visit. Most of these participants (64%) experienced only 1 interval with ≥5% weight loss. Seventy-six participants (20%) experienced 2 intervals with ≥5% weight loss, 42 participants (11%) experienced 3 intervals with ≥5% weight loss, and 13 participants (3%) experienced 4 intervals of ≥5% weight loss. The average amount of weight lost across these intervals was 6.5 kg (range: 2.4-23.6 kg). The median percent weight loss across these intervals was 7.2% (IQR: 5.9%-9.7%), with a minimum of 5% and a maximum of 26%.

Table 2 shows the bivariate associations between calendar time and various potential confounders with the outcome being ≥5% weight loss from the previous visit. As shown, many variables were significantly associated with ≥5% weight loss in the unadjusted models. Female gender, ever IDU, poverty, higher levels of BMI, lower CD4+ cell counts, higher viral loads, low albumin levels, trying to lose weight, nausea, fever, cough, thrush, and higher levels of REE per kilogram of FFM were all significantly associated with a higher 6-month risk of weight loss, whereas white race, diarrhea, HAART use, NRTI use, and higher levels of caloric intake (per kilogram of weight) were significantly associated with a lower risk of weight loss. In these bivariate analyses, however, calendar period was not significantly associated with ≥5% weight loss.

Table 2
Table 2
Image Tools

In the final multivariate model, several covariates remained significantly independently associated with ≥5% weight loss. In addition, we found that the effect of calendar time on ≥5% weight loss was significantly modified by whether or not a participant reported that he or she had tried to lose weight since the previous visit (Table 3). As shown, there was a significant increase in the 6-month risk of ≥5% weight loss in the later HAART years (1998-2003) compared with the early HAART years (1995-1997), but only among participants who reported they were not trying to lose weight (RR = 1.52; 95% confidence interval (CI): 1.17-1.97). Among those who were trying to lose weight, there was no difference in the risk of ≥5% weight loss between calendar periods (RR = 0.84; P = 0.32). In addition, ever IDU, poverty, BMI >28 kg/m2, CD4 count ≤200 cells/μL, viral load >105 copies/mL, nausea, and thrush were significantly independently associated with a higher risk of ≥5% weight loss. Furthermore, report of any diarrhea was significantly associated with a lower risk of ≥5% weight loss. Adjustment for duration of HAART use did not affect the association between calendar period and risk of weight loss.

Table 3
Table 3
Image Tools

Figure 1 shows the relation between calendar time (as a continuous variable) and the 6-month risk of ≥5% weight loss only among those who were not trying to lose weight. As shown, the risk of ≥5% weight loss steadily increased on a log-linear scale between 1995 and 2004. With each passing year, the 6-month risk of ≥5% weight loss increased significantly by approximately 5% (P = 0.01) among those who were not trying to lose weight. The relation was the same after adjusting for IDU (ever vs. never), poverty, overweight or obese (BMI >25 kg/m2), low CD4+ cell count, high viral load, diarrhea, nausea, and thrush.

Figure 1
Figure 1
Image Tools

Table 4 compares the characteristics of weight loss intervals in the early HAART years with those of weight loss intervals in the later HAART years among the subset of intervals where weight loss was unintentional (ie, not trying to lose weight). There were no differences in mean weight, BMI, or triceps skinfold thickness at the start or the end (data not shown) of the weight loss intervals. There was a higher proportion of weight loss intervals in the overweight category of BMI (25 to <30 kg/m2) in the later HAART years than in the early HAART years (40% vs. 27%; P = 0.06). There were no differences in proportion of weight loss intervals in the obese category of BMI (BMI ≥30 kg/m2), however. The average amount of weight lost in the early HAART years (6.3 ± 0.2 kg) was virtually identical to the amount of weight lost in the later HAART years (6.3 ± 0.4 kg). Neither were there any significant differences between the 2 periods in the absolute amount of fat, FFM, BCM, or triceps skinfold thickness lost. As might be expected, mean CD4 counts were higher and mean viral loads were lower among weight loss intervals in the later calendar period. Furthermore, the proportion of intervals with any opportunistic infections was significantly lower in the later period compared with the early period. The prevalence of low albumin levels was higher in the later period than in the earlier period, although this difference was not statistically significant. There was no difference in the prevalence of any diarrhea by calendar period; however the prevalence of severe diarrhea (≥6 loose or watery stools per day) was slightly lower in the post-1997 period (3%) versus the earlier period (7%; P = 0.12). The prevalence of thrush decreased by approximately half from the early HAART era to the later HAART era. As expected, the prevalence of HAART and NNRTI use was higher in the later period; however, the use of NRTIs remained constant over time. No significant differences between the 2 periods were observed for gender, race, IDU, housing insecurity, poverty, nausea, fever, energy intake per kilogram of weight, or REE per kilogram of FFM.

