Cleveland Clinic Journal of Medicine

We thank those who reviewed manuscripts submitted to the Cleveland Clinic Journal of Medicine in 2018. Reviewing papers for the Journal—both for specialty content and for relevance to our readership—is an arduous task that involves considerable time and effort. Our publication decisions depend in no small part on the timely efforts of reviewers, and we are indebted to them for contributing their expertise this past year.   
—Brian F. Mandell, MD, PhD, Editor in Chief

We thank those who reviewed manuscripts submitted to the Cleveland Clinic Journal of Medicine in 2018. Reviewing papers for the Journal—both for specialty content and for relevance to our readership—is an arduous task that involves considerable time and effort. Our publication decisions depend in no small part on the timely efforts of reviewers, and we are indebted to them for contributing their expertise this past year.   
—Brian F. Mandell, MD, PhD, Editor in Chief

We thank those who reviewed manuscripts submitted to the Cleveland Clinic Journal of Medicine in 2018. Reviewing papers for the Journal—both for specialty content and for relevance to our readership—is an arduous task that involves considerable time and effort. Our publication decisions depend in no small part on the timely efforts of reviewers, and we are indebted to them for contributing their expertise this past year.   
—Brian F. Mandell, MD, PhD, Editor in Chief

MGUS is present in 3% to 4% of the population over age 50 and is more common in older men, African Americans, and Africans.^1–6

The overall risk of progression to myeloma and related disorders is less than or equal to 1% per year depending on the subtype of the M protein (higher risk with IgM than non-IgM and light-chain MGUS).^7,8 While the risk of malignant transformation is low, multiple myeloma is almost always preceded by the presence of an asymptomatic and often unrecognized monoclonal protein.

WHEN SHOULD WE LOOK FOR AN M PROTEIN?

An M protein is typically an incidental finding when a patient is being assessed for any of a number of presenting symptoms or conditions. A large retrospective study⁹ found that screening for MGUS was mostly performed by internal medicine physicians. The indications for testing were anemia, bone-related issues, elevated creatinine, elevated erythrocyte sedimentation rate, and neuropathy.

A low anion gap is not a major indicator of an M protein unless in a high concentration, in which case other manifestations would be present, such as renal failure, which would guide the diagnosis. Polyclonal hypergammaglobulinemia as a cause of low anion gap is far more common than MGUS.

HOW SHOULD WE SCREEN FOR AN M PROTEIN?

Serum protein electrophoresis is an initial test used to identify an M protein and has a key role in quantifying it (Figure 2). An M protein appears as a narrow spike on the agarose gel and should be distinguished from the broad band seen in polyclonal gammopathies associated with cirrhosis and chronic infectious and inflammatory conditions, among others.¹² A major disadvantage of serum protein electrophoresis is that it cannot detect an M protein in very low concentrations or determine its identity.

Serum immunofixation is more sensitive than serum protein electrophoresis and should always be ordered in conjunction with it, mostly to ensure detecting tiny amounts of M protein and to identify the type of its heavy chain and light-chain components.¹³

The serum free light-chain assay is also considered an essential part of the screening process to detect light-chain MGUS and light-chain myeloma. As many as 16% of myeloma patients secrete only light chains, which may not be identified on serum immunofixation.^{3,6,7,10,14,15} In general, a low kappa-lambda ratio (< 0.26) indicates the overproduction of lambda light chains, and a high ratio (> 1.65) indicates the overproduction of kappa light chains.

The serum free light-chain assay helps detect abnormal secretion of monoclonal light chains before they appear in the urine once the kidney tubules become saturated and unable to reabsorb them.

Of note, the free light-chain ratio can be abnormal (< 0.26 or > 1.65) in chronic kidney disease. Thus, it may be challenging to discern whether an abnormal light-chain ratio is related to impaired light-chain clearance by the kidneys or to MGUS. In general, kappa light chains are more elevated than lambda light chains in chronic kidney disease, but the ratio should not be considerably skewed. A kappa-lambda ratio below 0.37 or above 3 is rarely seen in chronic kidney disease and should prompt workup for MGUS.¹⁶

Tests in combination. The sensitivity of screening for M proteins ranges from 82% with serum protein electrophoresis alone to 93% with the addition of serum immunofixation and to 98% with the serum free light-chain assay.¹⁵ The latter can replace urine protein electrophoresis and immunofixation when screening for M protein, given its higher sensitivity.^15,17 An important caveat is that urine dipstick testing does not detect urine light chains.

Table 3 lists the initial laboratory tests required in patients with MGUS.

MGUS

Based on the International Myeloma Working Group consensus, a formal diagnosis of MGUS is established when a serum M protein is detected and measured at a concentration less than 3 g/dL on serum protein electrophoresis along with less than 10% clonal plasma cells in the bone marrow.^{1–6,14,18,19} Nevertheless, bone marrow biopsy can be omitted in certain patients as discussed below. The absence of myeloma-related organ damage—particularly osteolytic bone lesions, anemia, otherwise unexplained renal failure, and hypercalcemia—is fundamental and necessary for a diagnosis of MGUS.

Smoldering multiple myeloma

Compared with patients with MGUS, patients with smoldering multiple myeloma have higher M protein concentrations (≥ 3 g/dL) or 10% or more clonal plasma cells in the marrow or both, and are at higher risk of progression to symptomatic multiple myeloma. Nevertheless, like patients with MGUS, they have no myeloma symptoms or evidence of end-organ damage.

Symptomatic multiple myeloma

By definition, patients with multiple myeloma develop organ damage related to their malignancy and need therapy to halt disease progression. Multiple myeloma causes clinical manifestations through cellular infiltration of the bone and bone marrow (anemia, osteolysis, and hypercalcemia) and light chain-induced toxicity (renal tubular damage and cast nephropathy).

In 2014, the definition of multiple myeloma was updated to include 3 new myeloma-defining events that herald a significantly higher risk of progression from smoldering to symptomatic multiple myeloma, and now constitute an integral part of the diagnosis of symptomatic multiple myeloma. These are:

• Focal lesions (> 1 lesion larger than 5 mm) visible on magnetic resonance imaging
• ≥ 60% clonal plasma cells on bone marrow biopsy
• Ratio of involved to uninvolved serum free light chains ≥ 100 (the involved light chain is the one detected on serum protein electrophoresis and immunofixation).¹⁴

Bone pain, symptoms of anemia, and decreased urine output may suggest myeloma, but are not diagnostic. Although the “CRAB” criteria (elevated calcium, renal failure, anemia, and bone lesions) define multiple myeloma, the presence of anemia, hypercalcemia, or renal dysfunction do not by themselves mark transformation from MGUS to multiple myeloma. Thus, other causes need to be considered, since the risk of transformation is so low. Importantly, hyperparathyroidism must be ruled out if hypercalcemia is present in a patient with MGUS.¹⁰

Waldenström macroglobulinemia

Waldenström macroglobulinemia, also called lymphoplasmacytic lymphoma, is an indolent non-Hodgkin B-cell lymphoma that can invade the marrow, liver, spleen, and lymph nodes, leading to anemia and organomegaly. It features a monoclonal IgM protein that can be associated with increased blood viscosity, cold agglutinin disease, peripheral neuropathy, and cryoglobulinemia.

Waldenström macroglobulinemia should be suspected in any patient with IgM type M protein and symptoms related to hyperviscosity (headache, blurry vision, lightheadedness, shortness of breath, unexplained epistaxis,  gum bleeding); systemic symptoms (fever, weight loss, and night sweats); and abdominal pain (due to organomegaly).²³

Monoclonal gammopathy of renal significance

Monoclonal gammopathy of renal significance (MGRS) is a newly recognized entity defined by kidney dysfunction associated with an M protein without evidence of myeloma or other lymphoid disorders.²⁴ Multiple disorders have been included in this category with different underlying mechanisms of kidney injury. This entity is beyond the scope of this discussion.

Light-chain amyloidosis

Misfolded light-chain deposition leading to organ dysfunction is the hallmark of light-chain amyloidosis, which constitutes a subset of MGRS. An abnormal light-chain ratio, especially if skewed toward lambda should trigger an investigation for light-chain amyloidosis.¹⁰

Abnormal light chains may infiltrate any organ or tissue, but of greatest concern is infiltration of the myocardium with ensuing heart failure manifestations. N-terminal pro-B-type natriuretic peptide (NT-proBNP) is a sensitive marker for cardiac amyloidosis in the presence of suggestive features on transthoracic echocardiography (eg, left ventricular hypertrophy) but is not specific as it can be elevated in heart failure regardless of the underlying cause.¹⁰

Glomerular injury with nephrotic syndrome may also point toward renal involvement by light-chain amyloidosis and establishes a key distinctive factor from myeloma in which tubular injury is the main mechanism of kidney dysfunction.

Clinical clues for light-chain amyloidosis include heart failure symptoms, neuropathy, and macroglossia. If any of these symptoms and signs is present, we recommend electrocardiography (look for low voltage in limb leads), transthoracic echocardiography, measuring the NT-proBNP level, and urinalysis to look for albuminuria. Notably, carpal tunnel syndrome may be a very early clinical manifestation of amyloidosis, but by itself it is nonspecific. Light-chain amyloidosis is a common cause of macroglossia in adults.^10,25

Neuropathy associated with M proteins is a clinical entity related to a multitude of disorders that may necessitate treating the underlying cellular clone responsible for the secretion of the toxic M protein. These disorders include light-chain amyloidosis, POEMS (polyneuropathy, organomegaly, endocrinopathy, M protein, and skin changes or sclerotic bone lesions) syndrome, and IgM-related neuropathies with anti-myelin-associated glycoprotein antibodies.^3,10,11,14

Notably, weight loss and fatigue in a patient with MGUS may be the first signs of light-chain amyloidosis or Waldenström macroglobulinemia and should prompt further evaluation.²⁵

Research has helped to refine the diagnostic workup and recognize subsets of patients with MGUS at different risks of progression to myeloma and related disorders. Factors predicting progression are ^1,6,7,26,27:

• The amount of the M protein
• The type of M protein (IgG vs non-IgG)
• An abnormal free light-chain ratio.

Half of patients with MGUS fall into the low-risk category, which is defined by IgG-type serum M protein in a concentration less than 1.5 g/dL and a normal serum free light-chain ratio (kappa-lambda 0.26–1.65).^5,27 The absolute risk of progression at 20 years is only 5% for patients with low-risk MGUS, compared with 58% in patients with high-risk MGUS (positive for all 3 risk factors).⁵

The presence of less than 10% plasma cells in the bone marrow is required to satisfy the definition of MGUS, but bone marrow biopsy can be omitted for patients with low-risk MGUS, given the slim chance of finding a significant percentage of clonal plasma cells in the marrow and the inherently low risk of progression.^5,10 Skeletal surveys are often deferred for low-risk MGUS, but we obtain them in all our patients to ensure the absence of plasmacytomas, which need to be treated (typically with radiotherapy). Importantly, patients with unexplained bone pain (mostly in long bones, ribs, and spine, whereas joints are not typically involved) and a normal skeletal survey should undergo advanced imaging (whole-body magnetic resonance imaging or whole-body positron emission tomography and computed tomography) to detect bone lesions otherwise missed on plain radiography.^28,29

Most of the recommendations regarding follow-up are based on expert opinion, given the lack of randomized data. Most experts agree that all patients should be reevaluated 6 months after an M protein is detected, with laboratory surveillance tests (complete blood cell count, serum creatinine, serum calcium level, serum protein electrophoresis, and serum free light chains). Low-risk patients with a stable M protein level can be followed every 2 to 3 years.

Suspect malignant progression if the serum M protein level increases by 50% or more (with an absolute increase of ≥ 0.5 g/dL); the serum M protein level is 3 g/dL or higher; the serum free light-chain ratio is more than 100; or the patient has unexplained anemia, elevated creatinine, bone pain, fracture, or hypercalcemia.

Patients at intermediate or high risk should be followed annually after the initial 6-month visit.^5,7,10

A recent study highlighted the importance of risk stratification in reducing the costs associated with an overzealous diagnostic workup of patients with low-risk MGUS.³⁰ These savings are in addition to a reduction in patient anticipation and anxiety that universally occur before invasive procedures.

THE ROLE OF THE PRIMARY CARE PROVIDER AND THE HEMATOLOGIST

Once an M protein is identified, a comprehensive history, physical examination, and laboratory tests (serum protein electrophoresis to quantify the protein, serum immunofixation, serum free light chains, complete blood cell count, calcium, and creatinine) should be done, taking into consideration the differential diagnosis of monoclonal gammopathies discussed above. After MGUS is confirmed, the patient should be risk-stratified to determine the need for bone marrow biopsy and to predict the risk of progression to more serious conditions.

Referral to a hematologist is warranted for patients with intermediate- and high-risk MGUS, patients with abnormal serum free light-chain ratios, and those who show evidence of malignant progression. Patients with intermediate- and high-risk MGUS could be referred for bone marrow biopsy before assessment by a hematologist. The primary care provider may continue to follow patients with low-risk MGUS who do not display clinical or laboratory evidence of myeloma or related disorders.

The importance of educating patients to report any new worrisome symptom (eg, fatigue, neuropathy, weight loss, night sweats, bone pain) cannot be overemphasized, as some patients may progress to myeloma or other disorders between follow-up visits.

MGUS is present in 3% to 4% of the population over age 50 and is more common in older men, African Americans, and Africans.^1–6

The overall risk of progression to myeloma and related disorders is less than or equal to 1% per year depending on the subtype of the M protein (higher risk with IgM than non-IgM and light-chain MGUS).^7,8 While the risk of malignant transformation is low, multiple myeloma is almost always preceded by the presence of an asymptomatic and often unrecognized monoclonal protein.

WHEN SHOULD WE LOOK FOR AN M PROTEIN?

An M protein is typically an incidental finding when a patient is being assessed for any of a number of presenting symptoms or conditions. A large retrospective study⁹ found that screening for MGUS was mostly performed by internal medicine physicians. The indications for testing were anemia, bone-related issues, elevated creatinine, elevated erythrocyte sedimentation rate, and neuropathy.

A low anion gap is not a major indicator of an M protein unless in a high concentration, in which case other manifestations would be present, such as renal failure, which would guide the diagnosis. Polyclonal hypergammaglobulinemia as a cause of low anion gap is far more common than MGUS.

HOW SHOULD WE SCREEN FOR AN M PROTEIN?

Serum protein electrophoresis is an initial test used to identify an M protein and has a key role in quantifying it (Figure 2). An M protein appears as a narrow spike on the agarose gel and should be distinguished from the broad band seen in polyclonal gammopathies associated with cirrhosis and chronic infectious and inflammatory conditions, among others.¹² A major disadvantage of serum protein electrophoresis is that it cannot detect an M protein in very low concentrations or determine its identity.

Serum immunofixation is more sensitive than serum protein electrophoresis and should always be ordered in conjunction with it, mostly to ensure detecting tiny amounts of M protein and to identify the type of its heavy chain and light-chain components.¹³

The serum free light-chain assay is also considered an essential part of the screening process to detect light-chain MGUS and light-chain myeloma. As many as 16% of myeloma patients secrete only light chains, which may not be identified on serum immunofixation.^{3,6,7,10,14,15} In general, a low kappa-lambda ratio (< 0.26) indicates the overproduction of lambda light chains, and a high ratio (> 1.65) indicates the overproduction of kappa light chains.

The serum free light-chain assay helps detect abnormal secretion of monoclonal light chains before they appear in the urine once the kidney tubules become saturated and unable to reabsorb them.

Of note, the free light-chain ratio can be abnormal (< 0.26 or > 1.65) in chronic kidney disease. Thus, it may be challenging to discern whether an abnormal light-chain ratio is related to impaired light-chain clearance by the kidneys or to MGUS. In general, kappa light chains are more elevated than lambda light chains in chronic kidney disease, but the ratio should not be considerably skewed. A kappa-lambda ratio below 0.37 or above 3 is rarely seen in chronic kidney disease and should prompt workup for MGUS.¹⁶

Tests in combination. The sensitivity of screening for M proteins ranges from 82% with serum protein electrophoresis alone to 93% with the addition of serum immunofixation and to 98% with the serum free light-chain assay.¹⁵ The latter can replace urine protein electrophoresis and immunofixation when screening for M protein, given its higher sensitivity.^15,17 An important caveat is that urine dipstick testing does not detect urine light chains.

Table 3 lists the initial laboratory tests required in patients with MGUS.

MGUS

Based on the International Myeloma Working Group consensus, a formal diagnosis of MGUS is established when a serum M protein is detected and measured at a concentration less than 3 g/dL on serum protein electrophoresis along with less than 10% clonal plasma cells in the bone marrow.^{1–6,14,18,19} Nevertheless, bone marrow biopsy can be omitted in certain patients as discussed below. The absence of myeloma-related organ damage—particularly osteolytic bone lesions, anemia, otherwise unexplained renal failure, and hypercalcemia—is fundamental and necessary for a diagnosis of MGUS.

