madman
Super Moderator
Testosterone therapy and secondary erythrocytosis (2021)
Joshua White, Francis Petrella, and Jesse Ory
Secondary erythrocytosis is one of the most common adverse events associated with testosterone therapy (TT). Upon encountering this, clinicians will often either adjust TT dosing, stop therapy, order phlebotomy, or recommend a combination of these. Despite this, the evidence for secondary polycythemia causing harm during TT is scarce, and the hematocrit-based cutoffs present in multiple guidelines appear to be arbitrarily chosen. In this review, we present the pathophysiology behind TT and secondary erythrocytosis, the evidence connecting TT, secondary erythrocytosis, and major adverse cardiovascular events (MACE), and the data supporting varying interventions upon diagnosis of secondary erythrocytosis.
INTRODUCTION
Men with biochemical evidence of low testosterone and symptoms of hypogonadism may benefit from testosterone therapy (TT) [1]. Proposed benefits of TT include improved sexual function, bone mineral density as well as increased strength improved lipid profiles [2–5]. Between 2001 and 2011, the number of patients prescribed testosterone in the United States tripled [6, 7]. Prescribing patterns shifted in both Canada and the United States after 2 studies reported increased myocardial infarction and thromboembolic events associated with testosterone use [8, 9]. While these studies had questionable methodologies and controversial conclusions, following their publication, the rates of new testosterone users, as well as total testosterone users, decreased by 22–62% [7, 10]. In one large commercial health insurance population-based study, the rate of testosterone usage in men over 30 in the United States in 2016 was 1.67% [7].
The most common dose-limiting effect of TT is erythrocytosis [11, 12]. This is defined as an erythrocyte mass exceeding 125% predicted based on sex and body mass [13]. More practical definitions vary depending on the guideline, ranging from a hematocrit of 48% up to 55% [1, 14, 15]. Polycythemia and erythrocytosis have been used interchangeably, though polycythemia implies an increase in all blood cells and so erythrocytosis is more accurate when describing the increased erythrocyte mass related to TT [13]. Erythrocytosis confers an increased blood viscosity and concerns arise from the potential increased risk of thromboembolic events including myocardial infarction and cerebrovascular accidents.
There is controversial evidence surrounding increased risks of MACE and venous thromboembolic events (VTE) and their association with TT. In a recent study of 39,622 men, Walker et al. found that TT was associated with an increased short-term risk for VTE [16]. In contrast, another study of 805 hypogonadal men found that long-term TT reduced risks of both MACE and mortality [17]. Amongst randomized controlled trials (RCTs), meta-analyses do not support an increased risk of MACE; however, these trials are not designed to detect MACE as their primary outcome and are often underpowered to do so. Unfortunately, none of these aforementioned trials evaluated secondary polycythemia as a potential independent risk factor for these adverse events. Amongst multiple guidelines currently, polycythemia is grounds for TT cessation or modification, without evidence that this actually increases the risk.
Discontinuation of TT results in normalization of the erythrocyte mass parameters with time, though in many men, the positive effects, or at least perceived benefits of TT make discontinuation challenging to do. Altering the testosterone formulation or dose prescribed may change the risk of secondary erythrocytosis in patients, though there is a lack of head-to-head trials comparing formulations. Another treatment option includes phlebotomy though the benefits of this practice are not clear in men with secondary erythrocytosis from TT. Herein, we will review the pathophysiology of polycythemia and testosterone therapy, and review the evidence regarding necessary intervention for men who develop polycythemia while on TT.
PATHOPHYSIOLOGY OF ERYTHROCYTOSIS AND TT
Polycythemia can be divided into primary and secondary according to its etiology. Primary polycythemia, also known as polycythemia vera (PCV), is due to an over-production of erythrocytes secondary to intrinsic cellular defects within the bone marrow. PCV is often due to a mutation in JAK2, a tyrosine kinase, leading to the proliferation of erythrocytes independent of cellular control. The secondary subtype is due to either a physiological response to decreased tissue oxygenation or from inappropriate stimulation of erythropoiesis, such as with TT [18].
