madman
Super Moderator
ABSTRACT
Objective
Elevated hematocrit (Hct) can result in increased risk of major adverse cardiac events(MACE) in men receiving testosterone therapy (TTh). However, the impact of the magnitude of the change in Hct from baseline after starting TTh has never been assessed.
Methods
To assess whether an increase in Hct after initiating TTh is associated with an increased risk of MACE within 3 and 24 months of initiating TTh, we queried the TriNetX Research network database for men over the age of 18 with Hct values obtained within six months before starting TTh, and who had follow-up Hct measurements within three and 24 months after beginning TTh from 2010-2021. Men with and without a subsequent increase in Hct after initiating TTh were propensity matched. MACE was defined as myocardial infarction,stroke, or death.
Results
After matching, 10,511 men who experienced an any increase in Hct after initiating TTh and an equal number of controls who did have an increase in Hct were included. Compared to controls who did not have an increase in Hct after starting TTh, the men who had an increase in subsequent Hct had a significantly increased risk of major adverse cardiac events compared with men with no change in Hct.
Conclusions
We demonstrate that increases in Hct from baseline are associated with increased risk of MACE, compared to men whose Hct remains stable while receiving TTh.
Impact Statement
Clinicians should monitor Hct after initiating TTh, as large increase from baseline may be a risk factor for MACE.
INTRODUCTION
Testosterone therapy (TTh) is recommended for men with symptoms of hypogonadism associated with testosterone deficiency.1 Indications for TTh include delayed puberty, sexual dysfunction, hypopituitarism, and adult men with testosterone levels below an age-specific physiological range exhibiting corresponding symptoms. The primary goal of TTh is to reduce clinical symptoms and restore testosterone levels to age-appropriate ranges.2
However, TTh is associated with an increase in hemoglobin (Hb) and hematocrit (Hct) levels. The Endocrine society recommends against starting therapy in patients with elevated baseline Hct due to concerns about an increased risk of major adverse cardiovascular events (MACE) including myocardial infarctions and cerebrovascular accidents potentially due to increase in blood viscosity.1, 3-5 TTh is postulated to affect Hct by multifactorial mechanisms including decreasing hepcidin production, increasing erythropoietin setpoint and production, andactivating peripheral bone marrow estrogen receptors leading to telomere stability.6 Depending on the different societies, Hct levels of 48% to 55% are regarded as contraindications for TTh given concerns of increased risk of MACE.1, 2, 5, 7, 8
While recent studies have shown that TTh-induced secondary polycythemia (Hct > 52%) is an independent risk factor for MACE within the first year of treatment 9, it is not clear whether erythrocytosis – i.e., degree of change in Hct from baseline Hct after starting TTh confers additional risk for MACE. Herein we aimed to determine the incidence of MACE associated with the degree of Hct change from baseline in adult men receiving testosterone therapy for hypogonadism.
DISCUSSION
The link between TTh and MACE is actively debated, with recent evidence supporting that erythrocytosis may be an important intermediary between these two.9 While guidelines surrounding erythrocytosis have thus far involved an absolute upper limit cutoff for safety, studies have not investigated whether the magnitude of change in Hct while on TTh is a risk for MACE. We separated Hct into 5 separate quintiles and investigated how risks of MACE changed as Hct increased through these ranges while on TTh compared to men who did not experience an increase. We found that increases in Hct conferred an subsequent risk of MACE, regardless of the final upper limit. The risk of MACE was highest in those men who underwent the largest increases in Hct.
This study has several important implications. Until recently, studies that have informed FDA warnings, or those that have showed TTh to be safe with respect to MACE have not looked at an underlying cause for the association. Recently, Hct over 52% has been found to be an important intermediary linking TTh and MACE.9This study looked at a simple cutoff of 52%, and so did not provide any information on whether a large increase in Hct, while still remaining under 52%, was important. Our study suggests that all men, regardless of initial Hct, can potentially be at risk for MACE if their Hct increases. Men can have a Hct below guideline based cutoffs, but if the overall increase is large enough despite not reaching the upper limit o fcutoff for changing dose or therapeutic phlebotomy, they may still experience risk for an adverse event. This may mean more careful tracking of Hct while on TTh instead of simply looking for an upper limit during the first 2 years of therapy.
