Fish Oil vs. Drugs for Hypertriglyceridemia in HIV+ Patients
Fish Oil vs. Drugs for Hypertriglyceridemia in HIV+ Patients
Fish oil led to a significant reduction in TG values. Fibrates were associated with greater reductions in TG values compared with fish oil in HIV-infected patients in routine clinical care. The reduction in TG values associated with atorvastatin compared with fish oil was not statistically significant.
Fenofibrate and gemfibrozil demonstrated similar substantial reductions in TG values. Gemfibrozil was used less frequently, suggesting that clinicians preferentially use fenofibrate, consistent with general population trends across the United States for much greater fenofibrate use without increasing use of gemfibrozil. Fenofibrate may have a lower risk of rhabdomyolysis than gemfibrozil in combination with statins and the advantage of daily dosing versus twice daily. Of the other statins commonly used in HIV-infected patients, pravastatin and rosuvastatin, the latter has previously been shown to have similar reductions in TG values compared with atorvastatin, whereas pravastatin has less reduction in TG values. Although niacin is another treatment option for hypertriglyceridemia, it was not included in this study because of its rare use and small sample size within the CNICS Cohort.
This study demonstrates the difficulty in achieving TG goals with monotherapy in HIV-infected patients. The reduced response to dyslipidemia therapies in HIV-infected patients has been demonstrated previously. Only 23% achieved TG < 150 mg/dL posttreatment. This number would be lower if patients with TG >810 mg/dL had not been excluded, as these patients are unlikely to reach goal with monotherapy.
Although combination therapy was not the focus of this study, fish oil represents a particularly attractive option, as it is well tolerated and seems devoid of drug-interactions. Guidelines recommend caution when using statins and fibrates in combination because of the increased risk of toxicity, however in clinical practice their combined use is not uncommon. Statin therapy has the strongest evidence for reducing cardiovascular risk, whereas the evidence supporting fibrate and fish oil therapy is moderate. Clinical trials have shown conflicting results for cardiovascular risk reduction with fibrates, but there seems to be a benefit in patients with high TGs and low HDL cholesterol. Fish oil has demonstrated efficacy in reducing cardiovascular risk, particularly as secondary prevention. Additional studies are needed to determine the impact of fish oil on reducing TG values in combination with other medications among those individuals who do not reach their targets on monotherapy.
This study found a much smaller reduction in TG values associated with fish oil than earlier trials which demonstrated reductions ranging from 15% to 46%8–11,30 Although this may be due in part to differences in patient populations and the extent of baseline hypertriglyceridemia in participants, we suspect that much of this difference is because of the way fish oil is used in clinical care settings compared with clinical trials. Dosage information was not available for the majority of patients included, however, among the subset of patients taking fish oil in the entire CNICS cohort (N = 720) with complete dosage data (N = 97, 14%) 61% of patients had a dose of 2 g or less and only 2% had a dose of 4 g or greater. In contrast, while data were also incomplete, among those with doses for gemfibrozil and fenofibrate, most were on standard or high doses (eg, 600 mg of gemfibrozil two times a day, varies based on brand for fenofibrate). Mean atorvastatin dose was 22 mg. Doses used in randomized clinical trials of HIV-infected patients have typically used higher doses, 3–6 g, compared with the doses used in clinical care in this cohort. Furthermore, a dose effect with greater impact from higher doses has previously been suggested. The relatively low doses used in clinical care suggest that the effectiveness of fish oil in this study may be underestimated but reflective of how it is used in clinical practice. Variations in the amount of omega-3 fatty acids contained in the various fish oil brands likely also contributed to the findings.
The range of eligible baseline TG values was restricted to maintain comparability across treatments; however, baseline TG values were still significantly higher in patients receiving fibrates. This may underestimate the comparative effects of fish oil and atorvastatin because the baseline TG value is an effect measure modifier. Previous randomized trials have demonstrated that the higher the baseline TG value is, the greater the therapeutic response.