Table 4
Table 4
Image Tools
Back to Top | Article Outline

DISCUSSION

Despite the widespread use of HAART, we found that the risk of 5% or more weight loss over an average period of 6.5 months steadily increased from 1995 to 2003 among most participants in our cohort who were not trying to lose weight. More specifically, among our cohort of HIV-positive participants followed since 1995, the 6-month risk of ≥5% unintentional weight loss was 52% higher in the later HAART years (1998-2003) compared with the pre- and/or early HAART years (1995-1997), after adjusting for several other significant determinants of weight loss. Furthermore, the 6-month risk of unintentional weight loss increased steadily, by approximately 5% per year, despite apparently adequate control of HIV infection as determined by lower HIV viral loads and higher CD4+ cell counts.

We examined the 6-month risk of weight loss by calendar period among the entire cohort and found that the increase in risk of weight loss with calendar time occurred only among most of the intervals (77%) when participants reported they were not trying to lose weight (ie, not dieting). Participants who reported that they were trying to lose weight (dieters) were different from those who were not trying to lose weight. Dieters were significantly more likely to be female and overweight or obese (BMI ≥25 kg/m2) and were more likely to report lower caloric intake. Although dieters were more likely to lose weight than nondieters (particularly in the earlier period), there was no increase in 6-month risk of ≥5% weight loss over time in this group.

The studies by Smit et al,8 Dworkin et al,9 and Mocroft et al10 mentioned previously all report declines in the clinical diagnosis of HIV wasting syndrome with the advent of HAART. As stated earlier, these diagnoses were based on the CDC definition of HIV-related wasting (ie, >10% loss of baseline body weight plus chronic diarrhea or chronic weakness and documented fever in the absence of a concurrent illness or condition other than HIV infection that could explain the findings). Because the amount of time since baseline is not specified in the definition, weight loss by the CDC definition could have occurred over varying periods. These studies have led to anecdotal reports that unintentional weight loss (or wasting) is no longer occurring or is no longer an important issue in HIV-infected patients. Our finding of an increasing risk of ≥5% unintentional weight loss over calendar time does not necessarily contradict the findings from these previous studies. Our findings are not based on a clinical diagnosis of wasting syndrome but, instead, on actual weight measurements taken in a standardized fashion at study visits approximately 6 months apart. Although there may be fewer diagnoses of HIV wasting syndrome by clinicians in recent years, including in our cohort, our analyses reveal that there is actually a steady increase in the 6-month risk of ≥5% unintentional weight loss with each passing year. The difference between these results is likely a result of the fact that dramatic declines (>10%) in weight are rarely seen anymore. In fact, in our cohort, the risk of more severe weight loss (≥10% over 6 months) was low (approximately 3%) and remained at this level throughout the study period.

In an attempt to distinguish generalized HIV-associated weight loss from the more recent syndrome of regional peripheral lipoatrophy commonly seen in HIV patients on HAART, we examined differences in the “look” of weight loss (among nondieters) between the 2 calendar periods. We found that weight loss intervals in the later HAART years were not different with respect to body composition than weight loss intervals in the earlier HAART years. Although there was a somewhat higher proportion of participants who were overweight in the later HAART period (P = 0.06), there were no significant differences in mean weight, BMI, or triceps skinfold thickness at the start of the weight loss intervals between the 2 periods. Furthermore, there were no differences in initial fat mass, FFM, or BCM among weight loss intervals by calendar period (data not shown). The change in body composition over the weight loss intervals did not differ between the 2 periods for any of the parameters we examined. We did not begin measuring waist and hip circumferences until September 2000; thus, we were unable to compare the prevalence of central adiposity between the 2 calendar periods.

Participants who lost weight in the later HAART era did differ from those who lost weight in the early HAART era in several ways. Weight loss intervals in the later period were characterized by significantly higher CD4+ cell counts, lower HIV viral loads, and fewer opportunistic infections, likely attributable to the benefits of HAART use in that period. Furthermore, thrush was less commonly seen in weight loss intervals in the later period. This corroborates our previous assertion that unintentional weight loss continues to occur despite better control of HIV infection.