It is not known why an increase in Hct from any baseline level may increase the risk of MACE. Clinical evidence does support that the change in baseline may be more important than the final upper limit. Walker et al. found that men on TTh had a higher risk of VTE during the first 6 months of TTh. On exploratory analysis, they found this risk was highest in the first 3 months of therapy, with no increased risk between month 3 and 6.14 Upon starting TTh, Hct begins to rise within the first month of therapy, and peaks at month 3.15 This increased risk of VTE within the first 3 months then may be explained by the rise in Hct. Other research has found that the rate of MACE is higher in the first year of TTh, which may further support that the rise in Hct is more important than final value.16
Rates of MACE were greatest in the cohort with a baseline Hct of 36-39% when compared to all other cohorts but this is likely explained by the more unhealthy individuals in this cohort. For example in the cohort with baseline Hct of 36-39% that stayed in this range after starting TTh, rates of hypertension were 51.5%, hyperlipidemia was 45.1%, type II diabetes 30.7%, prior stroke was 4.1%. Compared to men with a baseline Hct of 44-47% that stayed in this range after starting TTh: hypertension was 38.5%, hyperlipidemia was 44.0%, type II diabetes 15.8%, prior stroke was 0.74%. Thus the higher overall frequency of MACE is likely explained by a more unhealthy cohort for those with Hct of 36-39%. This is why each cohort is propensity matched to its own same baseline Hct values to demonstrate the risk associated with each increases in Hct quintile compared to men who saw no increase in Hct after starting TTh.
One of the major strengths of our study is the number of patients we were able to compare. One of the main criticisms of studies linking TTh and MACE is that they are not adequately powered to show causation.17 With over 16,000 men in our analysis after matching,we were able to have adequate power for our primary outcome. Another strength of our study is separating Hct into quintiles and using this to investigate an exposure-response relationship between Hct and MACE. While a simple cutoff of 52% has been shown to be correlated with increased risk of MACE, finding an exposure-response relationship further strengthens the plausibility that Hct increases are an important factor in MACE risk in men on TTh. One of the major limitations of our study is that due to using an anonymized database, we lose the ability to investigate individual patient factors in the analysis including testosterone preparation used,compliance, or dosages as well as patient diet or exercise routines at baseline. Unfortunately,because the TriNetX database obfuscates any counts where the incidence is 10 or less to protect HIPAA, we are unable to obtain precise counts of the sub-categories of MACE (stroke, MI, and death). Regardless, because we still do show a consistent increase in every starting quintile of Hct, we believe our findings still are intriguing. Because some cohorts that experienced an increase in Hct after starting testosterone had smaller numbers of patients compared to the cohort not experiencing an increase and other times, those experiencing an increase had larger numbers than those not experiencing an increase, we could not consistently keep the target population for propensity score modeling to only those experiencing an increase and this could limit the generalizability of our study. TriNetX database also only allows for propensity score matching rather than regression analysis limiting our analytical approach and requiring categorical treatment of Hct values rather than as a continuous variable. Additionally, we do not know what interventions were undertaken upon development of erythrocytosis, or if these men were appropriately diagnosed with hypogonadism. More and more clinics are offering TTh without proper guideline-based diagnosis which may increase their risk for adverse effects.18-20 Lastly,we were not able to determine which testosterone modalities these men were on. Since longer acting testosterone modalities have a higher risk of erythrocytosis than short-acting modalities, itis theoretically possible that these would also have a higher risk of MACE, but this has not been investigated yet.21, 22 Finally, assessing whether a decline in Hct after starting testosterone is or is not associated with future MACE would have been intriguing, unfortunately, too few men saw a decrease in Hct upon starting testosterone to allow for analysis of these cohorts.
This study expands the likelihood that increases in Hct account for the incidence of MACE possibly seen in men on TT. A large RCT (TRAVERSE Trial) is currently being undertaken to further elucidate this connection but unfortunately has not included Hct-based increases as a factor in their analysis. While the decision on how or when to act upon a rising Hct once starting testosterone therapy is a conversation that patients and their doctors will have to consider. Certainly, these men who are experiencing a rise in Hct could be encouraged to talk with their primary care physician to determine if any other steps could help to decrease their risk of MACE – such as testing for hyperlipidemia or starting treatment if hypertensive. Further research is needed to determine what amount of Hct increase should raise concerns amongst clinicians to warrant intervention.
CONCLUSIONS
In this study, we find that an increase in Hct from baseline after starting testosterone therapy is associated with an increased risk of MACE even when the increase is below traditional cut-off values, such as 52%. Clinicians should monitor Hct levels after initiating testosterone recognizing and counseling patients that increases in Hct may place patients a thigher risk of MACE. Further studies are needed to confirm these findings and to assess when clinicians should intervene when Hct is rising after starting testosterone therapy.