Smoldering multiple myeloma

Compared with patients with MGUS, patients with smoldering multiple myeloma have higher M protein concentrations (≥ 3 g/dL) or 10% or more clonal plasma cells in the marrow or both, and are at higher risk of progression to symptomatic multiple myeloma. Nevertheless, like patients with MGUS, they have no myeloma symptoms or evidence of end-organ damage.

Symptomatic multiple myeloma

By definition, patients with multiple myeloma develop organ damage related to their malignancy and need therapy to halt disease progression. Multiple myeloma causes clinical manifestations through cellular infiltration of the bone and bone marrow (anemia, osteolysis, and hypercalcemia) and light chain-induced toxicity (renal tubular damage and cast nephropathy).

In 2014, the definition of multiple myeloma was updated to include 3 new myeloma-defining events that herald a significantly higher risk of progression from smoldering to symptomatic multiple myeloma, and now constitute an integral part of the diagnosis of symptomatic multiple myeloma. These are:

• Focal lesions (> 1 lesion larger than 5 mm) visible on magnetic resonance imaging
• ≥ 60% clonal plasma cells on bone marrow biopsy
• Ratio of involved to uninvolved serum free light chains ≥ 100 (the involved light chain is the one detected on serum protein electrophoresis and immunofixation).¹⁴

Bone pain, symptoms of anemia, and decreased urine output may suggest myeloma, but are not diagnostic. Although the “CRAB” criteria (elevated calcium, renal failure, anemia, and bone lesions) define multiple myeloma, the presence of anemia, hypercalcemia, or renal dysfunction do not by themselves mark transformation from MGUS to multiple myeloma. Thus, other causes need to be considered, since the risk of transformation is so low. Importantly, hyperparathyroidism must be ruled out if hypercalcemia is present in a patient with MGUS.¹⁰

Waldenström macroglobulinemia

Waldenström macroglobulinemia, also called lymphoplasmacytic lymphoma, is an indolent non-Hodgkin B-cell lymphoma that can invade the marrow, liver, spleen, and lymph nodes, leading to anemia and organomegaly. It features a monoclonal IgM protein that can be associated with increased blood viscosity, cold agglutinin disease, peripheral neuropathy, and cryoglobulinemia.

Waldenström macroglobulinemia should be suspected in any patient with IgM type M protein and symptoms related to hyperviscosity (headache, blurry vision, lightheadedness, shortness of breath, unexplained epistaxis,  gum bleeding); systemic symptoms (fever, weight loss, and night sweats); and abdominal pain (due to organomegaly).²³

Monoclonal gammopathy of renal significance

Monoclonal gammopathy of renal significance (MGRS) is a newly recognized entity defined by kidney dysfunction associated with an M protein without evidence of myeloma or other lymphoid disorders.²⁴ Multiple disorders have been included in this category with different underlying mechanisms of kidney injury. This entity is beyond the scope of this discussion.

Light-chain amyloidosis

Misfolded light-chain deposition leading to organ dysfunction is the hallmark of light-chain amyloidosis, which constitutes a subset of MGRS. An abnormal light-chain ratio, especially if skewed toward lambda should trigger an investigation for light-chain amyloidosis.¹⁰

Abnormal light chains may infiltrate any organ or tissue, but of greatest concern is infiltration of the myocardium with ensuing heart failure manifestations. N-terminal pro-B-type natriuretic peptide (NT-proBNP) is a sensitive marker for cardiac amyloidosis in the presence of suggestive features on transthoracic echocardiography (eg, left ventricular hypertrophy) but is not specific as it can be elevated in heart failure regardless of the underlying cause.¹⁰

Glomerular injury with nephrotic syndrome may also point toward renal involvement by light-chain amyloidosis and establishes a key distinctive factor from myeloma in which tubular injury is the main mechanism of kidney dysfunction.

Clinical clues for light-chain amyloidosis include heart failure symptoms, neuropathy, and macroglossia. If any of these symptoms and signs is present, we recommend electrocardiography (look for low voltage in limb leads), transthoracic echocardiography, measuring the NT-proBNP level, and urinalysis to look for albuminuria. Notably, carpal tunnel syndrome may be a very early clinical manifestation of amyloidosis, but by itself it is nonspecific. Light-chain amyloidosis is a common cause of macroglossia in adults.^10,25

Neuropathy associated with M proteins is a clinical entity related to a multitude of disorders that may necessitate treating the underlying cellular clone responsible for the secretion of the toxic M protein. These disorders include light-chain amyloidosis, POEMS (polyneuropathy, organomegaly, endocrinopathy, M protein, and skin changes or sclerotic bone lesions) syndrome, and IgM-related neuropathies with anti-myelin-associated glycoprotein antibodies.^3,10,11,14

Notably, weight loss and fatigue in a patient with MGUS may be the first signs of light-chain amyloidosis or Waldenström macroglobulinemia and should prompt further evaluation.²⁵

Research has helped to refine the diagnostic workup and recognize subsets of patients with MGUS at different risks of progression to myeloma and related disorders. Factors predicting progression are ^1,6,7,26,27:

• The amount of the M protein
• The type of M protein (IgG vs non-IgG)
• An abnormal free light-chain ratio.

Half of patients with MGUS fall into the low-risk category, which is defined by IgG-type serum M protein in a concentration less than 1.5 g/dL and a normal serum free light-chain ratio (kappa-lambda 0.26–1.65).^5,27 The absolute risk of progression at 20 years is only 5% for patients with low-risk MGUS, compared with 58% in patients with high-risk MGUS (positive for all 3 risk factors).⁵

The presence of less than 10% plasma cells in the bone marrow is required to satisfy the definition of MGUS, but bone marrow biopsy can be omitted for patients with low-risk MGUS, given the slim chance of finding a significant percentage of clonal plasma cells in the marrow and the inherently low risk of progression.^5,10 Skeletal surveys are often deferred for low-risk MGUS, but we obtain them in all our patients to ensure the absence of plasmacytomas, which need to be treated (typically with radiotherapy). Importantly, patients with unexplained bone pain (mostly in long bones, ribs, and spine, whereas joints are not typically involved) and a normal skeletal survey should undergo advanced imaging (whole-body magnetic resonance imaging or whole-body positron emission tomography and computed tomography) to detect bone lesions otherwise missed on plain radiography.^28,29

Most of the recommendations regarding follow-up are based on expert opinion, given the lack of randomized data. Most experts agree that all patients should be reevaluated 6 months after an M protein is detected, with laboratory surveillance tests (complete blood cell count, serum creatinine, serum calcium level, serum protein electrophoresis, and serum free light chains). Low-risk patients with a stable M protein level can be followed every 2 to 3 years.

Suspect malignant progression if the serum M protein level increases by 50% or more (with an absolute increase of ≥ 0.5 g/dL); the serum M protein level is 3 g/dL or higher; the serum free light-chain ratio is more than 100; or the patient has unexplained anemia, elevated creatinine, bone pain, fracture, or hypercalcemia.

Patients at intermediate or high risk should be followed annually after the initial 6-month visit.^5,7,10

A recent study highlighted the importance of risk stratification in reducing the costs associated with an overzealous diagnostic workup of patients with low-risk MGUS.³⁰ These savings are in addition to a reduction in patient anticipation and anxiety that universally occur before invasive procedures.

THE ROLE OF THE PRIMARY CARE PROVIDER AND THE HEMATOLOGIST

Once an M protein is identified, a comprehensive history, physical examination, and laboratory tests (serum protein electrophoresis to quantify the protein, serum immunofixation, serum free light chains, complete blood cell count, calcium, and creatinine) should be done, taking into consideration the differential diagnosis of monoclonal gammopathies discussed above. After MGUS is confirmed, the patient should be risk-stratified to determine the need for bone marrow biopsy and to predict the risk of progression to more serious conditions.

Referral to a hematologist is warranted for patients with intermediate- and high-risk MGUS, patients with abnormal serum free light-chain ratios, and those who show evidence of malignant progression. Patients with intermediate- and high-risk MGUS could be referred for bone marrow biopsy before assessment by a hematologist. The primary care provider may continue to follow patients with low-risk MGUS who do not display clinical or laboratory evidence of myeloma or related disorders.

The importance of educating patients to report any new worrisome symptom (eg, fatigue, neuropathy, weight loss, night sweats, bone pain) cannot be overemphasized, as some patients may progress to myeloma or other disorders between follow-up visits.

MGUS is present in 3% to 4% of the population over age 50 and is more common in older men, African Americans, and Africans.^1–6

The overall risk of progression to myeloma and related disorders is less than or equal to 1% per year depending on the subtype of the M protein (higher risk with IgM than non-IgM and light-chain MGUS).^7,8 While the risk of malignant transformation is low, multiple myeloma is almost always preceded by the presence of an asymptomatic and often unrecognized monoclonal protein.

WHEN SHOULD WE LOOK FOR AN M PROTEIN?

An M protein is typically an incidental finding when a patient is being assessed for any of a number of presenting symptoms or conditions. A large retrospective study⁹ found that screening for MGUS was mostly performed by internal medicine physicians. The indications for testing were anemia, bone-related issues, elevated creatinine, elevated erythrocyte sedimentation rate, and neuropathy.

A low anion gap is not a major indicator of an M protein unless in a high concentration, in which case other manifestations would be present, such as renal failure, which would guide the diagnosis. Polyclonal hypergammaglobulinemia as a cause of low anion gap is far more common than MGUS.

HOW SHOULD WE SCREEN FOR AN M PROTEIN?

Serum protein electrophoresis is an initial test used to identify an M protein and has a key role in quantifying it (Figure 2). An M protein appears as a narrow spike on the agarose gel and should be distinguished from the broad band seen in polyclonal gammopathies associated with cirrhosis and chronic infectious and inflammatory conditions, among others.¹² A major disadvantage of serum protein electrophoresis is that it cannot detect an M protein in very low concentrations or determine its identity.

Serum immunofixation is more sensitive than serum protein electrophoresis and should always be ordered in conjunction with it, mostly to ensure detecting tiny amounts of M protein and to identify the type of its heavy chain and light-chain components.¹³

The serum free light-chain assay is also considered an essential part of the screening process to detect light-chain MGUS and light-chain myeloma. As many as 16% of myeloma patients secrete only light chains, which may not be identified on serum immunofixation.^{3,6,7,10,14,15} In general, a low kappa-lambda ratio (< 0.26) indicates the overproduction of lambda light chains, and a high ratio (> 1.65) indicates the overproduction of kappa light chains.

The serum free light-chain assay helps detect abnormal secretion of monoclonal light chains before they appear in the urine once the kidney tubules become saturated and unable to reabsorb them.

Of note, the free light-chain ratio can be abnormal (< 0.26 or > 1.65) in chronic kidney disease. Thus, it may be challenging to discern whether an abnormal light-chain ratio is related to impaired light-chain clearance by the kidneys or to MGUS. In general, kappa light chains are more elevated than lambda light chains in chronic kidney disease, but the ratio should not be considerably skewed. A kappa-lambda ratio below 0.37 or above 3 is rarely seen in chronic kidney disease and should prompt workup for MGUS.¹⁶

Tests in combination. The sensitivity of screening for M proteins ranges from 82% with serum protein electrophoresis alone to 93% with the addition of serum immunofixation and to 98% with the serum free light-chain assay.¹⁵ The latter can replace urine protein electrophoresis and immunofixation when screening for M protein, given its higher sensitivity.^15,17 An important caveat is that urine dipstick testing does not detect urine light chains.

Table 3 lists the initial laboratory tests required in patients with MGUS.

MGUS

Based on the International Myeloma Working Group consensus, a formal diagnosis of MGUS is established when a serum M protein is detected and measured at a concentration less than 3 g/dL on serum protein electrophoresis along with less than 10% clonal plasma cells in the bone marrow.^{1–6,14,18,19} Nevertheless, bone marrow biopsy can be omitted in certain patients as discussed below. The absence of myeloma-related organ damage—particularly osteolytic bone lesions, anemia, otherwise unexplained renal failure, and hypercalcemia—is fundamental and necessary for a diagnosis of MGUS.

Smoldering multiple myeloma

Compared with patients with MGUS, patients with smoldering multiple myeloma have higher M protein concentrations (≥ 3 g/dL) or 10% or more clonal plasma cells in the marrow or both, and are at higher risk of progression to symptomatic multiple myeloma. Nevertheless, like patients with MGUS, they have no myeloma symptoms or evidence of end-organ damage.

Symptomatic multiple myeloma

By definition, patients with multiple myeloma develop organ damage related to their malignancy and need therapy to halt disease progression. Multiple myeloma causes clinical manifestations through cellular infiltration of the bone and bone marrow (anemia, osteolysis, and hypercalcemia) and light chain-induced toxicity (renal tubular damage and cast nephropathy).

In 2014, the definition of multiple myeloma was updated to include 3 new myeloma-defining events that herald a significantly higher risk of progression from smoldering to symptomatic multiple myeloma, and now constitute an integral part of the diagnosis of symptomatic multiple myeloma. These are:

• Focal lesions (> 1 lesion larger than 5 mm) visible on magnetic resonance imaging
• ≥ 60% clonal plasma cells on bone marrow biopsy
• Ratio of involved to uninvolved serum free light chains ≥ 100 (the involved light chain is the one detected on serum protein electrophoresis and immunofixation).¹⁴

Bone pain, symptoms of anemia, and decreased urine output may suggest myeloma, but are not diagnostic. Although the “CRAB” criteria (elevated calcium, renal failure, anemia, and bone lesions) define multiple myeloma, the presence of anemia, hypercalcemia, or renal dysfunction do not by themselves mark transformation from MGUS to multiple myeloma. Thus, other causes need to be considered, since the risk of transformation is so low. Importantly, hyperparathyroidism must be ruled out if hypercalcemia is present in a patient with MGUS.¹⁰

Waldenström macroglobulinemia

Waldenström macroglobulinemia, also called lymphoplasmacytic lymphoma, is an indolent non-Hodgkin B-cell lymphoma that can invade the marrow, liver, spleen, and lymph nodes, leading to anemia and organomegaly. It features a monoclonal IgM protein that can be associated with increased blood viscosity, cold agglutinin disease, peripheral neuropathy, and cryoglobulinemia.

Waldenström macroglobulinemia should be suspected in any patient with IgM type M protein and symptoms related to hyperviscosity (headache, blurry vision, lightheadedness, shortness of breath, unexplained epistaxis,  gum bleeding); systemic symptoms (fever, weight loss, and night sweats); and abdominal pain (due to organomegaly).²³

Monoclonal gammopathy of renal significance

Monoclonal gammopathy of renal significance (MGRS) is a newly recognized entity defined by kidney dysfunction associated with an M protein without evidence of myeloma or other lymphoid disorders.²⁴ Multiple disorders have been included in this category with different underlying mechanisms of kidney injury. This entity is beyond the scope of this discussion.

Light-chain amyloidosis

Misfolded light-chain deposition leading to organ dysfunction is the hallmark of light-chain amyloidosis, which constitutes a subset of MGRS. An abnormal light-chain ratio, especially if skewed toward lambda should trigger an investigation for light-chain amyloidosis.¹⁰

Abnormal light chains may infiltrate any organ or tissue, but of greatest concern is infiltration of the myocardium with ensuing heart failure manifestations. N-terminal pro-B-type natriuretic peptide (NT-proBNP) is a sensitive marker for cardiac amyloidosis in the presence of suggestive features on transthoracic echocardiography (eg, left ventricular hypertrophy) but is not specific as it can be elevated in heart failure regardless of the underlying cause.¹⁰

Glomerular injury with nephrotic syndrome may also point toward renal involvement by light-chain amyloidosis and establishes a key distinctive factor from myeloma in which tubular injury is the main mechanism of kidney dysfunction.

Clinical clues for light-chain amyloidosis include heart failure symptoms, neuropathy, and macroglossia. If any of these symptoms and signs is present, we recommend electrocardiography (look for low voltage in limb leads), transthoracic echocardiography, measuring the NT-proBNP level, and urinalysis to look for albuminuria. Notably, carpal tunnel syndrome may be a very early clinical manifestation of amyloidosis, but by itself it is nonspecific. Light-chain amyloidosis is a common cause of macroglossia in adults.^10,25

Neuropathy associated with M proteins is a clinical entity related to a multitude of disorders that may necessitate treating the underlying cellular clone responsible for the secretion of the toxic M protein. These disorders include light-chain amyloidosis, POEMS (polyneuropathy, organomegaly, endocrinopathy, M protein, and skin changes or sclerotic bone lesions) syndrome, and IgM-related neuropathies with anti-myelin-associated glycoprotein antibodies.^3,10,11,14

Notably, weight loss and fatigue in a patient with MGUS may be the first signs of light-chain amyloidosis or Waldenström macroglobulinemia and should prompt further evaluation.²⁵

Research has helped to refine the diagnostic workup and recognize subsets of patients with MGUS at different risks of progression to myeloma and related disorders. Factors predicting progression are ^1,6,7,26,27:

• The amount of the M protein
• The type of M protein (IgG vs non-IgG)
• An abnormal free light-chain ratio.