In addition to calendar time, we found several other independent determinants of weight loss. Use of injection drugs, living below the federal poverty level, higher BMI (≥25 kg/m2), lower CD4+ cell count, higher HIV viral load, and presence of nausea or fever over the weight loss interval were all significantly associated with a higher risk of unintentional weight loss. Overall, our cohort, like the general US population, is slowly gaining weight over time. Among the overall cohort, the proportion of participants in the overweight category of BMI (25 to <30 kg/m2) was significantly higher in the post-1997 period than in the pre-1997 period (35% vs. 30%; P = 0.03). The proportion of participants in the obese category (BMI ≥30 kg/m2) did not differ significantly between calendar periods (16% post-1997 vs. 15% pre-1997; P = 0.58). Because BMI is positively associated with risk of weight loss, it could be inferred that the increase in 6-month risk of ≥5% weight loss is a result of the increase in the proportion of overweight participants over time. Our results showed an increase in risk of weight loss even after adjusting for the increase in BMI over time, however. In addition, there was no interaction between BMI and calendar time.

There are some limitations to these data. It is possible that calendar time (or calendar period) is a marker for another determinant of weight loss that we did not consider in this analysis. We considered potential confounders from every domain available to us in our data, however, including demographics, body composition, health status and symptoms, duration of HIV infection, HIV treatment, duration of HAART use, REE, and dietary intake, and found that none were associated with calendar time. Furthermore, we found that ever IDU was an independent determinant of weight loss. In our cohort, few of the injection drug users are currently using drugs (∼3%). Ever IDU in our cohort is thus more likely to be a surrogate marker of some other sociodemographic characteristic rather than a measure of the actual effect of drug use. Finally, as stated earlier, participants who were included in the analysis differed from those who were excluded, thereby limiting the generalizability of our findings.

In summary, we have found that the risk of ≥5% unintentional weight loss over 6-month intervals is on the rise in our cohort of HIV-infected participants. Previously, we have shown that weight loss of this magnitude is associated with a significantly increased risk of mortality among HAART-treated HIV-infected participants.6 Therefore, attention to weight loss in the current HIV climate, where patients are maintaining more normal nutritional status, remains important so as to maximize care. This is particularly important among segments of the HIV patient population who have symptoms of illness, have less control of the virus, or are of lower socioeconomic status.

Back to Top | Article Outline

REFERENCES

1. Coodley GO, Loveless MO, Merrill TM. The HIV wasting syndrome: a review. J Acquir Immune Defic Syndr Hum Retrovirol. 1994;7:681-694.

2. Nahlen BL, Chu SY, Nwanyanwu OC, et al. HIV wasting syndrome in the U.S. AIDS. 1993;7:183-188.

3. Suttmann U, Ockenga J, Selberg O, et al. Incidence and prognostic value of malnutrition and wasting in human immunodeficiency virus-infected outpatients. J Acquir Immune Defic Syndr Hum Retrovirol. 1995;8:239-246.

4. Moore RD, Chaisson RE. Natural history of HIV infection in the era of combination antiretroviral therapy. AIDS. 1999;13:1933-1942.

5. Wanke CA, Silva M, Knox TA, et al. Weight loss and wasting remain common complications in individuals infected with human immunodeficiency virus in the era of highly active antiretroviral therapy. Clin Infect Dis. 2000;31:803-805.

6. Tang AM, Forrester J, Spiegelman D, et al. Weight loss and survival in HIV-positive patients in the era of highly active antiretroviral therapy. J Acquir Immune Defic Syndr. 2002;31:230-236.

7. Centers for Disease Control and Prevention. Revision of the CDC surveillance case definition for acquired immunodeficiency syndrome. MMWR Morb Mortal Wkly Rep. 1987;36(Suppl):1S-15S.

8. Smit E, Skolasky RL, Dobs AS, et al. Changes in the incidence and predictors of wasting syndrome related to human immunodeficiency virus infection, 1987-1999. Am J Epidemiol. 2002;156:211-218.

9. Dworkin MS, Williamson JM. AIDS wasting syndrome: trends, influence on opportunistic infections, and survival. J Acquir Immune Defic Syndr. 2003;33:267-273.

10. Mocroft A, Sabin CA, Youle M, et al. Changes in AIDS-defining illnesses in a London clinic, 1987-1998. J Acquir Immune Defic Syndr. 1999;21:401-407.

11. Wheeler DA, Gibert CL, Launer CA, et al. The Terry Beirn Community Programs for Clinical Research on AIDS. Weight loss as a predictor of survival and disease progression in HIV infection. J Acquir Immune Defic Syndr Hum Retrovirol. 1998;18:80-85.

12. Shevitz AH, Knox TA, Spiegelman D, et al. Elevated resting energy expenditure among HIV-seropositive persons receiving highly active antiretroviral therapy. AIDS. 1999;13:1351-1357.