Hematocrit increase after TT is usually seen within the first few months of treatment and returns to baseline within one year of treatment cessation [19]. The mechanism behind secondary erythrocytosis from TT is multifactorial. See Fig. 1 for a schematic representation of the current understanding of the pathophysiology.
Hepcidin is a hepatic protein that acts as a regulator of iron metabolism; increasing levels of hepcidin decrease gut absorption of iron increases intracellular iron storage and thus decreases hemoglobin production. The role of hepcidin was first proposed by Bachman et al. as a key player in the relationship between testosterone and erythropoiesis [20]. This study followed 109 men for 20 weeks during TT and found that increased testosterone levels inhibited hepcidin by more than 50% in all age groups and in a dose-dependent manner. This suppression was persistent throughout the duration of TT. The marked decrease in hepcidin is hypothesized to increase iron metabolism, systemic absorption of iron, and erythropoiesis.
Another mechanism behind secondary polycythemia involves erythropoietin (EPO) [21]. Cellular hypoxia stimulates EPO, a renal cytokine, causing an increase in red blood cell production directly in the bone marrow. Bachman et al. were also able to demonstrate that TT causes a transient spike in EPO. This results in a new set point for EPO expression, where cytokine release is triggered by a smaller drop in hematocrit [21]. EPO levels failed to decline after subsequent hemoglobin rises, demonstrating the possibility for uninhibited stimulation.
Estradiol, a breakdown product of testosterone via aromatase, may also play a role in polycythemia. Calado et al. found that estradiol increased hematopoietic telomerase, an enzyme that prevents the shortening of telomeres during cell division [22]. Therefore the increase in estradiol via increased aromatization in men on TT may increase telomerase activity, resulting in increased hematopoietic stem cell proliferation and survival [23].
The relationship between obstructive sleep apnea (OSA) and TT leading to erythrocytosis is not yet completely elucidated [24]. OSA by itself is believed to cause erythrocytosis via hypoxemia [25]. Combined with TT, it is possible that the effects on erythrocytosis may be compounded, either through higher metabolic requirements with elevated testosterone, changes in response to hypoxia, and physiologic changes to the airways [24]. Lundy et al. found a strong correlation between OSA and erythrocytosis in men on TT even after correcting for confounding factors. More than half of patients that developed erythrocytosis while on TT were found to have OSA. Optimizing co-morbid medical conditions such as OSA has been shown to not only decrease the risk of erythrocytosis but also increase the likelihood of symptom improvement while on TT [26].
While an increased red blood cell mass can lead to increased oxygen-carrying capacity, it may also have negative outcomes at a supraphysiological level. Erythrocytosis can lead to increased blood viscosity [27, 28]. At the molecular level, it has been shown to impact platelet adhesiveness and bleeding time [29]. Elevated hematocrit levels can also increase thromboxane A2 concentrations thereby activating new platelets and increasing platelet aggregation [16]. Studies have shown that polycythemia may also result in a systemic effect on venous return and cardiac function [13, 30]. PCV has been heavily associated with an increased risk of thromboembolic and MACE [31]. Despite this, adverse events that may result from the development of polycythemia while on TT is less clear [11].
WHO IS AT RISK FOR ERYTHROCYTOSIS?