Objective
Elevated hematocrit (Hct) can result in increased risk of major adverse cardiac events(MACE) in men receiving testosterone therapy (TTh). However, the impact of the magnitude of the change in Hct from baseline after starting TTh has never been assessed.
Methods
To assess whether an increase in Hct after initiating TTh is associated with an increased risk of MACE within 3 and 24 months of initiating TTh, we queried the TriNetX Research network database for men over the age of 18 with Hct values obtained within six months before starting TTh, and who had follow-up Hct measurements within three and 24 months after beginning TTh from 2010-2021. Men with and without a subsequent increase in Hct after initiating TTh were propensity matched. MACE was defined as myocardial infarction,stroke, or death.
Results
After matching, 10,511 men who experienced an any increase in Hct after initiating TTh and an equal number of controls who did have an increase in Hct were included. Compared to controls who did not have an increase in Hct after starting TTh, the men who had an increase in subsequent Hct had a significantly increased risk of major adverse cardiac events compared with men with no change in Hct.
Conclusions
We demonstrate that increases in Hct from baseline are associated with increased risk of MACE, compared to men whose Hct remains stable while receiving TTh.
Impact Statement
Clinicians should monitor Hct after initiating TTh, as large increase from baseline may be a risk factor for MACE.
INTRODUCTION
Testosterone therapy (TTh) is recommended for men with symptoms of hypogonadism associated with testosterone deficiency.1 Indications for TTh include delayed puberty, sexual dysfunction, hypopituitarism, and adult men with testosterone levels below an age-specific physiological range exhibiting corresponding symptoms. The primary goal of TTh is to reduce clinical symptoms and restore testosterone levels to age-appropriate ranges.2
However, TTh is associated with an increase in hemoglobin (Hb) and hematocrit (Hct) levels. The Endocrine society recommends against starting therapy in patients with elevated baseline Hct due to concerns about an increased risk of major adverse cardiovascular events (MACE) including myocardial infarctions and cerebrovascular accidents potentially due to increase in blood viscosity.1, 3-5 TTh is postulated to affect Hct by multifactorial mechanisms including decreasing hepcidin production, increasing erythropoietin setpoint and production, andactivating peripheral bone marrow estrogen receptors leading to telomere stability.6 Depending on the different societies, Hct levels of 48% to 55% are regarded as contraindications for TTh given concerns of increased risk of MACE.1, 2, 5, 7, 8
While recent studies have shown that TTh-induced secondary polycythemia (Hct > 52%) is an independent risk factor for MACE within the first year of treatment 9, it is not clear whether erythrocytosis – i.e., degree of change in Hct from baseline Hct after starting TTh confers additional risk for MACE. Herein we aimed to determine the incidence of MACE associated with the degree of Hct change from baseline in adult men receiving testosterone therapy for hypogonadism.
DISCUSSION
The link between TTh and MACE is actively debated, with recent evidence supporting that erythrocytosis may be an important intermediary between these two.9 While guidelines surrounding erythrocytosis have thus far involved an absolute upper limit cutoff for safety, studies have not investigated whether the magnitude of change in Hct while on TTh is a risk for MACE. We separated Hct into 5 separate quintiles and investigated how risks of MACE changed as Hct increased through these ranges while on TTh compared to men who did not experience an increase. We found that increases in Hct conferred an subsequent risk of MACE, regardless of the final upper limit. The risk of MACE was highest in those men who underwent the largest increases in Hct.
This study has several important implications. Until recently, studies that have informed FDA warnings, or those that have showed TTh to be safe with respect to MACE have not looked at an underlying cause for the association. Recently, Hct over 52% has been found to be an important intermediary linking TTh and MACE.9This study looked at a simple cutoff of 52%, and so did not provide any information on whether a large increase in Hct, while still remaining under 52%, was important. Our study suggests that all men, regardless of initial Hct, can potentially be at risk for MACE if their Hct increases. Men can have a Hct below guideline based cutoffs, but if the overall increase is large enough despite not reaching the upper limit o fcutoff for changing dose or therapeutic phlebotomy, they may still experience risk for an adverse event. This may mean more careful tracking of Hct while on TTh instead of simply looking for an upper limit during the first 2 years of therapy.