13. Silva M, Skolnik PR, Gorbach SL, et al. The effect of protease inhibitors on weight and body composition in HIV-infected patients. AIDS. 1998;12:1645-1651.

14. Wilson IB, Roubenoff R, Knox TA, et al. Relation of lean body mass to health-related quality of life in persons with HIV. J Acquir Immune Defic Syndr. 2000;24:137-146.

15. Lukaski HC. Use of bioelectrical impedance analysis to assess human body composition: a review. In: Livingston GE, ed. Nutritional Status Assessment of the Individual. Trumbull, CT: Food & Nutrition Press; 1989:189-204.

16. Kotler DP, Burastero S, Wang J, et al. Prediction of body cell mass, fat-free mass, and total body water with bioelectrical impedance analysis: effects of race, sex, and disease. Am J Clin Nutr. 1996;64(Suppl):489S-497S.

17. Durrleman S, Simon R. Flexible regression models with cubic splines. Stat Med. 1989;8:551-561.

18. Jones MP. Indicator and stratification methods for missing explanatory variables in multiple linear regression. J Am Stat Assoc. 1996;91:222-230.

Cited By:

This article has been cited 22 time(s).

Cochrane Database of Systematic Reviews
Megestrol acetate for treatment of anorexia-cachexia syndrome
Garcia, VR; Lopez-Briz, E; Sanchis, RC; Perales, JLG; Bort-Marti, S
Cochrane Database of Systematic Reviews, (3): -.
ARTN CD004310
CrossRef
Janac-Journal of the Association of Nurses in AIDS Care
HIV-Associated Wasting
Keithley, JK; Swanson, B
Janac-Journal of the Association of Nurses in AIDS Care, 24(1): S103-S111.
10.1016/j.jana.2012.06.013
CrossRef
Drugs
Mammalian cell-derived somatropin - A review of its use in the management of HIV-associated wasting
Goldsmith, DR; Wagstaff, AJ
Drugs, 66(3): 387-401.

Journal of Clinical Endocrinology & Metabolism
Testosterone supplementation of megestrol therapy does not enhance lean tissue accrual in men with human immunodeficiency virus-associated weight loss: A randomized, double-blind, placebo-controlled, multicenter trial
Mulligan, K; Zackin, R; Von Roenn, JH; Chesney, MA; Egorin, MJ; Sattler, FR; Benson, CA; Liu, T; Umbleja, T; Shriver, S; Auchus, RJ; Schambelan, M
Journal of Clinical Endocrinology & Metabolism, 92(2): 563-570.
10.1210/jc.2006-0954
CrossRef
Public Health Nutrition
Declines in dietary macronutrient intake in persons with HIV infection who develop depression
Isaac, R; Jacobson, D; Wankel, C; Hendricks, K; Knox, TA; Wilson, IB
Public Health Nutrition, 11(2): 124-131.
10.1017/S1368980007000067
CrossRef
American Journal of Clinical Nutrition
Micronutrients in HIV-positive persons receiving highly active antiretroviral therapy
Drain, PK; Kupka, R; Mugusi, F; Fawzi, WW
American Journal of Clinical Nutrition, 85(2): 333-345.

Clinical Therapeutics
Role of recombinant human growth hormone in HIV-Associated wasting and cachexia: Pathophysiology and rationale for treatment
Gelato, M; McNurlan, M; Freedland, E
Clinical Therapeutics, 29(): 2269-2288.
10.1016/j.clinthera.2007.11.004
CrossRef
American Journal of Physiology-Endocrinology and Metabolism
Skeletal and cardiac myopathy in HIV-1 transgenic rats
Pruznak, AM; Hong-Brown, L; Lantry, R; She, PX; Frost, RA; Vary, TC; Lang, CH
American Journal of Physiology-Endocrinology and Metabolism, 295(4): E964-E973.
10.1152/ajpendo.90482.2008
CrossRef
Clinical Infectious Diseases
Nutrition and HIV infection: Review of weight loss and wasting in the era of highly active antiretroviral therapy from the Nutrition for Healthy Living cohort
Mangili, A; Murman, DH; Zampini, AM; Wanke, CA
Clinical Infectious Diseases, 42(6): 836-842.