Half of patients with MGUS fall into the low-risk category, which is defined by IgG-type serum M protein in a concentration less than 1.5 g/dL and a normal serum free light-chain ratio (kappa-lambda 0.26–1.65).^5,27 The absolute risk of progression at 20 years is only 5% for patients with low-risk MGUS, compared with 58% in patients with high-risk MGUS (positive for all 3 risk factors).⁵

The presence of less than 10% plasma cells in the bone marrow is required to satisfy the definition of MGUS, but bone marrow biopsy can be omitted for patients with low-risk MGUS, given the slim chance of finding a significant percentage of clonal plasma cells in the marrow and the inherently low risk of progression.^5,10 Skeletal surveys are often deferred for low-risk MGUS, but we obtain them in all our patients to ensure the absence of plasmacytomas, which need to be treated (typically with radiotherapy). Importantly, patients with unexplained bone pain (mostly in long bones, ribs, and spine, whereas joints are not typically involved) and a normal skeletal survey should undergo advanced imaging (whole-body magnetic resonance imaging or whole-body positron emission tomography and computed tomography) to detect bone lesions otherwise missed on plain radiography.^28,29

Most of the recommendations regarding follow-up are based on expert opinion, given the lack of randomized data. Most experts agree that all patients should be reevaluated 6 months after an M protein is detected, with laboratory surveillance tests (complete blood cell count, serum creatinine, serum calcium level, serum protein electrophoresis, and serum free light chains). Low-risk patients with a stable M protein level can be followed every 2 to 3 years.

Suspect malignant progression if the serum M protein level increases by 50% or more (with an absolute increase of ≥ 0.5 g/dL); the serum M protein level is 3 g/dL or higher; the serum free light-chain ratio is more than 100; or the patient has unexplained anemia, elevated creatinine, bone pain, fracture, or hypercalcemia.

Patients at intermediate or high risk should be followed annually after the initial 6-month visit.^5,7,10

A recent study highlighted the importance of risk stratification in reducing the costs associated with an overzealous diagnostic workup of patients with low-risk MGUS.³⁰ These savings are in addition to a reduction in patient anticipation and anxiety that universally occur before invasive procedures.

THE ROLE OF THE PRIMARY CARE PROVIDER AND THE HEMATOLOGIST

Once an M protein is identified, a comprehensive history, physical examination, and laboratory tests (serum protein electrophoresis to quantify the protein, serum immunofixation, serum free light chains, complete blood cell count, calcium, and creatinine) should be done, taking into consideration the differential diagnosis of monoclonal gammopathies discussed above. After MGUS is confirmed, the patient should be risk-stratified to determine the need for bone marrow biopsy and to predict the risk of progression to more serious conditions.

Referral to a hematologist is warranted for patients with intermediate- and high-risk MGUS, patients with abnormal serum free light-chain ratios, and those who show evidence of malignant progression. Patients with intermediate- and high-risk MGUS could be referred for bone marrow biopsy before assessment by a hematologist. The primary care provider may continue to follow patients with low-risk MGUS who do not display clinical or laboratory evidence of myeloma or related disorders.

The importance of educating patients to report any new worrisome symptom (eg, fatigue, neuropathy, weight loss, night sweats, bone pain) cannot be overemphasized, as some patients may progress to myeloma or other disorders between follow-up visits.

When treating high blood pressure, how low should we try to go? Debate continues about optimal blood pressure goals after publication of guidelines from the American College of Cardiology and American Heart Association (ACC/AHA) in 2017 that set or permitted a treatment goal of less than 130 mm Hg, depending on the population.¹

In this article, we summarize the evolution of hypertension guidelines and the evidence behind them.

HOW THE GOALS EVOLVED

JNC 7, 2003: 140/90 or 130/80

The seventh report of the Joint National Committee on Prevention, Detection, Evaluation, and Treatment of High Blood Pressure (JNC 7),² published in 2003, specified treatment goals of:

• < 140/90 mm Hg for most patients
• < 130/80 mm Hg for those with diabetes or chronic kidney disease.

JNC 7 provided much-needed clarity and uniformity to managing hypertension. Since then, various scientific groups have published their own guidelines (Table 1).^1–9

ACC/AHA/CDC 2014: 140/90

In 2014, the ACC, AHA, and US Centers for Disease Control and Prevention (CDC) published an evidence-based algorithm for hypertension management.³ As in JNC 7, they suggested a blood pressure goal of less than 140/90 mm Hg, lifestyle modification, and polytherapy, eg, a thiazide diuretic for stage 1 hypertension (< 160/100 mm Hg) and combination therapy with a thiazide diuretic and an angiotensin-converting enzyme (ACE) inhibitor, angiotensin II receptor blocker (ARB), or calcium channel blocker for stage 2 hypertension (≥ 160/100 mm Hg).

JNC 8 2014: 140/90 or 150/90

Soon after, the much-anticipated report of the panel members appointed to the eighth JNC (JNC 8) was published.⁴ Previous JNC reports were written and published under the auspices of the National Heart, Lung, and Blood Institute, but while the JNC 8 report was being prepared, this government body announced it would no longer publish guidelines.

In contrast to JNC 7, the JNC 8 panel based its recommendations on a systematic review of randomized clinical trials. However, the process and methodology were controversial, especially as the panel excluded some important clinical trials from the analysis.

JNC 8 relaxed the targets in several subgroups, such as patients over age 60 and those with diabetes and chronic kidney disease, due to a lack of definitive evidence on the impact of blood pressure targets lower than 140/90 mm Hg in these groups. Thus, their goals were:

• < 140/90 mm Hg for patients under age 60
• < 150/90 mm Hg for patients age 60 and older.

Of note, a minority of the JNC 8 panel disagreed with the new targets and provided evidence for keeping the systolic blood pressure target below 140 mm Hg for patients 60 and older.⁵ Further, the JNC 8 report was not endorsed by several important societies, ie, the AHA, ACC, National Heart, Lung, and Blood Institute, and American Society of Hypertension (ASH). These issues compromised the acceptance and applicability of the guidelines.

ASH/ISH 2014: 140/90 or 150/90

Also in 2014, the ASH and the International Society of Hypertension released their own report.⁶ Their goals:

• < 140/90 mm Hg for most patients
• < 150/90 mm Hg for patients age 80 and older.

AHA/ACC/ASH 2015: Goals in subgroups

In 2015, the AHA, ACC, and ASH released a joint scientific statement outlining hypertension goals for specific patient populations⁷:

• < 150/90 mm Hg for those age 80 and older
• < 140/90 mm Hg for those with coronary artery disease
• < 130/80 mm Hg for those with comorbidities such as diabetes and cardiovascular disease.

ADA 2016: Goals for patients with diabetes

In 2016, the American Diabetes Association (ADA) set the following blood pressure goals for patients with diabetes⁸:

• < 140/90 mm Hg for adults with diabetes
• < 130/80 mm Hg for younger adults with diabetes and adults with a high risk of cardiovascular disease
• 120–160/80–105 mm Hg for pregnant patients with diabetes and preexisting hypertension who are treated with antihypertensive therapy.

In 2017, the American College of Physicians (ACP) and the American Academy of Family Physicians (AAFP) recommended a relaxed systolic blood pressure target, ie, below 150 mm Hg, for adults over age 60, but a tighter goal of less than 140 mm Hg for the same age group if they have transient ischemic attack, stroke, or high cardiovascular risk.⁹

ACC/AHA 2017: 130/80

The 2017 ACC/AHA guidelines recommended a more aggressive goal of below 130/80 for all, including patients age 65 and older.¹

This is a class I (strong) recommendation for patients with known cardiovascular disease or a 10-year risk of a cardiovascular event of 10% or higher, with a B-R level of evidence for the systolic goal (ie, moderate-quality, based on systematic review of randomized controlled trials) and a C-EO level of evidence for the diastolic goal (ie, based on expert opinion).

For patients who do not have cardiovascular disease and who are at lower risk of it, this is a class IIb (weak) recommendation, ie, it “may be reasonable,” with a B-NR level of evidence (moderate-quality, based on nonrandomized studies) for the systolic goal and C-EO (expert opinion) for the diastolic goal.

For many patients, this involves drug treatment. For those with known cardiovascular disease or a 10-year risk of an atherosclerotic cardiovascular disease event of 10% or higher, the ACC/AHA guidelines say that drug treatment “is recommended” if their average blood pressure is 130/80 mm Hg or higher (class I recommendation, based on strong evidence for the systolic threshold and expert option for the diastolic). For those without cardiovascular disease and at lower risk, drug treatment is recommended if their average blood pressure is 140/90 mm Hg or higher (also class I, but based on limited data).

EVERYONE AGREES ON LIFESTYLE

Although the guidelines differ in their blood pressure targets, they consistently recommend lifestyle modifications.

Lifestyle modifications, first described in JNC 7, included weight loss, sodium restriction, and the DASH diet, which is rich in fruits, vegetables, low-fat dairy products, whole grains, poultry, and fish, and low in red meat, sweets, cholesterol, and total and saturated fat.²

These recommendations were based on results from 3 large randomized controlled trials in patients with and without hypertension.^10–12 In patients with no history of hypertension, interventions to promote weight loss and sodium restriction significantly reduced blood pressure and the incidence of hypertension (the latter by as much as 77%) compared with usual care.^10,11

In patients with and without hypertension, lowering sodium intake in conjunction with the DASH diet was associated with substantially larger reductions in systolic blood pressure.¹²

The recommendation to lower sodium intake has not changed in the guideline revisions. Meanwhile, other modifications have been added, such as incorporating both aerobic and resistance exercise and moderating alcohol intake. These recommendations have a class I level of evidence (ie, strongest level) in the 2017 ACC/AHA guidelines.¹

HYPERTENSION BEGINS AT 130/80

The definition of hypertension changed in the 2017 ACC/AHA guidelines¹: previously set at 140/90 mm Hg or higher, it is now 130/80 mm Hg or higher for all age groups. Adults with systolic blood pressure of 130 to 139 mm Hg or diastolic blood pressure of 80 to 89 mm Hg are now classified as having stage 1 hypertension.

Under the new definition, the number of US adults who have hypertension expanded to 45.6% of the general population,¹³ up from 31.9% under the JNC 7 definition. Thus, overall, 103.3 million US adults now have hypertension, compared with 72.2 million under the JNC 7 criteria.

In addition, the new guidelines expanded the population of adults for whom antihypertensive drug treatment is recommended to 36.2% (81.9 million). However, this represents only a 1.9% absolute increase over the JNC 7 recommendations (34.3%) and a 5.1% absolute increase over the JNC 8 recommendations.¹⁴

SPRINT: INTENSIVE TREATMENT IS BENEFICIAL

The new ACC/AHA guidelines¹ were based on evidence from several trials, including the Systolic Blood Pressure Intervention Trial (SPRINT).¹⁵

This multicenter trial investigated the effect of intensive blood pressure treatment on cardiovascular disease risk.¹⁶ The primary outcome was a composite of myocardial infarction, acute coronary syndrome, stroke, and heart failure.

The trial enrolled 9,361 participants at least 50 years of age with systolic blood pressure 130 mm Hg or higher and at least 1 additional risk factor for cardiovascular disease. It excluded anyone with a history of diabetes mellitus, stroke, symptomatic heart failure, or end-stage renal disease.

Two interventions were compared: 

• Intensive treatment, with a systolic blood pressure goal of less than 120 mm Hg: the protocol called for polytherapy, even for participants who were 75 or older if their blood pressure was 140 mm Hg or higher
• Standard treatment, with a systolic blood pressure goal of less than 140 mm Hg: it used polytherapy for patients whose systolic blood pressure was 160 mm Hg or higher.

The trial was intended to last 5 years but was stopped early at a median of 3.26 years owing to a significantly lower rate of the primary composite outcome in the intensive-treatment group: 1.65% per year vs 2.19%, a 25% relative risk reduction (P < .001) or a 0.54% absolute risk reduction. We calculate the number needed to treat (NNT) for 1 year to prevent 1 event as 185, and over the 3.26 years of the trial, the investigators calculated the NNT as 61. Similarly, the rate of death from any cause was also lower with intensive treatment, 1.03% per year vs 1.40% per year, a 27% relative risk reduction (P = .003) or a 0.37% absolute risk reduction, NNT 270.

Using these findings, Bress et al¹⁶ estimated that implementing intensive blood pressure goals could prevent 107,500 deaths annually.

The downside is adverse effects. In SPRINT,¹⁵ the intensive-treatment group experienced significantly higher rates of serious adverse effects than the standard-treatment group, ie:

• Hypotension 2.4% vs 1.4%, P = .001
• Syncope 2.3% vs 1.7%, P = .05
• Electrolyte abnormalities 3.1% vs 2.3%, P = .02)
• Acute kidney injury or kidney failure 4.1% vs 2.5%, P < .001
• Any treatment-related adverse event 4.7% vs 2.5%, P = .001.

Thus, Bress et al¹⁶ estimated that fully implementing the intensive-treatment goals could cause an additional 56,100 episodes of hypotension per year, 34,400 cases of syncope, 43,400 serious electrolyte disorders, and 88,700 cases of acute kidney injury. All told, about 3 million Americans could suffer a serious adverse effect under the intensive-treatment goals.

SPRINT¹⁵ was stopped early, after 3.26 years instead of the planned 5 years. The true risk-benefit ratio may have been different if the trial had been extended longer.

In addition, SPRINT used automated office blood pressure measurements in which patients were seated alone and a device (Model 907, Omron Healthcare) took 3 blood pressure measurements at 1-minute intervals after 5 minutes of quiet rest. This was designed to reduce elevated blood pressure readings in the presence of a healthcare professional in a medical setting (ie, “white coat” hypertension).

Many physicians are still taking blood pressure manually, which tends to give higher readings. Therefore, if they aim for a lower goal, they may risk overtreating the patient.

About 50% of patients did not achieve the target systolic blood pressure (< 120 mm Hg) despite receiving an average of 2.8 antihypertensive medications in the intensive-treatment group and 1.8 in the standard-treatment group. The use of antihypertensive medications, however, was not a controlled variable in the trial, and practitioners chose the appropriate drugs for their patients.

Diastolic pressure, which can be markedly lower in older hypertensive patients, was largely ignored, although lower diastolic pressure may have contributed to higher syncope rates in response to alpha blockers and calcium blockers.

Moreover, the trial excluded those with significant comorbidities and those younger than 50 (the mean age was 67.9), which limits the generalizability of the results.

JNC 8 VS SPRINT GOALS: WHAT'S THE EFFECT ON OUTCOMES?

JNC 8⁴ recommended a relaxed target of less than 140/90 mm Hg for adults younger than 60, including those with chronic kidney disease or diabetes, and less than 150/90 mm Hg for adults 60 and older. The SPRINT findings upended those recommendations, showing that intensive treatment in adults age 75 or older significantly improved the composite cardiovascular disease outcome (2.59 vs 3.85 events per year; P < .001) and all-cause mortality (1.78 vs 2.63 events per year; P < .05) compared with standard treatment.¹⁷ Also, a subset review of SPRINT trial data found no difference in benefit based on chronic kidney disease status.¹⁸

A meta-analysis of 74 clinical trials (N = 306,273) offers a compromise between the SPRINT findings and the JNC 8 recommendations.¹⁹ It found that the beneficial effect of blood pressure treatment depended on the patient’s baseline systolic blood pressure. In those with a baseline systolic pressure of 160 mm Hg or higher, treatment reduced cardiovascular mortality by about 15% (relative risk [RR] 0.85; 95% confidence interval [CI] 0.77–0.95). In patients with systolic pressure below 140 mm Hg, treatment effects were neutral (RR 1.03, 95% CI 0.87–1.20) and not associated with any benefit as primary prevention, although data suggest it may reduce the risk of adverse outcomes in patients with coronary heart disease.

OTHER TRIALS THAT INFLUENCED THE GUIDELINES

SHEP and HYVET (the Systolic Hypertension in the Elderly Program²⁰ and the Hypertension in the Very Elderly Trial)²¹ supported intensive blood pressure treatment for older patients by reporting a reduction in fatal and nonfatal stroke risks for those with a systolic blood pressure above 160 mm Hg.

FEVER (the Felodipine Event Reduction study)²² found that treatment with a calcium channel blocker in even a low dose can significantly decrease cardiovascular events, cardiovascular disease, and heart failure compared with no treatment.

JATOS and VALISH (the Japanese Trial to Assess Optimal Systolic Blood Pressure in Elderly Hypertensive Patients²³ and the Valsartan in Elderly Isolated Systolic Hypertension study)²⁴ found that outcomes were similar with intensive vs standard treatment.

Ettehad et al²⁵ performed a meta-analysis of 123 studies with more than 600,000 participants that provided strong evidence supporting blood pressure treatment goals below 130/90 mm Hg, in line with the SPRINT trial results.