Patient characteristics at initial evaluation can help predict the likelihood of erythrocytosis and adverse effects. Patients with obstructive sleep apnea, advanced age, obesity, type II diabetes mellitus, elevated baseline hematocrit (>50%), and those who live in high altitudes are at higher risk of developing erythrocytosis after TT [32]. Testosterone formulation, dose, and pharmacokinetics collectively determine the risk of erythrocytosis [11]. The general hypothesis is that increased duration in supraphysiological testosterone levels results in an increased risk of erythrocytosis. Short-acting intramuscular (IM) formulations result in supraphysiological testosterone levels achieved days after administration [33]. Extended-release injectable testosterone and transdermal options maintain physiological testosterone levels more effectively and reduce the risk of secondary erythrocytosis [11, 33, 34]. Of the various formulations, intramuscular injections have an estimated rate of erythrocytosis of 40% followed by subcutaneous pellets at 35%. Short-acting TT through transdermal and Androgel has been associated with lower rates of polycythemia; 15% and 3% respectively. The safest formulations include intranasal testosterone (0–2%) and oral testosterone (0.03%) [11, 35, 36]
Comorbidities, as well as TT formulation and pharmacokinetics, must be considered prior to initiate TT in order to identify if there is a preferred treatment modality. This is especially important in individuals who may have comorbid health conditions where avoiding polycythemia is paramount: congestive heart failure [1], prothrombotic conditions such as factor V Leiden, prothrombin gene mutations, Lupus antiphospholipid syndrome, and having elevated homocysteine or factor VIII levels [11]. In patients with risk factors for cardiovascular disease, the prescribing physician may opt for more physiological dosing with a daily pill, patch, or nasal gel. It is important to note that the risk for developing erythrocytosis has not been shown to be associated with the duration of TT and that the greatest risk for erythrocytosis likely occurs at the onset of therapy [23, 37].
DOES SECONDARY ERYTHROCYTOSIS CONFER AN INCREASED RISK OF MACE/VTE?
While primary erythrocytosis has been well established as a risk factor for thromboembolic events, the risk of secondary erythrocytosis related to TT is less clear. Secondary erythrocytosis is a predictable side effect of testosterone replacement therapy with objective increases in hematocrit noted after one month of therapy. In fact, men receiving TT have a 315% greater risk for developing erythrocytosis when compared to controls [23]. The percentage increase in hematocrit continues to increase in a linear dose-dependent fashion [38]. Side effects of secondary erythrocytosis resulting from hyperviscosity include paresthesias, blurred vision, fatigue, and headaches [39]. The most feared complication of hyperviscosity is an increased risk of clotting. There are conflicting reports in the literature regarding the association between secondary erythrocytosis and MACE and VTE. This conflicting evidence in part contributes to the varying definitions of what is a “high” hematocrit while on TT.
The Endocrine Society uses a hematocrit threshold of >50% as a relative contraindication to initiating TT and >54% as an indication to discontinue treatment [1]. The European Association of Urology (EAU) guidelines on hypogonadism also state that the hematocrit should not exceed 54%, while recent Canadian guidelines cite 55% as the safe upper limit [15, 40]. The AUA guidelines on testosterone deficiency define polycythemia as a hematocrit of 52% and recommend stopping or reducing treatment if the hematocrit reaches 54% [14]. The upper limit of normal for hematocrit in most laboratory reference ranges in healthy adult males is 54%, which is where this value is likely derived. Interestingly, HCT > 48% was shown in the Framingham Heart Study to be associated with an increased risk of coronary artery disease and mortality [41]. In another large scale Swedish study evaluating hematocrit values and the risk of subsequent myocardial infarction, Toss et al. found that hematocrit ≥49% was associated with a 1.4-fold increased risk of myocardial infarction compared with men with a hematocrit ≤44% [42]. The major limitation to using these studies is that they involve population sampling, and do not investigate men on testosterone. Regardless, a hematocrit of ≥54% appears to be the consistent threshold to discontinuing or reducing treatment utilized by major urologic governing bodies, while the evidence for this specific cutoff is lacking. Definitions of associated terms related to erythrocytosis and polycythemia can be found in Table 1
There are no randomized or prospective studies that have documented a direct relationship between TT-related erythrocytosis and thromboembolic events [11]. Studies by Basaria et al. [43], Finkle et al. [8], and Vigen et al. [9], all claim an increased risk of MACE associated with TT, which prompted warnings and regulatory changes by the U.S. Food and Drug Administration (FDA). The FDA required manufacturers to change their labeling to indicate a possible increased risk of heart attacks and stroked in patients taking testosterone [44].