It is not known why an increase in Hct from any baseline level may increase the risk of MACE. Clinical evidence does support that the change in baseline may be more important than the final upper limit. Walker et al. found that men on TTh had a higher risk of VTE during the first 6 months of TTh. On exploratory analysis, they found this risk was highest in the first 3 months of therapy, with no increased risk between month 3 and 6.14 Upon starting TTh, Hct begins to rise within the first month of therapy, and peaks at month 3.15 This increased risk of VTE within the first 3 months then may be explained by the rise in Hct. Other research has found that the rate of MACE is higher in the first year of TTh, which may further support that the rise in Hct is more important than final value.16
Rates of MACE were greatest in the cohort with a baseline Hct of 36-39% when compared to all other cohorts but this is likely explained by the more unhealthy individuals in this cohort. For example in the cohort with baseline Hct of 36-39% that stayed in this range after starting TTh, rates of hypertension were 51.5%, hyperlipidemia was 45.1%, type II diabetes 30.7%, prior stroke was 4.1%. Compared to men with a baseline Hct of 44-47% that stayed in this range after starting TTh: hypertension was 38.5%, hyperlipidemia was 44.0%, type II diabetes 15.8%, prior stroke was 0.74%. Thus the higher overall frequency of MACE is likely explained by a more unhealthy cohort for those with Hct of 36-39%. This is why each cohort is propensity matched to its own same baseline Hct values to demonstrate the risk associated with each increases in Hct quintile compared to men who saw no increase in Hct after starting TTh.
One of the major strengths of our study is the number of patients we were able to compare. One of the main criticisms of studies linking TTh and MACE is that they are not adequately powered to show causation.17 With over 16,000 men in our analysis after matching,we were able to have adequate power for our primary outcome. Another strength of our study is separating Hct into quintiles and using this to investigate an exposure-response relationship between Hct and MACE. While a simple cutoff of 52% has been shown to be correlated with increased risk of MACE, finding an exposure-response relationship further strengthens the plausibility that Hct increases are an important factor in MACE risk in men on TTh. One of the major limitations of our study is that due to using an anonymized database, we lose the ability to investigate individual patient factors in the analysis including testosterone preparation used,compliance, or dosages as well as patient diet or exercise routines at baseline. Unfortunately,because the TriNetX database obfuscates any counts where the incidence is 10 or less to protect HIPAA, we are unable to obtain precise counts of the sub-categories of MACE (stroke, MI, and death). Regardless, because we still do show a consistent increase in every starting quintile of Hct, we believe our findings still are intriguing. Because some cohorts that experienced an increase in Hct after starting testosterone had smaller numbers of patients compared to the cohort not experiencing an increase and other times, those experiencing an increase had larger numbers than those not experiencing an increase, we could not consistently keep the target population for propensity score modeling to only those experiencing an increase and this could limit the generalizability of our study. TriNetX database also only allows for propensity score matching rather than regression analysis limiting our analytical approach and requiring categorical treatment of Hct values rather than as a continuous variable. Additionally, we do not know what interventions were undertaken upon development of erythrocytosis, or if these men were appropriately diagnosed with hypogonadism. More and more clinics are offering TTh without proper guideline-based diagnosis which may increase their risk for adverse effects.18-20 Lastly,we were not able to determine which testosterone modalities these men were on. Since longer acting testosterone modalities have a higher risk of erythrocytosis than short-acting modalities, itis theoretically possible that these would also have a higher risk of MACE, but this has not been investigated yet.21, 22 Finally, assessing whether a decline in Hct after starting testosterone is or is not associated with future MACE would have been intriguing, unfortunately, too few men saw a decrease in Hct upon starting testosterone to allow for analysis of these cohorts.
This study expands the likelihood that increases in Hct account for the incidence of MACE possibly seen in men on TT. A large RCT (TRAVERSE Trial) is currently being undertaken to further elucidate this connection but unfortunately has not included Hct-based increases as a factor in their analysis. While the decision on how or when to act upon a rising Hct once starting testosterone therapy is a conversation that patients and their doctors will have to consider. Certainly, these men who are experiencing a rise in Hct could be encouraged to talk with their primary care physician to determine if any other steps could help to decrease their risk of MACE – such as testing for hyperlipidemia or starting treatment if hypertensive. Further research is needed to determine what amount of Hct increase should raise concerns amongst clinicians to warrant intervention.
CONCLUSIONS
In this study, we find that an increase in Hct from baseline after starting testosterone therapy is associated with an increased risk of MACE even when the increase is below traditional cut-off values, such as 52%. Clinicians should monitor Hct levels after initiating testosterone recognizing and counseling patients that increases in Hct may place patients a thigher risk of MACE. Further studies are needed to confirm these findings and to assess when clinicians should intervene when Hct is rising after starting testosterone therapy.