BLOOD PRESSURE ISN’T EVERYTHING

Other trials remind us that although blood pressure is important, it is not the only factor affecting cardiovascular risk.

HOPE (the Heart Outcomes Prevention Evaluation)²⁶ investigated the use of ramipril (an ACE inhibitor) in preventing myocardial infarction, stroke, or cardiovascular death in patients at high risk of cardiovascular events. The study included 9,297 participants over age 55 (mean age 66) with a baseline blood pressure 139/79 mm Hg. Follow-up was 4.5 years.

Ramipril was better than placebo, with significantly fewer patients experiencing adverse end points in the ramipril group compared with the placebo group:

• Myocardial infarction 9.9% vs 12.3%, RR 0.80, P < .001
• Cardiovascular death 6.1% vs 8.1%, RR  0.74, P < .001
• Stroke 3.4% vs 4.9%, RR = .68, P < .001
• The composite end point 14.0% vs 17.8%, RR 0.78, P < .001).

Results were even better in the subset of patients who had diabetes.²⁷ However, the decrease in blood pressure attributable to anti­hypertensive therapy with ramipril was minimal (3–4 mm Hg systolic and 1–2 mm Hg diastolic). This slight change should not have been enough to produce significant differences in clinical outcomes, a major limitation of this trial. The investigators speculated that the positive results may be due to a class effect of ACE inhibitors.²⁶

HOPE 3^28–30 explored the effect of blood pressure- and cholesterol-controlling drugs on the same primary end points but in patients at intermediate risk of major cardiovascular events. Investigators randomized the 12,705 patients to 4 treatment groups:

• Blood pressure control with candesartan (an ARB) plus hydrochlorothiazide (a thiazide diuretic)
• Cholesterol control with rosuvastatin (a statin)
• Blood pressure plus cholesterol control
• Placebo.

Therapy was started at a systolic blood pressure above 140 mm Hg.

Compared with placebo, the rate of composite events was significantly reduced in the rosuvastatin group (3.7% vs 4.8%, HR 0.76, P = .002)²⁸ and the candesartan-hydrochlorothiazide-rosuvastatin group (3.6% vs 5.0%, HR 0.71; P = .005)²⁹ but not in the candesartan-hydrochlorothiazide group (4.1% vs 4.4%; HR 0.93; P = .40).³⁰

In addition, a subgroup analysis comparing active treatment vs placebo found a significant reduction in major cardiovascular events for treated patients whose baseline systolic blood pressure was in the upper third (> 143.5 mm Hg, mean 154.1 mm Hg), while treated patients in the lower middle and lower thirds had no significant reduction.³⁰

These results suggest that intensive treatment to achieve a systolic blood pressure below 140 mm Hg in patients at intermediate risk may not be helpful. Nevertheless, there seems to be agreement that intensive treatment generally leads to a reduction in cardiovascular events. The results also show the benefit of lowering cholesterol.

Bundy et al³¹ performed a meta-analysis that provides support for intensive antihypertensive treatment. Reviewing 42 clinical trials in more than 144,000 patients, they found that treating to reach a target systolic blood pressure of 120 to 124 mm Hg can reduce cardiovascular events and all-cause mortality.

The trade-off is a minimal increase in the risk of adverse events. Also, the risk-benefit ratio of intensive treatment seems to vary in different patient subgroups.

WHAT ABOUT PATIENTS WITH COMORBIDITIES?

The debate over intensive vs standard treatment in blood pressure management extends beyond hypertension and includes important comorbidities such as diabetes, stroke, and renal disease. Patients with a history of stroke or end-stage renal disease have only a minimal mention in the AHA/ACC guidelines.

Diabetes

Emdin et al,³² in a meta-analysis of 40 trials that included more than 100,000 patients with diabetes, concluded that a 10-mm Hg lowering of systolic blood pressure significantly reduces the rates of all-cause mortality, cardiovascular disease, coronary heart disease, stroke, albuminuria, and retinopathy. Stratifying the results according to the systolic blood pressure achieved (≥ 130 or < 130 mm Hg), the relative risks of mortality, coronary heart disease, cardiovascular disease, heart failure, and albuminuria were actually lower in the higher stratum than in the lower.

ACCORD (the Action to Control Cardiovascular Risk in Diabetes)³³ study provides contrary results. It examined intensive and standard blood pressure control targets in patients with type 2 diabetes at high risk of cardiovascular events, using primary outcome measures similar to those in SPRINT. It found no significant difference in fatal and nonfatal cardiovascular events between the intensive and standard blood pressure target arms.

Despite those results, the ACC/AHA guidelines still advocate for more intensive treatment (goal < 130/80 mm Hg) in all patients, including those with diabetes.¹

The ADA position statement (September 2017) recommended a target below 140/90 mm Hg in patients with diabetes and hypertension.⁸ However, they also noted that lower systolic and diastolic blood pressure targets, such as below 130/80 mm Hg, may be appropriate for patients at high risk of cardiovascular disease “if they can be achieved without undue treatment burden.”⁸ Thus, it is not clear which blood pressure targets in patients with diabetes are the best.

Stroke

In patients with stroke, AHA/ACC guidelines¹ recommend treatment if the blood pressure is 140/90 mm Hg or higher because antihypertensive therapy has been associated with a decrease in the recurrence of transient ischemic attack and stroke. The ideal target blood pressure is not known, but a goal of less than 130/80 mm Hg may be reasonable.

In the Secondary Prevention of Small Subcortical Strokes (SPS3) trial, a retrospective open-label trial, a target blood pressure below 130/80 mm Hg in patients with a history of lacunar stroke was associated with a lower risk of intracranial hemorrhage, but the difference was not statistically significant.³⁴ For this reason, the ACC/AHA guidelines consider it reasonable to aim for a systolic blood pressure below 130 mm Hg in these patients.¹

Renal disease

The ACC/AHA guidelines do not address how to manage hypertension in patients with end-stage renal disease, but for patients with chronic kidney disease they recommend a blood pressure target below 130/80 mm Hg.¹ This recommendation is derived from the SPRINT trial,¹⁵ in which patients with stage 3 or 4 chronic kidney disease accounted for 28% of the study population. In that subgroup, intensive blood pressure control seemed to provide the same benefits for reduction in cardiovascular death and all-cause mortality.

TREAT PATIENTS, NOT NUMBERS

Blood pressure targets should be applied in the appropriate clinical context and on a patient-by-patient basis. In clinical practice, one size does not always fit all, as special cases exist.

For example, blood pressure can oscillate widely in patients with autonomic nerve disorders, making it difficult to strive for a specific target, especially an intensive one. Thus, it may be necessary to allow higher systolic blood pressure in these patients. Similarly, patients with diabetes or chronic kidney disease may be at higher risk of kidney injury with more intensive blood pressure management.

Treating numbers rather than patients may result in unbalanced patient care. The optimal approach to blood pressure management relies on a comprehensive risk factor assessment and shared decision-making with the patient before setting specific blood pressure targets.

OUR APPROACH

We aim for a blood pressure goal below 130/80 mm Hg for all patients with cardiovascular disease, according to the AHA/ACC guidelines. We aim for that same target in patients without cardiovascular disease but who have an elevated estimated cardiovascular risk (> 10%) over the next 10 years.

We recognize, however, that the benefits of aggressive blood pressure reduction may not be as clear in all patients, such as those with diabetes. We also recognize that some patient subgroups are at high risk of adverse events, including those with low diastolic pressure, chronic kidney disease, a history of falls, and older age. In those patients, we are extremely judicious when titrating antihypertensive medications. We often make smaller titrations, at longer intervals, and with more frequent laboratory testing and in-office follow-up.

Our process of managing hypertension through intensive blood pressure control to achieve lower systolic blood pressure targets requires a concerted effort among healthcare providers at all levels. It especially requires more involvement and investment from primary care providers to individualize treatment in their patients. This process has helped us to reach our treatment goals while limiting adverse effects of lower blood pressure targets.

MOVING FORWARD

Hypertension is a major risk factor for cardiovascular disease, and intensive blood pressure control has the potential to significantly reduce rates of morbidity and death associated with cardiovascular disease. Thus, a general consensus on the definition of hypertension and treatment goals is essential to reduce the risk of cardiovascular events in this large patient population.

Intensive blood pressure treatment has shown efficacy, but it has a small accompanying risk of adverse events, which varies in patient subgroups and affects the benefit-risk ratio of this therapy. For example, the cardiovascular benefit of intensive treatment is less clear in diabetic patients, and the risk of adverse events may be higher in older patients with chronic kidney disease.

Moving forward, more research is needed into the effects of intensive and standard treatment on patients of all ages, those with common comorbid conditions, and those with other important factors such as diastolic hypertension.

Finally, the various medical societies should collaborate on hypertension guideline development. This would require considerable planning and coordination but would ultimately be useful in creating a generalizable approach to hypertension management.

When treating high blood pressure, how low should we try to go? Debate continues about optimal blood pressure goals after publication of guidelines from the American College of Cardiology and American Heart Association (ACC/AHA) in 2017 that set or permitted a treatment goal of less than 130 mm Hg, depending on the population.¹

In this article, we summarize the evolution of hypertension guidelines and the evidence behind them.

HOW THE GOALS EVOLVED

JNC 7, 2003: 140/90 or 130/80

The seventh report of the Joint National Committee on Prevention, Detection, Evaluation, and Treatment of High Blood Pressure (JNC 7),² published in 2003, specified treatment goals of:

• < 140/90 mm Hg for most patients
• < 130/80 mm Hg for those with diabetes or chronic kidney disease.

JNC 7 provided much-needed clarity and uniformity to managing hypertension. Since then, various scientific groups have published their own guidelines (Table 1).^1–9

ACC/AHA/CDC 2014: 140/90

In 2014, the ACC, AHA, and US Centers for Disease Control and Prevention (CDC) published an evidence-based algorithm for hypertension management.³ As in JNC 7, they suggested a blood pressure goal of less than 140/90 mm Hg, lifestyle modification, and polytherapy, eg, a thiazide diuretic for stage 1 hypertension (< 160/100 mm Hg) and combination therapy with a thiazide diuretic and an angiotensin-converting enzyme (ACE) inhibitor, angiotensin II receptor blocker (ARB), or calcium channel blocker for stage 2 hypertension (≥ 160/100 mm Hg).

JNC 8 2014: 140/90 or 150/90

Soon after, the much-anticipated report of the panel members appointed to the eighth JNC (JNC 8) was published.⁴ Previous JNC reports were written and published under the auspices of the National Heart, Lung, and Blood Institute, but while the JNC 8 report was being prepared, this government body announced it would no longer publish guidelines.

In contrast to JNC 7, the JNC 8 panel based its recommendations on a systematic review of randomized clinical trials. However, the process and methodology were controversial, especially as the panel excluded some important clinical trials from the analysis.

JNC 8 relaxed the targets in several subgroups, such as patients over age 60 and those with diabetes and chronic kidney disease, due to a lack of definitive evidence on the impact of blood pressure targets lower than 140/90 mm Hg in these groups. Thus, their goals were:

• < 140/90 mm Hg for patients under age 60
• < 150/90 mm Hg for patients age 60 and older.

Of note, a minority of the JNC 8 panel disagreed with the new targets and provided evidence for keeping the systolic blood pressure target below 140 mm Hg for patients 60 and older.⁵ Further, the JNC 8 report was not endorsed by several important societies, ie, the AHA, ACC, National Heart, Lung, and Blood Institute, and American Society of Hypertension (ASH). These issues compromised the acceptance and applicability of the guidelines.

ASH/ISH 2014: 140/90 or 150/90

Also in 2014, the ASH and the International Society of Hypertension released their own report.⁶ Their goals:

• < 140/90 mm Hg for most patients
• < 150/90 mm Hg for patients age 80 and older.

AHA/ACC/ASH 2015: Goals in subgroups

In 2015, the AHA, ACC, and ASH released a joint scientific statement outlining hypertension goals for specific patient populations⁷:

• < 150/90 mm Hg for those age 80 and older
• < 140/90 mm Hg for those with coronary artery disease
• < 130/80 mm Hg for those with comorbidities such as diabetes and cardiovascular disease.

ADA 2016: Goals for patients with diabetes

In 2016, the American Diabetes Association (ADA) set the following blood pressure goals for patients with diabetes⁸:

• < 140/90 mm Hg for adults with diabetes
• < 130/80 mm Hg for younger adults with diabetes and adults with a high risk of cardiovascular disease
• 120–160/80–105 mm Hg for pregnant patients with diabetes and preexisting hypertension who are treated with antihypertensive therapy.

In 2017, the American College of Physicians (ACP) and the American Academy of Family Physicians (AAFP) recommended a relaxed systolic blood pressure target, ie, below 150 mm Hg, for adults over age 60, but a tighter goal of less than 140 mm Hg for the same age group if they have transient ischemic attack, stroke, or high cardiovascular risk.⁹

ACC/AHA 2017: 130/80

The 2017 ACC/AHA guidelines recommended a more aggressive goal of below 130/80 for all, including patients age 65 and older.¹

This is a class I (strong) recommendation for patients with known cardiovascular disease or a 10-year risk of a cardiovascular event of 10% or higher, with a B-R level of evidence for the systolic goal (ie, moderate-quality, based on systematic review of randomized controlled trials) and a C-EO level of evidence for the diastolic goal (ie, based on expert opinion).

For patients who do not have cardiovascular disease and who are at lower risk of it, this is a class IIb (weak) recommendation, ie, it “may be reasonable,” with a B-NR level of evidence (moderate-quality, based on nonrandomized studies) for the systolic goal and C-EO (expert opinion) for the diastolic goal.

For many patients, this involves drug treatment. For those with known cardiovascular disease or a 10-year risk of an atherosclerotic cardiovascular disease event of 10% or higher, the ACC/AHA guidelines say that drug treatment “is recommended” if their average blood pressure is 130/80 mm Hg or higher (class I recommendation, based on strong evidence for the systolic threshold and expert option for the diastolic). For those without cardiovascular disease and at lower risk, drug treatment is recommended if their average blood pressure is 140/90 mm Hg or higher (also class I, but based on limited data).

EVERYONE AGREES ON LIFESTYLE

Although the guidelines differ in their blood pressure targets, they consistently recommend lifestyle modifications.

Lifestyle modifications, first described in JNC 7, included weight loss, sodium restriction, and the DASH diet, which is rich in fruits, vegetables, low-fat dairy products, whole grains, poultry, and fish, and low in red meat, sweets, cholesterol, and total and saturated fat.²

These recommendations were based on results from 3 large randomized controlled trials in patients with and without hypertension.^10–12 In patients with no history of hypertension, interventions to promote weight loss and sodium restriction significantly reduced blood pressure and the incidence of hypertension (the latter by as much as 77%) compared with usual care.^10,11

In patients with and without hypertension, lowering sodium intake in conjunction with the DASH diet was associated with substantially larger reductions in systolic blood pressure.¹²

The recommendation to lower sodium intake has not changed in the guideline revisions. Meanwhile, other modifications have been added, such as incorporating both aerobic and resistance exercise and moderating alcohol intake. These recommendations have a class I level of evidence (ie, strongest level) in the 2017 ACC/AHA guidelines.¹

HYPERTENSION BEGINS AT 130/80

The definition of hypertension changed in the 2017 ACC/AHA guidelines¹: previously set at 140/90 mm Hg or higher, it is now 130/80 mm Hg or higher for all age groups. Adults with systolic blood pressure of 130 to 139 mm Hg or diastolic blood pressure of 80 to 89 mm Hg are now classified as having stage 1 hypertension.

Under the new definition, the number of US adults who have hypertension expanded to 45.6% of the general population,¹³ up from 31.9% under the JNC 7 definition. Thus, overall, 103.3 million US adults now have hypertension, compared with 72.2 million under the JNC 7 criteria.

In addition, the new guidelines expanded the population of adults for whom antihypertensive drug treatment is recommended to 36.2% (81.9 million). However, this represents only a 1.9% absolute increase over the JNC 7 recommendations (34.3%) and a 5.1% absolute increase over the JNC 8 recommendations.¹⁴

SPRINT: INTENSIVE TREATMENT IS BENEFICIAL

The new ACC/AHA guidelines¹ were based on evidence from several trials, including the Systolic Blood Pressure Intervention Trial (SPRINT).¹⁵

This multicenter trial investigated the effect of intensive blood pressure treatment on cardiovascular disease risk.¹⁶ The primary outcome was a composite of myocardial infarction, acute coronary syndrome, stroke, and heart failure.

The trial enrolled 9,361 participants at least 50 years of age with systolic blood pressure 130 mm Hg or higher and at least 1 additional risk factor for cardiovascular disease. It excluded anyone with a history of diabetes mellitus, stroke, symptomatic heart failure, or end-stage renal disease.

Two interventions were compared: 

• Intensive treatment, with a systolic blood pressure goal of less than 120 mm Hg: the protocol called for polytherapy, even for participants who were 75 or older if their blood pressure was 140 mm Hg or higher
• Standard treatment, with a systolic blood pressure goal of less than 140 mm Hg: it used polytherapy for patients whose systolic blood pressure was 160 mm Hg or higher.