Basaria et al. reported instances of elevated hematocrit, but did not report any numbers for comparison, or use polycythemia as a predictive variable for MACE. In a contemporary series by Walker et al., they too found an increased risk of MACE associated with testosterone use, though again polycythemia specifically was not evaluated as a potential contributing factor [16]. In a meta-analysis of all randomized controlled trials for TT and cardiovascular risk by Corona et al., they concluded that the existing evidence does not support a causal role between TT and adverse CV events when hypogonadism is appropriately diagnosed and treated [45]. While there is no convincing evidence that links polycythemia in patients who are on TT and MACE, physicians should be prepared to discuss the risks with their patients in a shared-decision making approach.
EVIDENCE FOR INTERVENING UPON DEVELOPING SECONDARY ERYTHROCYTOSIS
If a patient’s hematocrit level rises to or above 54%, both the EAU and AUA recommend intervening. The EAU recommends investigating for other contributing causes such as those outlined in the risk factors section [40]. Erythropoiesis tends to occur primarily in the first 6 months of treatment and then reaches a plateau [37, 46]. The hematocrit and hemoglobin tend to return to baseline after 3–12 months once TT is discontinued [21, 47]. Adjusting the formulation, dose or pharmacokinetic parameters of TT may also decrease erythrocytosis by creating a more physiological hormone profile. In some patients, testosterone replacement therapy is an essential component of gender affirmation and it is important for the physician to be aware of these situations. Furthermore, it is important for clinicians to consider a lower threshold for erythrocytosis in transgender males.
Therapeutic phlebotomy is one way in which patients may “treat” erythrocytosis. Phlebotomy is a mainstay of treatment in PCV and there are no absolute contraindications [48]. In a 2013 study, Marchioli et al. randomized 365 patients with PV to a hematocrit target <45% or to a target of 45–50% [49]. After 31 months, there was a statistically significant increase in the number of major thromboembolic events in the group targeted to a higher hematocrit range (HR = 3.91) [49]. Phlebotomy is effective in the management of erythrocytosis in patients with PV in reducing thromboembolic events and this benefit may be conferred onto patients with secondary erythrocytosis secondary to TT, though there is no high-quality evidence to support this claim. Phlebotomy has been shown in one recent observational study to be effective at lowering the mean hemoglobin levels in patients receiving TT. Hazegh et al. identified 5498 patients on TT who donated blood consistently and found that over a one-year period dropped their hemoglobin from 17.86 g/dL to 16.8 g/dL on average. The donation frequency in patients receiving TT also ranged between 1–29 over a two-year period compared to 1–12 in age-matched controls [50]. There are no evidence-based guidelines that outline the frequency or volume of phlebotomy protocols in patients receiving TT.
While phlebotomy may seem like an appealing solution in the management of erythrocytosis, some concerns were highlighted by Chin-Yee et al., who evaluated 39 patients presenting for blood donation over a two-year period. They found that hematocrit levels ≥54% were found at 25% of appointments and that 44% of repeat donors (n = 12) had persistently elevated hematocrit at subsequent donations [51]. Their findings raise the concern about potential misperceptions by patients and healthcare providers that phlebotomy can reduce or eliminate the risks of secondary erythrocytosis related to TT.
*Whether or not phlebotomy has a role to play in the management of secondary polycythemia deserves further study. At this point, phlebotomy appears to be a safe approach in reducing hematocrit but physicians should be aware that it may not be sufficient in reducing hematocrit levels on a long-term basis. Phlebotomy may be used as a temporizing measure to allow dose/formulation alterations in patients with secondary erythrocytosis receiving T.
CONCLUSION
Secondary erythrocytosis is the most common adverse effect of TT. The mechanism driving hematopoiesis with TT is multifactorial. Injectables and pellets confer a greater risk for erythrocytosis than other available formulations. There is an unclear relationship between TT, erythrocytosis, and MACE or other thromboembolic events. A large double-blind and placebo-controlled study in prospectively evaluating the risk of long-term vascular events in hypogonadal men is currently underway and should help to answer this question. Hematocrit should be monitored at baseline prior to starting TT, and checked more frequently in the six months—the first year of initiating treatment. In men at risk of secondary erythrocytosis, using a testosterone modality with a lower risk of polycythemia should be considered.