American Journal of Clinical Nutrition
Evaluation of high-protein supplementation in weight-stable HIV-positive subjects with a history of weight loss: a randomized, double-blind, multicenter trial
Sattler, FR; Rajicic, N; Mulligan, K; Yarasheski, KE; Koletar, SL; Zolopa, A; Alston, B; Zackin, SR; Bistrian, B
American Journal of Clinical Nutrition, 88(5): 1313-1321.
10.3945/ajcn.2006.23583
CrossRef
Plos One
Increasing Rates of Obesity among HIV-Infected Persons during the HIV Epidemic
Crum-Cianflone, N; Roediger, MP; Eberly, L; Headd, M; Marconi, V; Ganesan, A; Weintrob, A; Barthel, RV; Fraser, S; Agan, BK
Plos One, 5(4): -.
ARTN e10106
CrossRef
Nursing Clinics of North America
The impact of nutrition on physiologic changes in persons who have HIV
Thomas, AM; Mkandawire, SC
Nursing Clinics of North America, 41(3): 455-+.
10.1016/j.cnur.2006.05.002
CrossRef
Janac-Journal of the Association of Nurses in AIDS Care
HIV Infection and Obesity: A Review of the Evidence
Keithley, JK; Duloy, AMS; Swanson, B; Zeller, JM
Janac-Journal of the Association of Nurses in AIDS Care, 20(4): 260-274.
10.1016/j.jana.2009.02.006
CrossRef
Drug and Alcohol Dependence
Drug use and other risk factors related to lower body mass index among HIV-infected individuals
Quach, LA; Wanke, CA; Schmid, CH; Gorbach, SL; Mwamburi, DM; Mayer, KH; Spiegelman, D; Tang, AM
Drug and Alcohol Dependence, 95(): 30-36.
10.1016/j.drugalcdep.2007.12.004
CrossRef
AIDS Care-Psychological and Socio-Medical Aspects of AIDS/Hiv
Drug use and weight loss in HIV-infected Hispanic men
Forrester, JE; Tucker, KL; Skinner, S; Terrin, N
AIDS Care-Psychological and Socio-Medical Aspects of AIDS/Hiv, 20(7): 868-875.
10.1080/09540120701767174
CrossRef
Clinical Infectious Diseases
Weight and body shape changes in a treatment-naive population after 6 months of nevirapine-based generic highly active antiretroviral therapy in South India
Saghayam, S; Kumarasamy, N; Cecelia, AJ; Solomon, S; Mayer, K; Wanke, C
Clinical Infectious Diseases, 44(2): 295-300.

Current Medical Research and Opinion
Prevalence and cost of HIV-associated weight loss in a managed care population
Siddiqui, J; Phillips, AL; Freedland, ES; Sklar, AR; Darkow, T; Harley, CR
Current Medical Research and Opinion, 25(5): 1307-1317.
10.1185/03007990902902119
CrossRef
American Journal of Drug and Alcohol Abuse
Body composition, gender, and illicit drug use in an urban cohort
Cofrancesco, J; Brown, TT; Luo, RF; John, M; Stewart, KJ; Dobs, AS
American Journal of Drug and Alcohol Abuse, 33(3): 467-474.
10.1080/00952990701301616
CrossRef
Epidemiology and Infection
Markov modelling of HIV infection evolution in the HAART era
Binquet, C; le Teuff, G; Abrahamovicz, M; Mahboubi, A; Yazdanpanah, Y; Rey, D; Rabaud, C; Chirouze, C; Berger, JL; Faller, JP; Chavanet, P; Quantin, C; Piroth, L
Epidemiology and Infection, 137(9): 1272-1282.
10.1017/S0950268808001775
CrossRef
Bmc Infectious Diseases
Predictors of early death in a cohort of Ethiopian patients treated with HAART
Jerene, D; Endale, A; Hailu, Y; Lindtjorn, B
Bmc Infectious Diseases, 6(): -.
ARTN 136
CrossRef
Psychophysiology
Psychiatric and neurophysiological predictors of obesity in HIV/AIDS
Bauer, LO
Psychophysiology, 45(6): 1055-1063.
10.1111/j.1469-8986.2008.00706.x
CrossRef
JAIDS Journal of Acquired Immune Deficiency Syndromes
Evolution and Predictors of Change in Total Bone Mineral Density Over Time in HIV-Infected Men and Women in the Nutrition for Healthy Living Study
Jacobson, DL; Spiegelman, D; Knox, TK; Wilson, IB
JAIDS Journal of Acquired Immune Deficiency Syndromes, 49(3): 298-308.
10.1097/QAI.0b013e3181893e8e
PDF (127) | CrossRef
Back to Top | Article Outline
Keywords:

HIV; weight loss; HIV wasting syndrome; calendar time

© 2005 Lippincott Williams & Wilkins, Inc.

Login

Search for Similar Articles
You may search for similar articles that contain these same keywords or you may modify the keyword list to augment your search.