The trial was intended to last 5 years but was stopped early at a median of 3.26 years owing to a significantly lower rate of the primary composite outcome in the intensive-treatment group: 1.65% per year vs 2.19%, a 25% relative risk reduction (P < .001) or a 0.54% absolute risk reduction. We calculate the number needed to treat (NNT) for 1 year to prevent 1 event as 185, and over the 3.26 years of the trial, the investigators calculated the NNT as 61. Similarly, the rate of death from any cause was also lower with intensive treatment, 1.03% per year vs 1.40% per year, a 27% relative risk reduction (P = .003) or a 0.37% absolute risk reduction, NNT 270.

Using these findings, Bress et al¹⁶ estimated that implementing intensive blood pressure goals could prevent 107,500 deaths annually.

The downside is adverse effects. In SPRINT,¹⁵ the intensive-treatment group experienced significantly higher rates of serious adverse effects than the standard-treatment group, ie:

• Hypotension 2.4% vs 1.4%, P = .001
• Syncope 2.3% vs 1.7%, P = .05
• Electrolyte abnormalities 3.1% vs 2.3%, P = .02)
• Acute kidney injury or kidney failure 4.1% vs 2.5%, P < .001
• Any treatment-related adverse event 4.7% vs 2.5%, P = .001.

Thus, Bress et al¹⁶ estimated that fully implementing the intensive-treatment goals could cause an additional 56,100 episodes of hypotension per year, 34,400 cases of syncope, 43,400 serious electrolyte disorders, and 88,700 cases of acute kidney injury. All told, about 3 million Americans could suffer a serious adverse effect under the intensive-treatment goals.

SPRINT¹⁵ was stopped early, after 3.26 years instead of the planned 5 years. The true risk-benefit ratio may have been different if the trial had been extended longer.

In addition, SPRINT used automated office blood pressure measurements in which patients were seated alone and a device (Model 907, Omron Healthcare) took 3 blood pressure measurements at 1-minute intervals after 5 minutes of quiet rest. This was designed to reduce elevated blood pressure readings in the presence of a healthcare professional in a medical setting (ie, “white coat” hypertension).

Many physicians are still taking blood pressure manually, which tends to give higher readings. Therefore, if they aim for a lower goal, they may risk overtreating the patient.

About 50% of patients did not achieve the target systolic blood pressure (< 120 mm Hg) despite receiving an average of 2.8 antihypertensive medications in the intensive-treatment group and 1.8 in the standard-treatment group. The use of antihypertensive medications, however, was not a controlled variable in the trial, and practitioners chose the appropriate drugs for their patients.

Diastolic pressure, which can be markedly lower in older hypertensive patients, was largely ignored, although lower diastolic pressure may have contributed to higher syncope rates in response to alpha blockers and calcium blockers.

Moreover, the trial excluded those with significant comorbidities and those younger than 50 (the mean age was 67.9), which limits the generalizability of the results.

JNC 8 VS SPRINT GOALS: WHAT'S THE EFFECT ON OUTCOMES?

JNC 8⁴ recommended a relaxed target of less than 140/90 mm Hg for adults younger than 60, including those with chronic kidney disease or diabetes, and less than 150/90 mm Hg for adults 60 and older. The SPRINT findings upended those recommendations, showing that intensive treatment in adults age 75 or older significantly improved the composite cardiovascular disease outcome (2.59 vs 3.85 events per year; P < .001) and all-cause mortality (1.78 vs 2.63 events per year; P < .05) compared with standard treatment.¹⁷ Also, a subset review of SPRINT trial data found no difference in benefit based on chronic kidney disease status.¹⁸

A meta-analysis of 74 clinical trials (N = 306,273) offers a compromise between the SPRINT findings and the JNC 8 recommendations.¹⁹ It found that the beneficial effect of blood pressure treatment depended on the patient’s baseline systolic blood pressure. In those with a baseline systolic pressure of 160 mm Hg or higher, treatment reduced cardiovascular mortality by about 15% (relative risk [RR] 0.85; 95% confidence interval [CI] 0.77–0.95). In patients with systolic pressure below 140 mm Hg, treatment effects were neutral (RR 1.03, 95% CI 0.87–1.20) and not associated with any benefit as primary prevention, although data suggest it may reduce the risk of adverse outcomes in patients with coronary heart disease.

OTHER TRIALS THAT INFLUENCED THE GUIDELINES

SHEP and HYVET (the Systolic Hypertension in the Elderly Program²⁰ and the Hypertension in the Very Elderly Trial)²¹ supported intensive blood pressure treatment for older patients by reporting a reduction in fatal and nonfatal stroke risks for those with a systolic blood pressure above 160 mm Hg.

FEVER (the Felodipine Event Reduction study)²² found that treatment with a calcium channel blocker in even a low dose can significantly decrease cardiovascular events, cardiovascular disease, and heart failure compared with no treatment.

JATOS and VALISH (the Japanese Trial to Assess Optimal Systolic Blood Pressure in Elderly Hypertensive Patients²³ and the Valsartan in Elderly Isolated Systolic Hypertension study)²⁴ found that outcomes were similar with intensive vs standard treatment.

Ettehad et al²⁵ performed a meta-analysis of 123 studies with more than 600,000 participants that provided strong evidence supporting blood pressure treatment goals below 130/90 mm Hg, in line with the SPRINT trial results.

BLOOD PRESSURE ISN’T EVERYTHING

Other trials remind us that although blood pressure is important, it is not the only factor affecting cardiovascular risk.

HOPE (the Heart Outcomes Prevention Evaluation)²⁶ investigated the use of ramipril (an ACE inhibitor) in preventing myocardial infarction, stroke, or cardiovascular death in patients at high risk of cardiovascular events. The study included 9,297 participants over age 55 (mean age 66) with a baseline blood pressure 139/79 mm Hg. Follow-up was 4.5 years.

Ramipril was better than placebo, with significantly fewer patients experiencing adverse end points in the ramipril group compared with the placebo group:

• Myocardial infarction 9.9% vs 12.3%, RR 0.80, P < .001
• Cardiovascular death 6.1% vs 8.1%, RR  0.74, P < .001
• Stroke 3.4% vs 4.9%, RR = .68, P < .001
• The composite end point 14.0% vs 17.8%, RR 0.78, P < .001).

Results were even better in the subset of patients who had diabetes.²⁷ However, the decrease in blood pressure attributable to anti­hypertensive therapy with ramipril was minimal (3–4 mm Hg systolic and 1–2 mm Hg diastolic). This slight change should not have been enough to produce significant differences in clinical outcomes, a major limitation of this trial. The investigators speculated that the positive results may be due to a class effect of ACE inhibitors.²⁶

HOPE 3^28–30 explored the effect of blood pressure- and cholesterol-controlling drugs on the same primary end points but in patients at intermediate risk of major cardiovascular events. Investigators randomized the 12,705 patients to 4 treatment groups:

• Blood pressure control with candesartan (an ARB) plus hydrochlorothiazide (a thiazide diuretic)
• Cholesterol control with rosuvastatin (a statin)
• Blood pressure plus cholesterol control
• Placebo.

Therapy was started at a systolic blood pressure above 140 mm Hg.

Compared with placebo, the rate of composite events was significantly reduced in the rosuvastatin group (3.7% vs 4.8%, HR 0.76, P = .002)²⁸ and the candesartan-hydrochlorothiazide-rosuvastatin group (3.6% vs 5.0%, HR 0.71; P = .005)²⁹ but not in the candesartan-hydrochlorothiazide group (4.1% vs 4.4%; HR 0.93; P = .40).³⁰

In addition, a subgroup analysis comparing active treatment vs placebo found a significant reduction in major cardiovascular events for treated patients whose baseline systolic blood pressure was in the upper third (> 143.5 mm Hg, mean 154.1 mm Hg), while treated patients in the lower middle and lower thirds had no significant reduction.³⁰

These results suggest that intensive treatment to achieve a systolic blood pressure below 140 mm Hg in patients at intermediate risk may not be helpful. Nevertheless, there seems to be agreement that intensive treatment generally leads to a reduction in cardiovascular events. The results also show the benefit of lowering cholesterol.

Bundy et al³¹ performed a meta-analysis that provides support for intensive antihypertensive treatment. Reviewing 42 clinical trials in more than 144,000 patients, they found that treating to reach a target systolic blood pressure of 120 to 124 mm Hg can reduce cardiovascular events and all-cause mortality.

The trade-off is a minimal increase in the risk of adverse events. Also, the risk-benefit ratio of intensive treatment seems to vary in different patient subgroups.

WHAT ABOUT PATIENTS WITH COMORBIDITIES?

The debate over intensive vs standard treatment in blood pressure management extends beyond hypertension and includes important comorbidities such as diabetes, stroke, and renal disease. Patients with a history of stroke or end-stage renal disease have only a minimal mention in the AHA/ACC guidelines.

Diabetes

Emdin et al,³² in a meta-analysis of 40 trials that included more than 100,000 patients with diabetes, concluded that a 10-mm Hg lowering of systolic blood pressure significantly reduces the rates of all-cause mortality, cardiovascular disease, coronary heart disease, stroke, albuminuria, and retinopathy. Stratifying the results according to the systolic blood pressure achieved (≥ 130 or < 130 mm Hg), the relative risks of mortality, coronary heart disease, cardiovascular disease, heart failure, and albuminuria were actually lower in the higher stratum than in the lower.

ACCORD (the Action to Control Cardiovascular Risk in Diabetes)³³ study provides contrary results. It examined intensive and standard blood pressure control targets in patients with type 2 diabetes at high risk of cardiovascular events, using primary outcome measures similar to those in SPRINT. It found no significant difference in fatal and nonfatal cardiovascular events between the intensive and standard blood pressure target arms.

Despite those results, the ACC/AHA guidelines still advocate for more intensive treatment (goal < 130/80 mm Hg) in all patients, including those with diabetes.¹

The ADA position statement (September 2017) recommended a target below 140/90 mm Hg in patients with diabetes and hypertension.⁸ However, they also noted that lower systolic and diastolic blood pressure targets, such as below 130/80 mm Hg, may be appropriate for patients at high risk of cardiovascular disease “if they can be achieved without undue treatment burden.”⁸ Thus, it is not clear which blood pressure targets in patients with diabetes are the best.

Stroke

In patients with stroke, AHA/ACC guidelines¹ recommend treatment if the blood pressure is 140/90 mm Hg or higher because antihypertensive therapy has been associated with a decrease in the recurrence of transient ischemic attack and stroke. The ideal target blood pressure is not known, but a goal of less than 130/80 mm Hg may be reasonable.

In the Secondary Prevention of Small Subcortical Strokes (SPS3) trial, a retrospective open-label trial, a target blood pressure below 130/80 mm Hg in patients with a history of lacunar stroke was associated with a lower risk of intracranial hemorrhage, but the difference was not statistically significant.³⁴ For this reason, the ACC/AHA guidelines consider it reasonable to aim for a systolic blood pressure below 130 mm Hg in these patients.¹

Renal disease

The ACC/AHA guidelines do not address how to manage hypertension in patients with end-stage renal disease, but for patients with chronic kidney disease they recommend a blood pressure target below 130/80 mm Hg.¹ This recommendation is derived from the SPRINT trial,¹⁵ in which patients with stage 3 or 4 chronic kidney disease accounted for 28% of the study population. In that subgroup, intensive blood pressure control seemed to provide the same benefits for reduction in cardiovascular death and all-cause mortality.

TREAT PATIENTS, NOT NUMBERS

Blood pressure targets should be applied in the appropriate clinical context and on a patient-by-patient basis. In clinical practice, one size does not always fit all, as special cases exist.

For example, blood pressure can oscillate widely in patients with autonomic nerve disorders, making it difficult to strive for a specific target, especially an intensive one. Thus, it may be necessary to allow higher systolic blood pressure in these patients. Similarly, patients with diabetes or chronic kidney disease may be at higher risk of kidney injury with more intensive blood pressure management.

Treating numbers rather than patients may result in unbalanced patient care. The optimal approach to blood pressure management relies on a comprehensive risk factor assessment and shared decision-making with the patient before setting specific blood pressure targets.

OUR APPROACH

We aim for a blood pressure goal below 130/80 mm Hg for all patients with cardiovascular disease, according to the AHA/ACC guidelines. We aim for that same target in patients without cardiovascular disease but who have an elevated estimated cardiovascular risk (> 10%) over the next 10 years.

We recognize, however, that the benefits of aggressive blood pressure reduction may not be as clear in all patients, such as those with diabetes. We also recognize that some patient subgroups are at high risk of adverse events, including those with low diastolic pressure, chronic kidney disease, a history of falls, and older age. In those patients, we are extremely judicious when titrating antihypertensive medications. We often make smaller titrations, at longer intervals, and with more frequent laboratory testing and in-office follow-up.

Our process of managing hypertension through intensive blood pressure control to achieve lower systolic blood pressure targets requires a concerted effort among healthcare providers at all levels. It especially requires more involvement and investment from primary care providers to individualize treatment in their patients. This process has helped us to reach our treatment goals while limiting adverse effects of lower blood pressure targets.

MOVING FORWARD

Hypertension is a major risk factor for cardiovascular disease, and intensive blood pressure control has the potential to significantly reduce rates of morbidity and death associated with cardiovascular disease. Thus, a general consensus on the definition of hypertension and treatment goals is essential to reduce the risk of cardiovascular events in this large patient population.

Intensive blood pressure treatment has shown efficacy, but it has a small accompanying risk of adverse events, which varies in patient subgroups and affects the benefit-risk ratio of this therapy. For example, the cardiovascular benefit of intensive treatment is less clear in diabetic patients, and the risk of adverse events may be higher in older patients with chronic kidney disease.

Moving forward, more research is needed into the effects of intensive and standard treatment on patients of all ages, those with common comorbid conditions, and those with other important factors such as diastolic hypertension.

Finally, the various medical societies should collaborate on hypertension guideline development. This would require considerable planning and coordination but would ultimately be useful in creating a generalizable approach to hypertension management.

When treating high blood pressure, how low should we try to go? Debate continues about optimal blood pressure goals after publication of guidelines from the American College of Cardiology and American Heart Association (ACC/AHA) in 2017 that set or permitted a treatment goal of less than 130 mm Hg, depending on the population.¹

In this article, we summarize the evolution of hypertension guidelines and the evidence behind them.

HOW THE GOALS EVOLVED

JNC 7, 2003: 140/90 or 130/80

The seventh report of the Joint National Committee on Prevention, Detection, Evaluation, and Treatment of High Blood Pressure (JNC 7),² published in 2003, specified treatment goals of:

• < 140/90 mm Hg for most patients
• < 130/80 mm Hg for those with diabetes or chronic kidney disease.

JNC 7 provided much-needed clarity and uniformity to managing hypertension. Since then, various scientific groups have published their own guidelines (Table 1).^1–9

ACC/AHA/CDC 2014: 140/90

In 2014, the ACC, AHA, and US Centers for Disease Control and Prevention (CDC) published an evidence-based algorithm for hypertension management.³ As in JNC 7, they suggested a blood pressure goal of less than 140/90 mm Hg, lifestyle modification, and polytherapy, eg, a thiazide diuretic for stage 1 hypertension (< 160/100 mm Hg) and combination therapy with a thiazide diuretic and an angiotensin-converting enzyme (ACE) inhibitor, angiotensin II receptor blocker (ARB), or calcium channel blocker for stage 2 hypertension (≥ 160/100 mm Hg).

JNC 8 2014: 140/90 or 150/90

Soon after, the much-anticipated report of the panel members appointed to the eighth JNC (JNC 8) was published.⁴ Previous JNC reports were written and published under the auspices of the National Heart, Lung, and Blood Institute, but while the JNC 8 report was being prepared, this government body announced it would no longer publish guidelines.

In contrast to JNC 7, the JNC 8 panel based its recommendations on a systematic review of randomized clinical trials. However, the process and methodology were controversial, especially as the panel excluded some important clinical trials from the analysis.

JNC 8 relaxed the targets in several subgroups, such as patients over age 60 and those with diabetes and chronic kidney disease, due to a lack of definitive evidence on the impact of blood pressure targets lower than 140/90 mm Hg in these groups. Thus, their goals were:

• < 140/90 mm Hg for patients under age 60
• < 150/90 mm Hg for patients age 60 and older.

Of note, a minority of the JNC 8 panel disagreed with the new targets and provided evidence for keeping the systolic blood pressure target below 140 mm Hg for patients 60 and older.⁵ Further, the JNC 8 report was not endorsed by several important societies, ie, the AHA, ACC, National Heart, Lung, and Blood Institute, and American Society of Hypertension (ASH). These issues compromised the acceptance and applicability of the guidelines.

ASH/ISH 2014: 140/90 or 150/90

Also in 2014, the ASH and the International Society of Hypertension released their own report.⁶ Their goals:

• < 140/90 mm Hg for most patients
• < 150/90 mm Hg for patients age 80 and older.