Joshua White, Francis Petrella, and Jesse Ory
Secondary erythrocytosis is one of the most common adverse events associated with testosterone therapy (TT). Upon encountering this, clinicians will often either adjust TT dosing, stop therapy, order phlebotomy, or recommend a combination of these. Despite this, the evidence for secondary polycythemia causing harm during TT is scarce, and the hematocrit-based cutoffs present in multiple guidelines appear to be arbitrarily chosen. In this review, we present the pathophysiology behind TT and secondary erythrocytosis, the evidence connecting TT, secondary erythrocytosis, and major adverse cardiovascular events (MACE), and the data supporting varying interventions upon diagnosis of secondary erythrocytosis.
INTRODUCTION
Men with biochemical evidence of low testosterone and symptoms of hypogonadism may benefit from testosterone therapy (TT) [1]. Proposed benefits of TT include improved sexual function, bone mineral density as well as increased strength improved lipid profiles [2–5]. Between 2001 and 2011, the number of patients prescribed testosterone in the United States tripled [6, 7]. Prescribing patterns shifted in both Canada and the United States after 2 studies reported increased myocardial infarction and thromboembolic events associated with testosterone use [8, 9]. While these studies had questionable methodologies and controversial conclusions, following their publication, the rates of new testosterone users, as well as total testosterone users, decreased by 22–62% [7, 10]. In one large commercial health insurance population-based study, the rate of testosterone usage in men over 30 in the United States in 2016 was 1.67% [7].
The most common dose-limiting effect of TT is erythrocytosis [11, 12]. This is defined as an erythrocyte mass exceeding 125% predicted based on sex and body mass [13]. More practical definitions vary depending on the guideline, ranging from a hematocrit of 48% up to 55% [1, 14, 15]. Polycythemia and erythrocytosis have been used interchangeably, though polycythemia implies an increase in all blood cells and so erythrocytosis is more accurate when describing the increased erythrocyte mass related to TT [13]. Erythrocytosis confers an increased blood viscosity and concerns arise from the potential increased risk of thromboembolic events including myocardial infarction and cerebrovascular accidents.
There is controversial evidence surrounding increased risks of MACE and venous thromboembolic events (VTE) and their association with TT. In a recent study of 39,622 men, Walker et al. found that TT was associated with an increased short-term risk for VTE [16]. In contrast, another study of 805 hypogonadal men found that long-term TT reduced risks of both MACE and mortality [17]. Amongst randomized controlled trials (RCTs), meta-analyses do not support an increased risk of MACE; however, these trials are not designed to detect MACE as their primary outcome and are often underpowered to do so. Unfortunately, none of these aforementioned trials evaluated secondary polycythemia as a potential independent risk factor for these adverse events. Amongst multiple guidelines currently, polycythemia is grounds for TT cessation or modification, without evidence that this actually increases the risk.
Discontinuation of TT results in normalization of the erythrocyte mass parameters with time, though in many men, the positive effects, or at least perceived benefits of TT make discontinuation challenging to do. Altering the testosterone formulation or dose prescribed may change the risk of secondary erythrocytosis in patients, though there is a lack of head-to-head trials comparing formulations. Another treatment option includes phlebotomy though the benefits of this practice are not clear in men with secondary erythrocytosis from TT. Herein, we will review the pathophysiology of polycythemia and testosterone therapy, and review the evidence regarding necessary intervention for men who develop polycythemia while on TT.
PATHOPHYSIOLOGY OF ERYTHROCYTOSIS AND TT
Polycythemia can be divided into primary and secondary according to its etiology. Primary polycythemia, also known as polycythemia vera (PCV), is due to an over-production of erythrocytes secondary to intrinsic cellular defects within the bone marrow. PCV is often due to a mutation in JAK2, a tyrosine kinase, leading to the proliferation of erythrocytes independent of cellular control. The secondary subtype is due to either a physiological response to decreased tissue oxygenation or from inappropriate stimulation of erythropoiesis, such as with TT [18].