Strengths of this study include the large sample size, geographic diversity, and comprehensive clinical data. This study examined the change in TG levels associated with individual therapies in clinical care with minimal exclusion criteria. Our findings are generalizable to similar populations of patients in routine care who represent a broader range of characteristics than patients who typically enroll in clinical trials. The restriction of baseline TG values <810 mg/dL resulted in baseline TG values of approximately 350 mg/dL, therefore, the generalizability of these findings is limited to those with moderate hypertriglyceridemia.
As with any observational study, there may be unmeasured confounders for which adjustment is not possible. The study lacks information regarding adherence, genetic factors, diet, and exercise. The advantage of the pre/post design is that it controls for within-subject time independent confounders. Lipid values were measured in clinical care so we could not confirm fasting status but have no reason to suspect that fasting status would vary between therapies. In addition, despite the greater variability in nonfasting TG values, their importance is increasingly appreciated given their association with cardiovascular disease (CVD). Contradictory to a previous pharmacokinetic study, gemfibrozil showed a greater reduction in TG values in patients receiving a PI-based regimen. The sample size of the study population did not allow inference to be made when hypertriglyceridemia treatments were stratified by ART classes. Furthermore, it is well known that antiretroviral medications differ in their effects on lipid profiles not only between classes but also within antiretroviral medication classes. Finally, because fish oil is available over the counter, it is certainly possible that additional patients were taking fish oil and this was not captured. Most, but not all of the fish oil captured was prescription fish oil, so findings may be somewhat less generalizable to over-the-counter fish oil use.
Another limitation was that the doses of medications were not always available. It is possible that the doses of fenofibrate, fish oil, and atorvastatin used were lower than optimal and may have been titrated after the posttreatment results. This concern is less likely with gemfibrozil because there is only 1 commonly used dose. However, the point of this study was to examine the effectiveness of TG-lowering medications as they are actually prescribed in clinical care. We focused on the initial time period after initiation of these medications because clinical inertia and lack of medication titration is a frequent occurrence and a shorter time frame lessened the likelihood that large changes in BMI or other factors would occur and impact results. Finally, this study focused only on the impact on TG levels. In addition to lowering TG levels, these medications may have other potential impacts, such as changes in low-density lipoprotein subfractions, which require further study.
Discussion
Fish oil led to a significant reduction in TG values. Fibrates were associated with greater reductions in TG values compared with fish oil in HIV-infected patients in routine clinical care. The reduction in TG values associated with atorvastatin compared with fish oil was not statistically significant.
Fenofibrate and gemfibrozil demonstrated similar substantial reductions in TG values. Gemfibrozil was used less frequently, suggesting that clinicians preferentially use fenofibrate, consistent with general population trends across the United States for much greater fenofibrate use without increasing use of gemfibrozil. Fenofibrate may have a lower risk of rhabdomyolysis than gemfibrozil in combination with statins and the advantage of daily dosing versus twice daily. Of the other statins commonly used in HIV-infected patients, pravastatin and rosuvastatin, the latter has previously been shown to have similar reductions in TG values compared with atorvastatin, whereas pravastatin has less reduction in TG values. Although niacin is another treatment option for hypertriglyceridemia, it was not included in this study because of its rare use and small sample size within the CNICS Cohort.
This study demonstrates the difficulty in achieving TG goals with monotherapy in HIV-infected patients. The reduced response to dyslipidemia therapies in HIV-infected patients has been demonstrated previously. Only 23% achieved TG < 150 mg/dL posttreatment. This number would be lower if patients with TG >810 mg/dL had not been excluded, as these patients are unlikely to reach goal with monotherapy.
Although combination therapy was not the focus of this study, fish oil represents a particularly attractive option, as it is well tolerated and seems devoid of drug-interactions. Guidelines recommend caution when using statins and fibrates in combination because of the increased risk of toxicity, however in clinical practice their combined use is not uncommon. Statin therapy has the strongest evidence for reducing cardiovascular risk, whereas the evidence supporting fibrate and fish oil therapy is moderate. Clinical trials have shown conflicting results for cardiovascular risk reduction with fibrates, but there seems to be a benefit in patients with high TGs and low HDL cholesterol. Fish oil has demonstrated efficacy in reducing cardiovascular risk, particularly as secondary prevention. Additional studies are needed to determine the impact of fish oil on reducing TG values in combination with other medications among those individuals who do not reach their targets on monotherapy.