AHA/ACC/ASH 2015: Goals in subgroups

In 2015, the AHA, ACC, and ASH released a joint scientific statement outlining hypertension goals for specific patient populations⁷:

• < 150/90 mm Hg for those age 80 and older
• < 140/90 mm Hg for those with coronary artery disease
• < 130/80 mm Hg for those with comorbidities such as diabetes and cardiovascular disease.

ADA 2016: Goals for patients with diabetes

In 2016, the American Diabetes Association (ADA) set the following blood pressure goals for patients with diabetes⁸:

• < 140/90 mm Hg for adults with diabetes
• < 130/80 mm Hg for younger adults with diabetes and adults with a high risk of cardiovascular disease
• 120–160/80–105 mm Hg for pregnant patients with diabetes and preexisting hypertension who are treated with antihypertensive therapy.

In 2017, the American College of Physicians (ACP) and the American Academy of Family Physicians (AAFP) recommended a relaxed systolic blood pressure target, ie, below 150 mm Hg, for adults over age 60, but a tighter goal of less than 140 mm Hg for the same age group if they have transient ischemic attack, stroke, or high cardiovascular risk.⁹

ACC/AHA 2017: 130/80

The 2017 ACC/AHA guidelines recommended a more aggressive goal of below 130/80 for all, including patients age 65 and older.¹

This is a class I (strong) recommendation for patients with known cardiovascular disease or a 10-year risk of a cardiovascular event of 10% or higher, with a B-R level of evidence for the systolic goal (ie, moderate-quality, based on systematic review of randomized controlled trials) and a C-EO level of evidence for the diastolic goal (ie, based on expert opinion).

For patients who do not have cardiovascular disease and who are at lower risk of it, this is a class IIb (weak) recommendation, ie, it “may be reasonable,” with a B-NR level of evidence (moderate-quality, based on nonrandomized studies) for the systolic goal and C-EO (expert opinion) for the diastolic goal.

For many patients, this involves drug treatment. For those with known cardiovascular disease or a 10-year risk of an atherosclerotic cardiovascular disease event of 10% or higher, the ACC/AHA guidelines say that drug treatment “is recommended” if their average blood pressure is 130/80 mm Hg or higher (class I recommendation, based on strong evidence for the systolic threshold and expert option for the diastolic). For those without cardiovascular disease and at lower risk, drug treatment is recommended if their average blood pressure is 140/90 mm Hg or higher (also class I, but based on limited data).

EVERYONE AGREES ON LIFESTYLE

Although the guidelines differ in their blood pressure targets, they consistently recommend lifestyle modifications.

Lifestyle modifications, first described in JNC 7, included weight loss, sodium restriction, and the DASH diet, which is rich in fruits, vegetables, low-fat dairy products, whole grains, poultry, and fish, and low in red meat, sweets, cholesterol, and total and saturated fat.²

These recommendations were based on results from 3 large randomized controlled trials in patients with and without hypertension.^10–12 In patients with no history of hypertension, interventions to promote weight loss and sodium restriction significantly reduced blood pressure and the incidence of hypertension (the latter by as much as 77%) compared with usual care.^10,11

In patients with and without hypertension, lowering sodium intake in conjunction with the DASH diet was associated with substantially larger reductions in systolic blood pressure.¹²

The recommendation to lower sodium intake has not changed in the guideline revisions. Meanwhile, other modifications have been added, such as incorporating both aerobic and resistance exercise and moderating alcohol intake. These recommendations have a class I level of evidence (ie, strongest level) in the 2017 ACC/AHA guidelines.¹

HYPERTENSION BEGINS AT 130/80

The definition of hypertension changed in the 2017 ACC/AHA guidelines¹: previously set at 140/90 mm Hg or higher, it is now 130/80 mm Hg or higher for all age groups. Adults with systolic blood pressure of 130 to 139 mm Hg or diastolic blood pressure of 80 to 89 mm Hg are now classified as having stage 1 hypertension.

Under the new definition, the number of US adults who have hypertension expanded to 45.6% of the general population,¹³ up from 31.9% under the JNC 7 definition. Thus, overall, 103.3 million US adults now have hypertension, compared with 72.2 million under the JNC 7 criteria.

In addition, the new guidelines expanded the population of adults for whom antihypertensive drug treatment is recommended to 36.2% (81.9 million). However, this represents only a 1.9% absolute increase over the JNC 7 recommendations (34.3%) and a 5.1% absolute increase over the JNC 8 recommendations.¹⁴

SPRINT: INTENSIVE TREATMENT IS BENEFICIAL

The new ACC/AHA guidelines¹ were based on evidence from several trials, including the Systolic Blood Pressure Intervention Trial (SPRINT).¹⁵

This multicenter trial investigated the effect of intensive blood pressure treatment on cardiovascular disease risk.¹⁶ The primary outcome was a composite of myocardial infarction, acute coronary syndrome, stroke, and heart failure.

The trial enrolled 9,361 participants at least 50 years of age with systolic blood pressure 130 mm Hg or higher and at least 1 additional risk factor for cardiovascular disease. It excluded anyone with a history of diabetes mellitus, stroke, symptomatic heart failure, or end-stage renal disease.

Two interventions were compared: 

• Intensive treatment, with a systolic blood pressure goal of less than 120 mm Hg: the protocol called for polytherapy, even for participants who were 75 or older if their blood pressure was 140 mm Hg or higher
• Standard treatment, with a systolic blood pressure goal of less than 140 mm Hg: it used polytherapy for patients whose systolic blood pressure was 160 mm Hg or higher.

The trial was intended to last 5 years but was stopped early at a median of 3.26 years owing to a significantly lower rate of the primary composite outcome in the intensive-treatment group: 1.65% per year vs 2.19%, a 25% relative risk reduction (P < .001) or a 0.54% absolute risk reduction. We calculate the number needed to treat (NNT) for 1 year to prevent 1 event as 185, and over the 3.26 years of the trial, the investigators calculated the NNT as 61. Similarly, the rate of death from any cause was also lower with intensive treatment, 1.03% per year vs 1.40% per year, a 27% relative risk reduction (P = .003) or a 0.37% absolute risk reduction, NNT 270.

Using these findings, Bress et al¹⁶ estimated that implementing intensive blood pressure goals could prevent 107,500 deaths annually.

The downside is adverse effects. In SPRINT,¹⁵ the intensive-treatment group experienced significantly higher rates of serious adverse effects than the standard-treatment group, ie:

• Hypotension 2.4% vs 1.4%, P = .001
• Syncope 2.3% vs 1.7%, P = .05
• Electrolyte abnormalities 3.1% vs 2.3%, P = .02)
• Acute kidney injury or kidney failure 4.1% vs 2.5%, P < .001
• Any treatment-related adverse event 4.7% vs 2.5%, P = .001.

Thus, Bress et al¹⁶ estimated that fully implementing the intensive-treatment goals could cause an additional 56,100 episodes of hypotension per year, 34,400 cases of syncope, 43,400 serious electrolyte disorders, and 88,700 cases of acute kidney injury. All told, about 3 million Americans could suffer a serious adverse effect under the intensive-treatment goals.

SPRINT¹⁵ was stopped early, after 3.26 years instead of the planned 5 years. The true risk-benefit ratio may have been different if the trial had been extended longer.

In addition, SPRINT used automated office blood pressure measurements in which patients were seated alone and a device (Model 907, Omron Healthcare) took 3 blood pressure measurements at 1-minute intervals after 5 minutes of quiet rest. This was designed to reduce elevated blood pressure readings in the presence of a healthcare professional in a medical setting (ie, “white coat” hypertension).

Many physicians are still taking blood pressure manually, which tends to give higher readings. Therefore, if they aim for a lower goal, they may risk overtreating the patient.

About 50% of patients did not achieve the target systolic blood pressure (< 120 mm Hg) despite receiving an average of 2.8 antihypertensive medications in the intensive-treatment group and 1.8 in the standard-treatment group. The use of antihypertensive medications, however, was not a controlled variable in the trial, and practitioners chose the appropriate drugs for their patients.

Diastolic pressure, which can be markedly lower in older hypertensive patients, was largely ignored, although lower diastolic pressure may have contributed to higher syncope rates in response to alpha blockers and calcium blockers.

Moreover, the trial excluded those with significant comorbidities and those younger than 50 (the mean age was 67.9), which limits the generalizability of the results.

JNC 8 VS SPRINT GOALS: WHAT'S THE EFFECT ON OUTCOMES?

JNC 8⁴ recommended a relaxed target of less than 140/90 mm Hg for adults younger than 60, including those with chronic kidney disease or diabetes, and less than 150/90 mm Hg for adults 60 and older. The SPRINT findings upended those recommendations, showing that intensive treatment in adults age 75 or older significantly improved the composite cardiovascular disease outcome (2.59 vs 3.85 events per year; P < .001) and all-cause mortality (1.78 vs 2.63 events per year; P < .05) compared with standard treatment.¹⁷ Also, a subset review of SPRINT trial data found no difference in benefit based on chronic kidney disease status.¹⁸

A meta-analysis of 74 clinical trials (N = 306,273) offers a compromise between the SPRINT findings and the JNC 8 recommendations.¹⁹ It found that the beneficial effect of blood pressure treatment depended on the patient’s baseline systolic blood pressure. In those with a baseline systolic pressure of 160 mm Hg or higher, treatment reduced cardiovascular mortality by about 15% (relative risk [RR] 0.85; 95% confidence interval [CI] 0.77–0.95). In patients with systolic pressure below 140 mm Hg, treatment effects were neutral (RR 1.03, 95% CI 0.87–1.20) and not associated with any benefit as primary prevention, although data suggest it may reduce the risk of adverse outcomes in patients with coronary heart disease.

OTHER TRIALS THAT INFLUENCED THE GUIDELINES

SHEP and HYVET (the Systolic Hypertension in the Elderly Program²⁰ and the Hypertension in the Very Elderly Trial)²¹ supported intensive blood pressure treatment for older patients by reporting a reduction in fatal and nonfatal stroke risks for those with a systolic blood pressure above 160 mm Hg.

FEVER (the Felodipine Event Reduction study)²² found that treatment with a calcium channel blocker in even a low dose can significantly decrease cardiovascular events, cardiovascular disease, and heart failure compared with no treatment.

JATOS and VALISH (the Japanese Trial to Assess Optimal Systolic Blood Pressure in Elderly Hypertensive Patients²³ and the Valsartan in Elderly Isolated Systolic Hypertension study)²⁴ found that outcomes were similar with intensive vs standard treatment.

Ettehad et al²⁵ performed a meta-analysis of 123 studies with more than 600,000 participants that provided strong evidence supporting blood pressure treatment goals below 130/90 mm Hg, in line with the SPRINT trial results.

BLOOD PRESSURE ISN’T EVERYTHING

Other trials remind us that although blood pressure is important, it is not the only factor affecting cardiovascular risk.

HOPE (the Heart Outcomes Prevention Evaluation)²⁶ investigated the use of ramipril (an ACE inhibitor) in preventing myocardial infarction, stroke, or cardiovascular death in patients at high risk of cardiovascular events. The study included 9,297 participants over age 55 (mean age 66) with a baseline blood pressure 139/79 mm Hg. Follow-up was 4.5 years.

Ramipril was better than placebo, with significantly fewer patients experiencing adverse end points in the ramipril group compared with the placebo group:

• Myocardial infarction 9.9% vs 12.3%, RR 0.80, P < .001
• Cardiovascular death 6.1% vs 8.1%, RR  0.74, P < .001
• Stroke 3.4% vs 4.9%, RR = .68, P < .001
• The composite end point 14.0% vs 17.8%, RR 0.78, P < .001).

Results were even better in the subset of patients who had diabetes.²⁷ However, the decrease in blood pressure attributable to anti­hypertensive therapy with ramipril was minimal (3–4 mm Hg systolic and 1–2 mm Hg diastolic). This slight change should not have been enough to produce significant differences in clinical outcomes, a major limitation of this trial. The investigators speculated that the positive results may be due to a class effect of ACE inhibitors.²⁶

HOPE 3^28–30 explored the effect of blood pressure- and cholesterol-controlling drugs on the same primary end points but in patients at intermediate risk of major cardiovascular events. Investigators randomized the 12,705 patients to 4 treatment groups:

• Blood pressure control with candesartan (an ARB) plus hydrochlorothiazide (a thiazide diuretic)
• Cholesterol control with rosuvastatin (a statin)
• Blood pressure plus cholesterol control
• Placebo.

Therapy was started at a systolic blood pressure above 140 mm Hg.

Compared with placebo, the rate of composite events was significantly reduced in the rosuvastatin group (3.7% vs 4.8%, HR 0.76, P = .002)²⁸ and the candesartan-hydrochlorothiazide-rosuvastatin group (3.6% vs 5.0%, HR 0.71; P = .005)²⁹ but not in the candesartan-hydrochlorothiazide group (4.1% vs 4.4%; HR 0.93; P = .40).³⁰

In addition, a subgroup analysis comparing active treatment vs placebo found a significant reduction in major cardiovascular events for treated patients whose baseline systolic blood pressure was in the upper third (> 143.5 mm Hg, mean 154.1 mm Hg), while treated patients in the lower middle and lower thirds had no significant reduction.³⁰

These results suggest that intensive treatment to achieve a systolic blood pressure below 140 mm Hg in patients at intermediate risk may not be helpful. Nevertheless, there seems to be agreement that intensive treatment generally leads to a reduction in cardiovascular events. The results also show the benefit of lowering cholesterol.

Bundy et al³¹ performed a meta-analysis that provides support for intensive antihypertensive treatment. Reviewing 42 clinical trials in more than 144,000 patients, they found that treating to reach a target systolic blood pressure of 120 to 124 mm Hg can reduce cardiovascular events and all-cause mortality.

The trade-off is a minimal increase in the risk of adverse events. Also, the risk-benefit ratio of intensive treatment seems to vary in different patient subgroups.

WHAT ABOUT PATIENTS WITH COMORBIDITIES?

The debate over intensive vs standard treatment in blood pressure management extends beyond hypertension and includes important comorbidities such as diabetes, stroke, and renal disease. Patients with a history of stroke or end-stage renal disease have only a minimal mention in the AHA/ACC guidelines.

Diabetes

Emdin et al,³² in a meta-analysis of 40 trials that included more than 100,000 patients with diabetes, concluded that a 10-mm Hg lowering of systolic blood pressure significantly reduces the rates of all-cause mortality, cardiovascular disease, coronary heart disease, stroke, albuminuria, and retinopathy. Stratifying the results according to the systolic blood pressure achieved (≥ 130 or < 130 mm Hg), the relative risks of mortality, coronary heart disease, cardiovascular disease, heart failure, and albuminuria were actually lower in the higher stratum than in the lower.

ACCORD (the Action to Control Cardiovascular Risk in Diabetes)³³ study provides contrary results. It examined intensive and standard blood pressure control targets in patients with type 2 diabetes at high risk of cardiovascular events, using primary outcome measures similar to those in SPRINT. It found no significant difference in fatal and nonfatal cardiovascular events between the intensive and standard blood pressure target arms.

Despite those results, the ACC/AHA guidelines still advocate for more intensive treatment (goal < 130/80 mm Hg) in all patients, including those with diabetes.¹

The ADA position statement (September 2017) recommended a target below 140/90 mm Hg in patients with diabetes and hypertension.⁸ However, they also noted that lower systolic and diastolic blood pressure targets, such as below 130/80 mm Hg, may be appropriate for patients at high risk of cardiovascular disease “if they can be achieved without undue treatment burden.”⁸ Thus, it is not clear which blood pressure targets in patients with diabetes are the best.

Stroke

In patients with stroke, AHA/ACC guidelines¹ recommend treatment if the blood pressure is 140/90 mm Hg or higher because antihypertensive therapy has been associated with a decrease in the recurrence of transient ischemic attack and stroke. The ideal target blood pressure is not known, but a goal of less than 130/80 mm Hg may be reasonable.

In the Secondary Prevention of Small Subcortical Strokes (SPS3) trial, a retrospective open-label trial, a target blood pressure below 130/80 mm Hg in patients with a history of lacunar stroke was associated with a lower risk of intracranial hemorrhage, but the difference was not statistically significant.³⁴ For this reason, the ACC/AHA guidelines consider it reasonable to aim for a systolic blood pressure below 130 mm Hg in these patients.¹

Renal disease

The ACC/AHA guidelines do not address how to manage hypertension in patients with end-stage renal disease, but for patients with chronic kidney disease they recommend a blood pressure target below 130/80 mm Hg.¹ This recommendation is derived from the SPRINT trial,¹⁵ in which patients with stage 3 or 4 chronic kidney disease accounted for 28% of the study population. In that subgroup, intensive blood pressure control seemed to provide the same benefits for reduction in cardiovascular death and all-cause mortality.