Hematocrit increase after TT is usually seen within the first few months of treatment and returns to baseline within one year of treatment cessation [19]. The mechanism behind secondary erythrocytosis from TT is multifactorial. See Fig. 1 for a schematic representation of the current understanding of the pathophysiology.
Hepcidin is a hepatic protein that acts as a regulator of iron metabolism; increasing levels of hepcidin decrease gut absorption of iron increases intracellular iron storage and thus decreases hemoglobin production. The role of hepcidin was first proposed by Bachman et al. as a key player in the relationship between testosterone and erythropoiesis [20]. This study followed 109 men for 20 weeks during TT and found that increased testosterone levels inhibited hepcidin by more than 50% in all age groups and in a dose-dependent manner. This suppression was persistent throughout the duration of TT. The marked decrease in hepcidin is hypothesized to increase iron metabolism, systemic absorption of iron, and erythropoiesis.
Another mechanism behind secondary polycythemia involves erythropoietin (EPO) [21]. Cellular hypoxia stimulates EPO, a renal cytokine, causing an increase in red blood cell production directly in the bone marrow. Bachman et al. were also able to demonstrate that TT causes a transient spike in EPO. This results in a new set point for EPO expression, where cytokine release is triggered by a smaller drop in hematocrit [21]. EPO levels failed to decline after subsequent hemoglobin rises, demonstrating the possibility for uninhibited stimulation.
Estradiol, a breakdown product of testosterone via aromatase, may also play a role in polycythemia. Calado et al. found that estradiol increased hematopoietic telomerase, an enzyme that prevents the shortening of telomeres during cell division [22]. Therefore the increase in estradiol via increased aromatization in men on TT may increase telomerase activity, resulting in increased hematopoietic stem cell proliferation and survival [23].
The relationship between obstructive sleep apnea (OSA) and TT leading to erythrocytosis is not yet completely elucidated [24]. OSA by itself is believed to cause erythrocytosis via hypoxemia [25]. Combined with TT, it is possible that the effects on erythrocytosis may be compounded, either through higher metabolic requirements with elevated testosterone, changes in response to hypoxia, and physiologic changes to the airways [24]. Lundy et al. found a strong correlation between OSA and erythrocytosis in men on TT even after correcting for confounding factors. More than half of patients that developed erythrocytosis while on TT were found to have OSA. Optimizing co-morbid medical conditions such as OSA has been shown to not only decrease the risk of erythrocytosis but also increase the likelihood of symptom improvement while on TT [26].
While an increased red blood cell mass can lead to increased oxygen-carrying capacity, it may also have negative outcomes at a supraphysiological level. Erythrocytosis can lead to increased blood viscosity [27, 28]. At the molecular level, it has been shown to impact platelet adhesiveness and bleeding time [29]. Elevated hematocrit levels can also increase thromboxane A2 concentrations thereby activating new platelets and increasing platelet aggregation [16]. Studies have shown that polycythemia may also result in a systemic effect on venous return and cardiac function [13, 30]. PCV has been heavily associated with an increased risk of thromboembolic and MACE [31]. Despite this, adverse events that may result from the development of polycythemia while on TT is less clear [11].
WHO IS AT RISK FOR ERYTHROCYTOSIS?