This study found a much smaller reduction in TG values associated with fish oil than earlier trials which demonstrated reductions ranging from 15% to 46%8–11,30 Although this may be due in part to differences in patient populations and the extent of baseline hypertriglyceridemia in participants, we suspect that much of this difference is because of the way fish oil is used in clinical care settings compared with clinical trials. Dosage information was not available for the majority of patients included, however, among the subset of patients taking fish oil in the entire CNICS cohort (N = 720) with complete dosage data (N = 97, 14%) 61% of patients had a dose of 2 g or less and only 2% had a dose of 4 g or greater. In contrast, while data were also incomplete, among those with doses for gemfibrozil and fenofibrate, most were on standard or high doses (eg, 600 mg of gemfibrozil two times a day, varies based on brand for fenofibrate). Mean atorvastatin dose was 22 mg. Doses used in randomized clinical trials of HIV-infected patients have typically used higher doses, 3–6 g, compared with the doses used in clinical care in this cohort. Furthermore, a dose effect with greater impact from higher doses has previously been suggested. The relatively low doses used in clinical care suggest that the effectiveness of fish oil in this study may be underestimated but reflective of how it is used in clinical practice. Variations in the amount of omega-3 fatty acids contained in the various fish oil brands likely also contributed to the findings.
The range of eligible baseline TG values was restricted to maintain comparability across treatments; however, baseline TG values were still significantly higher in patients receiving fibrates. This may underestimate the comparative effects of fish oil and atorvastatin because the baseline TG value is an effect measure modifier. Previous randomized trials have demonstrated that the higher the baseline TG value is, the greater the therapeutic response.
Strengths and Limitations
Strengths of this study include the large sample size, geographic diversity, and comprehensive clinical data. This study examined the change in TG levels associated with individual therapies in clinical care with minimal exclusion criteria. Our findings are generalizable to similar populations of patients in routine care who represent a broader range of characteristics than patients who typically enroll in clinical trials. The restriction of baseline TG values <810 mg/dL resulted in baseline TG values of approximately 350 mg/dL, therefore, the generalizability of these findings is limited to those with moderate hypertriglyceridemia.
As with any observational study, there may be unmeasured confounders for which adjustment is not possible. The study lacks information regarding adherence, genetic factors, diet, and exercise. The advantage of the pre/post design is that it controls for within-subject time independent confounders. Lipid values were measured in clinical care so we could not confirm fasting status but have no reason to suspect that fasting status would vary between therapies. In addition, despite the greater variability in nonfasting TG values, their importance is increasingly appreciated given their association with cardiovascular disease (CVD). Contradictory to a previous pharmacokinetic study, gemfibrozil showed a greater reduction in TG values in patients receiving a PI-based regimen. The sample size of the study population did not allow inference to be made when hypertriglyceridemia treatments were stratified by ART classes. Furthermore, it is well known that antiretroviral medications differ in their effects on lipid profiles not only between classes but also within antiretroviral medication classes. Finally, because fish oil is available over the counter, it is certainly possible that additional patients were taking fish oil and this was not captured. Most, but not all of the fish oil captured was prescription fish oil, so findings may be somewhat less generalizable to over-the-counter fish oil use.
Another limitation was that the doses of medications were not always available. It is possible that the doses of fenofibrate, fish oil, and atorvastatin used were lower than optimal and may have been titrated after the posttreatment results. This concern is less likely with gemfibrozil because there is only 1 commonly used dose. However, the point of this study was to examine the effectiveness of TG-lowering medications as they are actually prescribed in clinical care. We focused on the initial time period after initiation of these medications because clinical inertia and lack of medication titration is a frequent occurrence and a shorter time frame lessened the likelihood that large changes in BMI or other factors would occur and impact results. Finally, this study focused only on the impact on TG levels. In addition to lowering TG levels, these medications may have other potential impacts, such as changes in low-density lipoprotein subfractions, which require further study.