TREAT PATIENTS, NOT NUMBERS

Blood pressure targets should be applied in the appropriate clinical context and on a patient-by-patient basis. In clinical practice, one size does not always fit all, as special cases exist.

For example, blood pressure can oscillate widely in patients with autonomic nerve disorders, making it difficult to strive for a specific target, especially an intensive one. Thus, it may be necessary to allow higher systolic blood pressure in these patients. Similarly, patients with diabetes or chronic kidney disease may be at higher risk of kidney injury with more intensive blood pressure management.

Treating numbers rather than patients may result in unbalanced patient care. The optimal approach to blood pressure management relies on a comprehensive risk factor assessment and shared decision-making with the patient before setting specific blood pressure targets.

OUR APPROACH

We aim for a blood pressure goal below 130/80 mm Hg for all patients with cardiovascular disease, according to the AHA/ACC guidelines. We aim for that same target in patients without cardiovascular disease but who have an elevated estimated cardiovascular risk (> 10%) over the next 10 years.

We recognize, however, that the benefits of aggressive blood pressure reduction may not be as clear in all patients, such as those with diabetes. We also recognize that some patient subgroups are at high risk of adverse events, including those with low diastolic pressure, chronic kidney disease, a history of falls, and older age. In those patients, we are extremely judicious when titrating antihypertensive medications. We often make smaller titrations, at longer intervals, and with more frequent laboratory testing and in-office follow-up.

Our process of managing hypertension through intensive blood pressure control to achieve lower systolic blood pressure targets requires a concerted effort among healthcare providers at all levels. It especially requires more involvement and investment from primary care providers to individualize treatment in their patients. This process has helped us to reach our treatment goals while limiting adverse effects of lower blood pressure targets.

MOVING FORWARD

Hypertension is a major risk factor for cardiovascular disease, and intensive blood pressure control has the potential to significantly reduce rates of morbidity and death associated with cardiovascular disease. Thus, a general consensus on the definition of hypertension and treatment goals is essential to reduce the risk of cardiovascular events in this large patient population.

Intensive blood pressure treatment has shown efficacy, but it has a small accompanying risk of adverse events, which varies in patient subgroups and affects the benefit-risk ratio of this therapy. For example, the cardiovascular benefit of intensive treatment is less clear in diabetic patients, and the risk of adverse events may be higher in older patients with chronic kidney disease.

Moving forward, more research is needed into the effects of intensive and standard treatment on patients of all ages, those with common comorbid conditions, and those with other important factors such as diastolic hypertension.

Finally, the various medical societies should collaborate on hypertension guideline development. This would require considerable planning and coordination but would ultimately be useful in creating a generalizable approach to hypertension management.

Breast concerns account for approximately 3% of all female visits to a primary care practice.¹ The most common symptoms are breast lumps and breast pain.

Because breast cancer is the most common malignancy in women in the United States, affecting nearly 1 in 8 women in their lifetime, women with breast problems often fear the worst. However, only about 3.5% of women reporting a concern have cancer; most problems are benign (Table 1).¹

Here, we present an evidence-based review of common breast problems in primary care practice and discuss how to evaluate and manage them.

GENERAL APPROACH

The evaluation of a breast concern requires a systematic approach, beginning with a history that documents the onset, severity, and frequency of symptoms. If the concern is a lump or mass, ask whether it becomes more tender or increases in size at any point during the menstrual cycle.

Focus the physical examination on the cervical, supraclavicular, infraclavicular, and axillary lymph nodes and on the breast itself. Assess breast symmetry, note any skin changes such as dimpling, and check the nipples for discharge and inversion. Palpate the breasts for masses.

PALPABLE BREAST MASS: IMAGING NEEDED

If a mass is present, it is more likely to be malignant if any of the following is true:

• Firm to hard texture or indistinct margins
• Attached to the underlying deep fascia or skin
• Associated nipple inversion or skin dimpling.²

Breast masses are more likely benign if they have discrete, well-defined margins, are mobile with a soft to rubbery consistency, and change with the menstrual cycle. However, clinical features are unreliable indicators of cause, and thus additional investigation with breast imaging is warranted.

Mammography remains the diagnostic test of choice for all women age 30 or older who have a palpable breast mass. It is less effective in younger women because they are more likely to have extremely dense fibroglandular tissue that will limit its sensitivity to imaging.

Order diagnostic mammography, which includes additional views focused on the area of concern, rather than screening mammography, which includes only standard cranio­caudal and mediolateral oblique views. A skin marker should be applied over the palpable lump to aid imaging. Because a breast that contains a mass may be denser than the opposite breast or may show asymmetry, both breasts should be imaged. The sensitivity of diagnostic mammography varies from 85% to 90%, so a negative mammogram does not rule out malignancy.^2,3

Targeted ultrasonography of the palpable mass helps identify solid masses such as fibroadenomas or malignant tumors, classifies the margins (lobulated, smooth, or irregular), and assesses vascularity. Ultrasonography is particularly useful for characterizing cystic lesions (eg, simple, septated, or clustered cysts) and cysts with internal echoes. It can also identify lipomas or sebaceous cysts.

If the findings on both mammography and ultrasonography are benign, the likelihood of cancer is very low, with an estimated negative predictive value of 97% to 100%.^2,3 Additionally, the likelihood of nonmalignant findings on biopsy after benign imaging is approximately 99%.³

Although radiologic imaging can define palpable masses, it is intended as a clinical aid. Suspicious findings on clinical examination should never be ignored even if findings on imaging are reassuring, as studies have documented that about 5% of breast cancers may be detected on clinical breast examination alone.⁴

Other imaging tests such as magnetic resonance imaging may be considered occasionally if clinical suspicion remains high after negative mammography and ultrasonography, but they cannot confirm a diagnosis of malignancy. In that case, refer the patient to a surgeon for consideration of excisional biopsy.

Patients with an indeterminate lesion can return in 3 to 12 weeks for a follow-up examination and repeat imaging, which helps assess interval clinical stability. The latter option is especially helpful for patients with masses that are of low suspicion or for patients who prefer to avoid invasive tissue biopsy.

Patients with clinical and radiologic findings that suggest a benign cause can return for short-term follow-up in 6 months or in 12 months for their regular mammogram.

More than 50% of women experience breast pain at some point in their life.⁵ Of these, 35% report that the pain adversely affects their sleep, and 41% note that the pain detrimentally affects their sexual quality of life. Up to 66% of breast pain correlates directly with the patient’s menstrual cycle.⁵ Breast pain is rarely associated with malignancy.

Regardless of its severity and the low likelihood of malignancy, breast pain can be a significant source of distress for the patient, primarily because of concerns about underlying malignancy. If the patient has a focal area of pain on examination, order mammography in combination with targeted ultrasonography. The sensitivity and negative predictive value of benign findings on combination mammography and ultrasonography in this setting are as high as 100%. The incidence of underlying cancer in patients with focal breast pain and no palpable mass is approximately 1.2%.⁶

The long-term prognosis in women with diffuse, often bilateral breast pain (in the absence of additional clinical findings) is excellent. In one study, the incidence of a breast cancer diagnosis was 1.8% after a median of 51 months of follow-up.⁷ Therefore, patients presenting with diffuse pain, no palpable abnormalities, and benign imaging can be safely reassured. Magnetic resonance imaging is rarely indicated in patients with breast pain unless other clinical findings, such as a mass or skin changes, are noted and the results of mammography and ultrasonography are negative.

Treating breast pain

Treating breast pain remains a challenge. The first step is to reassure the patient about her prognosis and help her make appropriate lifestyle modifications.

A well-fitting bra. Suggest getting a professional bra fitting. Wearing a well-fitted bra that offers lift, support, and compression and reduces excess motion can help improve benign breast pain. A bra fitting is especially important for women with large breasts because it can be difficult for these women to get an accurate size. Wearing a lightly fitted bra at night may also provide comfort if there is nighttime pain with breast tissue movement.

Reducing daily caffeine intake is often advised for pain management, but strong evidence of its efficacy is lacking.

Anti-inflammatory drugs can be beneficial if used short-term, especially if costochondritis is suspected.

Danazol improves pain in more than 70% of patients with cyclical symptoms and in up to 48% of those with noncyclical symptoms.

Bromocriptine is effective in up to 54% of those with cyclical symptoms and in up to 33% of those with noncyclical symptoms.⁸ However, the US Food and Drug Administration (FDA) withdrew approval for this indication because of adverse effects.

Tamoxifen, in contrast, provides relief in 94% of those with cyclical symptoms and in 56% of those with noncyclical symptoms.⁹

Adverse effects, however, limit the use of danazol, bromocriptine, and tamoxifen, and they should be prescribed only for short-term use (3 to 6 months) and only in women with chronic debilitating pain.

A few small studies have evaluated alternative options.

Toremifene is a triphenylethylene derivative similar to tamoxifen that is also used in the adjuvant treatment of postmenopausal breast cancer (but with fewer adverse effects). It has been documented to have a significant effect on premenstrual breast pain, with a 64% reduction in breast pain scores compared with a 26% reduction with placebo.¹⁰ However, the FDA has not approved it for this indication, and it can be cost-prohibitive.

Over-the-counter medications that may provide relief for cyclic breast pain include vitamin E or B₆, products containing oil of Vitex agnus castus (chaste tree or chasteberry), and flaxseed.^11,12

Acupuncture has been evaluated in patients with noncyclic breast pain and was found to reduce pain by 56% to 67% in one study,¹³ although it did not affect quality of life.

NIPPLE DISCHARGE

From 5% to 7% of women seek medical attention for nipple discharge.^14,15 Breast cancer is found in 5% to 15% of women who undergo surgery for nipple discharge.^16,17

Review the patient’s current medications and inquire about health conditions such as thyroid dysfunction or visual field changes that suggest a pituitary mass (which can lead to nipple discharge by causing hormonal dysregulation or hyperprolactinemia).

Palpate the breasts for an underlying mass, look for lesions on the nipple, and assess the color of the fluid. Also note whether there is discharge from one or both breasts, whether it is spontaneous or expressive, and whether it occurs from a single or multiple ducts. Nipple lesions may require further testing with punch biopsy.

Nonlactational nipple discharge is classified as physiologic or pathologic. Physiologic nipple discharge is typically bilateral, involving multiple ducts, and is often clear or straw-colored but may also be green, gray, or brown.

White, opaque fluid is often related to galactorrhea as a result of hyperprolactinemia, hypothyroidism, or medications such as antipsychotic drugs (eg, haloperidol and fluphenazine) and gastrointestinal motility agents such as metoclopramide. Discharge also commonly results from benign underlying ductal abnormalities such as intraductal papilloma, periductal mastitis, and duct ectasia.

Pathologic nipple discharge is often unilateral and persistent, occurring spontaneously from a solitary duct, and may be bloody or serous.

For women with pathologic nipple discharge who are 30 or older, diagnostic imaging with mammography and subareolar ultrasonography is recommended. If the patient is younger than 30, ultrasonography of the subareolar region alone can be used. Targeted ultrasonography of any palpable area is also advised.

Cytologic assessment of the fluid is not recommended because it can often lead to a false-positive finding of atypical cells. Imaging studies such as ductography, duct lavage, ductoscopy, and magnetic resonance imaging are also generally unnecessary; instead, a persistent clinical concern should prompt a surgical referral for consideration of duct excision.

When a patient has pathologic nipple discharge with a negative physical examination and breast imaging, studies have shown that the risk of cancer is 3% or less.¹⁸

Patients with spontaneous bloody or serous single-duct discharge with negative results on mammography and ultrasonography should be reassured that they have a low risk of underlying cancer. If the patient prefers, one approachto management is follow-up mammography and ultrasonography at 6 months and clinical examination for up to 2 years or until the discharge resolves on its own.

On the other hand, if the discharge is distressing to the patient, subareolar duct excision can be performed with both a diagnostic and therapeutic purpose.

A rash on the nipple or areolar region warrants careful evaluation because it may be the first sign of Paget disease of the breast.

In the clinical breast examination, assess the extent of the rash and the presence of any underlying breast mass or nipple discharge. Dermatitis often starts on the areola and resolves quickly with topical therapy. However, Paget disease tends to start directly on the nipple itself, is unresponsive or only partially responsive to topical therapy, and progresses gradually, leading to erosions and ultimately effacement of the nipple itself.

If the clinical examination suggests mild dermatitis and the results of breast imaging are negative, treat the patient with a topical medication because benign conditions such as dermatitis and eczema are common. However, continued follow-up is mandatory until the rash completely resolves: Paget disease sometimes initially improves with topical therapy due to its inflammatory nature.

If you suspect Paget disease or the rash does not fully resolve after 2 to 3 weeks of topical therapy, refer the patient to a dermatologist for full-thickness punch biopsy to establish the diagnosis.

Paget disease of the breast may or may not be associated with underlying ductal carcinoma in situ or invasive breast cancer.¹⁹ The absence of clinical or imaging abnormalities in a patient with Paget disease does not rule out underlying malignancy.²⁰

DENSE BREASTS

Increased breast density has been shown to be a risk factor for breast cancer and may be prognostically useful when combined with the Tyrer-Cuzick model or the Gail model of breast cancer risk.²⁴

Additionally, increased density can mask cancers on mammography, significantly reducing its sensitivity. In women with heterogeneously or extremely dense breasts, the sensitivity of mammography for detecting cancer is only 25% to 50%.²¹ Due to this low sensitivity, supplemental imaging is helpful, particularly in women already at risk of breast cancer based on family history.

Supplemental screening

Digital mammography with tomosynthesis was approved by the FDA in 2011 for use in combination with standard digital mammography for breast cancer screening. Compared with traditional 2-dimensional mammography alone, adding 3-D tomosynthesis decreases the recall rate and increases the cancer detection rate.²⁵

Tomosynthesis tends to perform better in women with heterogeneously dense breasts (BI-RADS category C). There is no significant improvement in cancer detection in women with extremely dense breasts (BI-RADS category D).²⁶

Depending on the methodology, radiation exposure can be either higher or lower than with traditional mammography. However, in all forms, the very small amount of radiation is considered safe.

Whole breast ultrasonography. When whole breast ultrasonography is used to supplement mammography, the recall rate is higher than when mammography is used alone (14% vs 7%–11%).²² It also increases the cancer detection rate by 4.4 additional cancers per 1,000 examinations. However, the false-positive rate with whole breast ultrasonography is higher; the positive predictive value of combined mammography and ultrasonography is 11.2% vs 22.6% for mammography alone.²² Therefore, we do not generally recommend whole breast ultrasonography as a supplement to mammography in women with dense breast tissue unless other studies are not an option.

Molecular breast imaging is not widely available because it requires special equipment, injection of a radiopharamceutical (technetium Tc 99m sestamibi), and a radiologist who specializes in breast imaging to interpret the results. When it is available, however, it increases the cancer detection rate by 8.8 in 1,000 examinations; the positive predictive value is similar to that of screening mammography alone.²¹ It is particularly useful in patients with dense breasts who do not qualify for screening magnetic resonance imaging (lifetime risk of < 20% to 25%).

Technetium sestamibi is associated with a minimal amount of radiation exposure (2.4 mSv vs 1.2 mSV with standard mammography). However, this exposure is much less than background radiation exposure and is considered safe.²¹

Screening mammography can disclose abnormalities such as calcifications, masses, asymmetry, or architectural distortion.²⁷ Abnormalities are reported using standardized BI-RADS categories designated with the numbers 0 through 6 (Table 3).²³

A report of BI-RADS category 0 (incomplete), 4 (suspicious), or 5 (highly suspicious) requires additional workup.

Category 1 (negative) requires no further follow-up, and the patient should resume age-appropriate screening.

For patients with category 2 (benign) findings, routine screening is recommended, whereas patients with category 3 (probably benign) are advised to come back in 6 months for follow-up imaging.

Diagnostic mammography includes additional assessments for focal symptoms or areas of abnormality noted on screening imaging or clinical examination. These may include spot magnification views of areas of asymmetry, mass, architectural distortion, or calcifications. Ultrasonography of focal breast abnormalities can help determine if there is an underlying cyst or solid mass.

MANAGEMENT OF BENIGN FINDINGS ON BREAST BIOPSY

Benign breast disease is diagnosed when a patient with a palpable or radiographic abnormality undergoes breast biopsy with benign findings.^28,29 It can be largely grouped into 3 categories: nonproliferative, proliferative without atypia, and proliferative with atypia (Table 4).^28,29

If core-needle biopsy study results are benign, the next step is to establish radiologic-pathologic and clinical-pathologic concordance. If the findings on clinical examination or imaging are not consistent with those on pathologic study, excisional biopsy should be performed, as imaging-directed biopsy may not have adequately sampled the lesion.³⁰

Nonproliferative lesions account for about 65% of findings on core-needle biopsy and include simple cysts, fibroadenomas, columnar cell changes, apocrine metaplasia, and mild ductal hyperplasia of the usual type. These lesions do not significantly increase the risk of breast cancer; the relative risk is 1.2 to 1.4.^28,29 Additionally, the risk of “upstaging” after excisional biopsy—ie, to a higher-risk lesion or to malignancy—is minimal. Therefore, no additional action is necessary when these findings alone are noted on core-needle biopsy.