Patient characteristics at initial evaluation can help predict the likelihood of erythrocytosis and adverse effects. Patients with obstructive sleep apnea, advanced age, obesity, type II diabetes mellitus, elevated baseline hematocrit (>50%), and those who live in high altitudes are at higher risk of developing erythrocytosis after TT [32]. Testosterone formulation, dose, and pharmacokinetics collectively determine the risk of erythrocytosis [11]. The general hypothesis is that increased duration in supraphysiological testosterone levels results in an increased risk of erythrocytosis. Short-acting intramuscular (IM) formulations result in supraphysiological testosterone levels achieved days after administration [33]. Extended-release injectable testosterone and transdermal options maintain physiological testosterone levels more effectively and reduce the risk of secondary erythrocytosis [11, 33, 34]. Of the various formulations, intramuscular injections have an estimated rate of erythrocytosis of 40% followed by subcutaneous pellets at 35%. Short-acting TT through transdermal and Androgel has been associated with lower rates of polycythemia; 15% and 3% respectively. The safest formulations include intranasal testosterone (0–2%) and oral testosterone (0.03%) [11, 35, 36]
Comorbidities, as well as TT formulation and pharmacokinetics, must be considered prior to initiate TT in order to identify if there is a preferred treatment modality. This is especially important in individuals who may have comorbid health conditions where avoiding polycythemia is paramount: congestive heart failure [1], prothrombotic conditions such as factor V Leiden, prothrombin gene mutations, Lupus antiphospholipid syndrome, and having elevated homocysteine or factor VIII levels [11]. In patients with risk factors for cardiovascular disease, the prescribing physician may opt for more physiological dosing with a daily pill, patch, or nasal gel. It is important to note that the risk for developing erythrocytosis has not been shown to be associated with the duration of TT and that the greatest risk for erythrocytosis likely occurs at the onset of therapy [23, 37].
DOES SECONDARY ERYTHROCYTOSIS CONFER AN INCREASED RISK OF MACE/VTE?
While primary erythrocytosis has been well established as a risk factor for thromboembolic events, the risk of secondary erythrocytosis related to TT is less clear. Secondary erythrocytosis is a predictable side effect of testosterone replacement therapy with objective increases in hematocrit noted after one month of therapy. In fact, men receiving TT have a 315% greater risk for developing erythrocytosis when compared to controls [23]. The percentage increase in hematocrit continues to increase in a linear dose-dependent fashion [38]. Side effects of secondary erythrocytosis resulting from hyperviscosity include paresthesias, blurred vision, fatigue, and headaches [39]. The most feared complication of hyperviscosity is an increased risk of clotting. There are conflicting reports in the literature regarding the association between secondary erythrocytosis and MACE and VTE. This conflicting evidence in part contributes to the varying definitions of what is a “high” hematocrit while on TT.
The Endocrine Society uses a hematocrit threshold of >50% as a relative contraindication to initiating TT and >54% as an indication to discontinue treatment [1]. The European Association of Urology (EAU) guidelines on hypogonadism also state that the hematocrit should not exceed 54%, while recent Canadian guidelines cite 55% as the safe upper limit [15, 40]. The AUA guidelines on testosterone deficiency define polycythemia as a hematocrit of 52% and recommend stopping or reducing treatment if the hematocrit reaches 54% [14]. The upper limit of normal for hematocrit in most laboratory reference ranges in healthy adult males is 54%, which is where this value is likely derived. Interestingly, HCT > 48% was shown in the Framingham Heart Study to be associated with an increased risk of coronary artery disease and mortality [41]. In another large scale Swedish study evaluating hematocrit values and the risk of subsequent myocardial infarction, Toss et al. found that hematocrit ≥49% was associated with a 1.4-fold increased risk of myocardial infarction compared with men with a hematocrit ≤44% [42]. The major limitation to using these studies is that they involve population sampling, and do not investigate men on testosterone. Regardless, a hematocrit of ≥54% appears to be the consistent threshold to discontinuing or reducing treatment utilized by major urologic governing bodies, while the evidence for this specific cutoff is lacking. Definitions of associated terms related to erythrocytosis and polycythemia can be found in Table 1
There are no randomized or prospective studies that have documented a direct relationship between TT-related erythrocytosis and thromboembolic events [11]. Studies by Basaria et al. [43], Finkle et al. [8], and Vigen et al. [9], all claim an increased risk of MACE associated with TT, which prompted warnings and regulatory changes by the U.S. Food and Drug Administration (FDA). The FDA required manufacturers to change their labeling to indicate a possible increased risk of heart attacks and stroked in patients taking testosterone [44].
Basaria et al. reported instances of elevated hematocrit, but did not report any numbers for comparison, or use polycythemia as a predictive variable for MACE. In a contemporary series by Walker et al., they too found an increased risk of MACE associated with testosterone use, though again polycythemia specifically was not evaluated as a potential contributing factor [16]. In a meta-analysis of all randomized controlled trials for TT and cardiovascular risk by Corona et al., they concluded that the existing evidence does not support a causal role between TT and adverse CV events when hypogonadism is appropriately diagnosed and treated [45]. While there is no convincing evidence that links polycythemia in patients who are on TT and MACE, physicians should be prepared to discuss the risks with their patients in a shared-decision making approach.
EVIDENCE FOR INTERVENING UPON DEVELOPING SECONDARY ERYTHROCYTOSIS
If a patient’s hematocrit level rises to or above 54%, both the EAU and AUA recommend intervening. The EAU recommends investigating for other contributing causes such as those outlined in the risk factors section [40]. Erythropoiesis tends to occur primarily in the first 6 months of treatment and then reaches a plateau [37, 46]. The hematocrit and hemoglobin tend to return to baseline after 3–12 months once TT is discontinued [21, 47]. Adjusting the formulation, dose or pharmacokinetic parameters of TT may also decrease erythrocytosis by creating a more physiological hormone profile. In some patients, testosterone replacement therapy is an essential component of gender affirmation and it is important for the physician to be aware of these situations. Furthermore, it is important for clinicians to consider a lower threshold for erythrocytosis in transgender males.
Therapeutic phlebotomy is one way in which patients may “treat” erythrocytosis. Phlebotomy is a mainstay of treatment in PCV and there are no absolute contraindications [48]. In a 2013 study, Marchioli et al. randomized 365 patients with PV to a hematocrit target <45% or to a target of 45–50% [49]. After 31 months, there was a statistically significant increase in the number of major thromboembolic events in the group targeted to a higher hematocrit range (HR = 3.91) [49]. Phlebotomy is effective in the management of erythrocytosis in patients with PV in reducing thromboembolic events and this benefit may be conferred onto patients with secondary erythrocytosis secondary to TT, though there is no high-quality evidence to support this claim. Phlebotomy has been shown in one recent observational study to be effective at lowering the mean hemoglobin levels in patients receiving TT. Hazegh et al. identified 5498 patients on TT who donated blood consistently and found that over a one-year period dropped their hemoglobin from 17.86 g/dL to 16.8 g/dL on average. The donation frequency in patients receiving TT also ranged between 1–29 over a two-year period compared to 1–12 in age-matched controls [50]. There are no evidence-based guidelines that outline the frequency or volume of phlebotomy protocols in patients receiving TT.
While phlebotomy may seem like an appealing solution in the management of erythrocytosis, some concerns were highlighted by Chin-Yee et al., who evaluated 39 patients presenting for blood donation over a two-year period. They found that hematocrit levels ≥54% were found at 25% of appointments and that 44% of repeat donors (n = 12) had persistently elevated hematocrit at subsequent donations [51]. Their findings raise the concern about potential misperceptions by patients and healthcare providers that phlebotomy can reduce or eliminate the risks of secondary erythrocytosis related to TT.
*Whether or not phlebotomy has a role to play in the management of secondary polycythemia deserves further study. At this point, phlebotomy appears to be a safe approach in reducing hematocrit but physicians should be aware that it may not be sufficient in reducing hematocrit levels on a long-term basis. Phlebotomy may be used as a temporizing measure to allow dose/formulation alterations in patients with secondary erythrocytosis receiving T.
CONCLUSION
Secondary erythrocytosis is the most common adverse effect of TT. The mechanism driving hematopoiesis with TT is multifactorial. Injectables and pellets confer a greater risk for erythrocytosis than other available formulations. There is an unclear relationship between TT, erythrocytosis, and MACE or other thromboembolic events. A large double-blind and placebo-controlled study in prospectively evaluating the risk of long-term vascular events in hypogonadal men is currently underway and should help to answer this question. Hematocrit should be monitored at baseline prior to starting TT, and checked more frequently in the six months—the first year of initiating treatment. In men at risk of secondary erythrocytosis, using a testosterone modality with a lower risk of polycythemia should be considered.