Proliferative lesions without atypia account for about 30% of biopsy results and include usual ductal hyperplasia, sclerosing adenosis, columnar hyperplasia, papilloma, and radial scar. Generally, there is a slightly increased risk of subsequent breast cancer, with a relative risk of 1.7 to 2.1.²⁸ Usual ductal hyperplasia and columnar hyperplasia have little risk of upstaging with excision, and therefore, surgical consultation is not recommended.

Previously, surgical excision was recommended for any intraductal papilloma due to risk of upgrade in pathologic diagnosis at the time of excision. However, more recent data suggest that the upgrade rate is about 2.2% for a solitary papilloma that is less than 1 cm in diameter and without associated mass lesion (either clinically or radiographically), is concordant with radiographic findings, and has no associated atypical cells on biopsy.³¹ In this case, observation and short-interval clinical follow-up are reasonable. If there are multiple papillomas, the patient has symptoms such as persistent bloody nipple discharge, or any of the above criteria are not met, surgical excision is recommended.²⁸

Similarly, radial scars and complex sclerosing lesions are increasingly likely to be associated with malignancy based on size. Upstaging ranges from 0% to 12%. It is again important when evaluating radial scars that there is pathologic concordance and that there were no associated high-risk lesions on pathology. If this is the case, it is reasonable to clinically monitor patients with small radial scars, particularly in those who do not have an elevated risk of developing breast cancer.³⁰

For all patients who have undergone biopsy and whose pathology study results are benign, a thorough risk evaluation should be performed, including calculation of their lifetime risk of breast cancer. This can be done with the National Cancer Institute Breast Cancer Risk Assessment Tool, the International Breast Cancer Intervention Study (IBIS) risk calculator, or other model using family history as a basis for calculations. Patients found to have a lifetime risk of breast cancer of greater than 20% to 25% should be offered annual screening with magnetic resonance imaging in addition to mammography.

ATYPICAL HYPERPLASIA: INCREASED RISK

When biopsy study shows atypical ductal hyperplasia or atypical lobular hyperplasia, there is an increased risk of breast cancer.^28,32 The absolute overall risk of developing breast cancer in 25 years is 30%, and that risk is further stratified based on the number of foci of atypia noted in the specimen.²⁹

When core-needle biopsy study reveals atypical ductal hyperplasia in the tissue, there is a 15% to 30% risk of finding breast cancer with surgical excision.²⁸ Surgical excision is therefore recommended for atypical ductal hyperplasia noted on core-needle biopsy.²⁸

In contrast, when atypical lobular hyperplasia alone is noted, the risk of upstagingto malignancy varies widely—from 0% to 67%—although recent studies have noted risks of 1% to 3%.^33,34 Thus, the decision for surgical excision is more variable. Generally, if the atypical lobular hyperplasia is noted incidentally, is not associated with a higher grade lesion, and is concordant with imaging, it is reasonable to closely monitor with serial imaging and physical examination. Excision is unnecessary.³⁵

Patients found to have atypical hyperplasia on breast biopsy should receive counseling about risk-reducing medications. Selective estrogen receptor modulators such as tamoxifen and raloxifene have been shown to reduce the risk of breast cancer by as much as 86% in patients with atypical hyperplasia.³⁶ Similarly, aromatase inhibitors such as exemestane and anastrozole reduce breast cancer risk by approximately 65%.³⁷

Breast concerns account for approximately 3% of all female visits to a primary care practice.¹ The most common symptoms are breast lumps and breast pain.

Because breast cancer is the most common malignancy in women in the United States, affecting nearly 1 in 8 women in their lifetime, women with breast problems often fear the worst. However, only about 3.5% of women reporting a concern have cancer; most problems are benign (Table 1).¹

Here, we present an evidence-based review of common breast problems in primary care practice and discuss how to evaluate and manage them.

GENERAL APPROACH

The evaluation of a breast concern requires a systematic approach, beginning with a history that documents the onset, severity, and frequency of symptoms. If the concern is a lump or mass, ask whether it becomes more tender or increases in size at any point during the menstrual cycle.

Focus the physical examination on the cervical, supraclavicular, infraclavicular, and axillary lymph nodes and on the breast itself. Assess breast symmetry, note any skin changes such as dimpling, and check the nipples for discharge and inversion. Palpate the breasts for masses.

PALPABLE BREAST MASS: IMAGING NEEDED

If a mass is present, it is more likely to be malignant if any of the following is true:

• Firm to hard texture or indistinct margins
• Attached to the underlying deep fascia or skin
• Associated nipple inversion or skin dimpling.²

Breast masses are more likely benign if they have discrete, well-defined margins, are mobile with a soft to rubbery consistency, and change with the menstrual cycle. However, clinical features are unreliable indicators of cause, and thus additional investigation with breast imaging is warranted.

Mammography remains the diagnostic test of choice for all women age 30 or older who have a palpable breast mass. It is less effective in younger women because they are more likely to have extremely dense fibroglandular tissue that will limit its sensitivity to imaging.

Order diagnostic mammography, which includes additional views focused on the area of concern, rather than screening mammography, which includes only standard cranio­caudal and mediolateral oblique views. A skin marker should be applied over the palpable lump to aid imaging. Because a breast that contains a mass may be denser than the opposite breast or may show asymmetry, both breasts should be imaged. The sensitivity of diagnostic mammography varies from 85% to 90%, so a negative mammogram does not rule out malignancy.^2,3

Targeted ultrasonography of the palpable mass helps identify solid masses such as fibroadenomas or malignant tumors, classifies the margins (lobulated, smooth, or irregular), and assesses vascularity. Ultrasonography is particularly useful for characterizing cystic lesions (eg, simple, septated, or clustered cysts) and cysts with internal echoes. It can also identify lipomas or sebaceous cysts.

If the findings on both mammography and ultrasonography are benign, the likelihood of cancer is very low, with an estimated negative predictive value of 97% to 100%.^2,3 Additionally, the likelihood of nonmalignant findings on biopsy after benign imaging is approximately 99%.³

Although radiologic imaging can define palpable masses, it is intended as a clinical aid. Suspicious findings on clinical examination should never be ignored even if findings on imaging are reassuring, as studies have documented that about 5% of breast cancers may be detected on clinical breast examination alone.⁴

Other imaging tests such as magnetic resonance imaging may be considered occasionally if clinical suspicion remains high after negative mammography and ultrasonography, but they cannot confirm a diagnosis of malignancy. In that case, refer the patient to a surgeon for consideration of excisional biopsy.

Patients with an indeterminate lesion can return in 3 to 12 weeks for a follow-up examination and repeat imaging, which helps assess interval clinical stability. The latter option is especially helpful for patients with masses that are of low suspicion or for patients who prefer to avoid invasive tissue biopsy.

Patients with clinical and radiologic findings that suggest a benign cause can return for short-term follow-up in 6 months or in 12 months for their regular mammogram.

More than 50% of women experience breast pain at some point in their life.⁵ Of these, 35% report that the pain adversely affects their sleep, and 41% note that the pain detrimentally affects their sexual quality of life. Up to 66% of breast pain correlates directly with the patient’s menstrual cycle.⁵ Breast pain is rarely associated with malignancy.

Regardless of its severity and the low likelihood of malignancy, breast pain can be a significant source of distress for the patient, primarily because of concerns about underlying malignancy. If the patient has a focal area of pain on examination, order mammography in combination with targeted ultrasonography. The sensitivity and negative predictive value of benign findings on combination mammography and ultrasonography in this setting are as high as 100%. The incidence of underlying cancer in patients with focal breast pain and no palpable mass is approximately 1.2%.⁶

The long-term prognosis in women with diffuse, often bilateral breast pain (in the absence of additional clinical findings) is excellent. In one study, the incidence of a breast cancer diagnosis was 1.8% after a median of 51 months of follow-up.⁷ Therefore, patients presenting with diffuse pain, no palpable abnormalities, and benign imaging can be safely reassured. Magnetic resonance imaging is rarely indicated in patients with breast pain unless other clinical findings, such as a mass or skin changes, are noted and the results of mammography and ultrasonography are negative.

Treating breast pain

Treating breast pain remains a challenge. The first step is to reassure the patient about her prognosis and help her make appropriate lifestyle modifications.

A well-fitting bra. Suggest getting a professional bra fitting. Wearing a well-fitted bra that offers lift, support, and compression and reduces excess motion can help improve benign breast pain. A bra fitting is especially important for women with large breasts because it can be difficult for these women to get an accurate size. Wearing a lightly fitted bra at night may also provide comfort if there is nighttime pain with breast tissue movement.

Reducing daily caffeine intake is often advised for pain management, but strong evidence of its efficacy is lacking.

Anti-inflammatory drugs can be beneficial if used short-term, especially if costochondritis is suspected.

Danazol improves pain in more than 70% of patients with cyclical symptoms and in up to 48% of those with noncyclical symptoms.

Bromocriptine is effective in up to 54% of those with cyclical symptoms and in up to 33% of those with noncyclical symptoms.⁸ However, the US Food and Drug Administration (FDA) withdrew approval for this indication because of adverse effects.

Tamoxifen, in contrast, provides relief in 94% of those with cyclical symptoms and in 56% of those with noncyclical symptoms.⁹

Adverse effects, however, limit the use of danazol, bromocriptine, and tamoxifen, and they should be prescribed only for short-term use (3 to 6 months) and only in women with chronic debilitating pain.

A few small studies have evaluated alternative options.

Toremifene is a triphenylethylene derivative similar to tamoxifen that is also used in the adjuvant treatment of postmenopausal breast cancer (but with fewer adverse effects). It has been documented to have a significant effect on premenstrual breast pain, with a 64% reduction in breast pain scores compared with a 26% reduction with placebo.¹⁰ However, the FDA has not approved it for this indication, and it can be cost-prohibitive.

Over-the-counter medications that may provide relief for cyclic breast pain include vitamin E or B₆, products containing oil of Vitex agnus castus (chaste tree or chasteberry), and flaxseed.^11,12

Acupuncture has been evaluated in patients with noncyclic breast pain and was found to reduce pain by 56% to 67% in one study,¹³ although it did not affect quality of life.

NIPPLE DISCHARGE

From 5% to 7% of women seek medical attention for nipple discharge.^14,15 Breast cancer is found in 5% to 15% of women who undergo surgery for nipple discharge.^16,17

Review the patient’s current medications and inquire about health conditions such as thyroid dysfunction or visual field changes that suggest a pituitary mass (which can lead to nipple discharge by causing hormonal dysregulation or hyperprolactinemia).

Palpate the breasts for an underlying mass, look for lesions on the nipple, and assess the color of the fluid. Also note whether there is discharge from one or both breasts, whether it is spontaneous or expressive, and whether it occurs from a single or multiple ducts. Nipple lesions may require further testing with punch biopsy.

Nonlactational nipple discharge is classified as physiologic or pathologic. Physiologic nipple discharge is typically bilateral, involving multiple ducts, and is often clear or straw-colored but may also be green, gray, or brown.

White, opaque fluid is often related to galactorrhea as a result of hyperprolactinemia, hypothyroidism, or medications such as antipsychotic drugs (eg, haloperidol and fluphenazine) and gastrointestinal motility agents such as metoclopramide. Discharge also commonly results from benign underlying ductal abnormalities such as intraductal papilloma, periductal mastitis, and duct ectasia.

Pathologic nipple discharge is often unilateral and persistent, occurring spontaneously from a solitary duct, and may be bloody or serous.

For women with pathologic nipple discharge who are 30 or older, diagnostic imaging with mammography and subareolar ultrasonography is recommended. If the patient is younger than 30, ultrasonography of the subareolar region alone can be used. Targeted ultrasonography of any palpable area is also advised.

Cytologic assessment of the fluid is not recommended because it can often lead to a false-positive finding of atypical cells. Imaging studies such as ductography, duct lavage, ductoscopy, and magnetic resonance imaging are also generally unnecessary; instead, a persistent clinical concern should prompt a surgical referral for consideration of duct excision.

When a patient has pathologic nipple discharge with a negative physical examination and breast imaging, studies have shown that the risk of cancer is 3% or less.¹⁸

Patients with spontaneous bloody or serous single-duct discharge with negative results on mammography and ultrasonography should be reassured that they have a low risk of underlying cancer. If the patient prefers, one approachto management is follow-up mammography and ultrasonography at 6 months and clinical examination for up to 2 years or until the discharge resolves on its own.

On the other hand, if the discharge is distressing to the patient, subareolar duct excision can be performed with both a diagnostic and therapeutic purpose.

A rash on the nipple or areolar region warrants careful evaluation because it may be the first sign of Paget disease of the breast.

In the clinical breast examination, assess the extent of the rash and the presence of any underlying breast mass or nipple discharge. Dermatitis often starts on the areola and resolves quickly with topical therapy. However, Paget disease tends to start directly on the nipple itself, is unresponsive or only partially responsive to topical therapy, and progresses gradually, leading to erosions and ultimately effacement of the nipple itself.

If the clinical examination suggests mild dermatitis and the results of breast imaging are negative, treat the patient with a topical medication because benign conditions such as dermatitis and eczema are common. However, continued follow-up is mandatory until the rash completely resolves: Paget disease sometimes initially improves with topical therapy due to its inflammatory nature.

If you suspect Paget disease or the rash does not fully resolve after 2 to 3 weeks of topical therapy, refer the patient to a dermatologist for full-thickness punch biopsy to establish the diagnosis.

Paget disease of the breast may or may not be associated with underlying ductal carcinoma in situ or invasive breast cancer.¹⁹ The absence of clinical or imaging abnormalities in a patient with Paget disease does not rule out underlying malignancy.²⁰

DENSE BREASTS

Increased breast density has been shown to be a risk factor for breast cancer and may be prognostically useful when combined with the Tyrer-Cuzick model or the Gail model of breast cancer risk.²⁴

Additionally, increased density can mask cancers on mammography, significantly reducing its sensitivity. In women with heterogeneously or extremely dense breasts, the sensitivity of mammography for detecting cancer is only 25% to 50%.²¹ Due to this low sensitivity, supplemental imaging is helpful, particularly in women already at risk of breast cancer based on family history.

Supplemental screening

Digital mammography with tomosynthesis was approved by the FDA in 2011 for use in combination with standard digital mammography for breast cancer screening. Compared with traditional 2-dimensional mammography alone, adding 3-D tomosynthesis decreases the recall rate and increases the cancer detection rate.²⁵

Tomosynthesis tends to perform better in women with heterogeneously dense breasts (BI-RADS category C). There is no significant improvement in cancer detection in women with extremely dense breasts (BI-RADS category D).²⁶

Depending on the methodology, radiation exposure can be either higher or lower than with traditional mammography. However, in all forms, the very small amount of radiation is considered safe.

Whole breast ultrasonography. When whole breast ultrasonography is used to supplement mammography, the recall rate is higher than when mammography is used alone (14% vs 7%–11%).²² It also increases the cancer detection rate by 4.4 additional cancers per 1,000 examinations. However, the false-positive rate with whole breast ultrasonography is higher; the positive predictive value of combined mammography and ultrasonography is 11.2% vs 22.6% for mammography alone.²² Therefore, we do not generally recommend whole breast ultrasonography as a supplement to mammography in women with dense breast tissue unless other studies are not an option.

Molecular breast imaging is not widely available because it requires special equipment, injection of a radiopharamceutical (technetium Tc 99m sestamibi), and a radiologist who specializes in breast imaging to interpret the results. When it is available, however, it increases the cancer detection rate by 8.8 in 1,000 examinations; the positive predictive value is similar to that of screening mammography alone.²¹ It is particularly useful in patients with dense breasts who do not qualify for screening magnetic resonance imaging (lifetime risk of < 20% to 25%).

Technetium sestamibi is associated with a minimal amount of radiation exposure (2.4 mSv vs 1.2 mSV with standard mammography). However, this exposure is much less than background radiation exposure and is considered safe.²¹

Screening mammography can disclose abnormalities such as calcifications, masses, asymmetry, or architectural distortion.²⁷ Abnormalities are reported using standardized BI-RADS categories designated with the numbers 0 through 6 (Table 3).²³

A report of BI-RADS category 0 (incomplete), 4 (suspicious), or 5 (highly suspicious) requires additional workup.

Category 1 (negative) requires no further follow-up, and the patient should resume age-appropriate screening.

For patients with category 2 (benign) findings, routine screening is recommended, whereas patients with category 3 (probably benign) are advised to come back in 6 months for follow-up imaging.

Diagnostic mammography includes additional assessments for focal symptoms or areas of abnormality noted on screening imaging or clinical examination. These may include spot magnification views of areas of asymmetry, mass, architectural distortion, or calcifications. Ultrasonography of focal breast abnormalities can help determine if there is an underlying cyst or solid mass.

MANAGEMENT OF BENIGN FINDINGS ON BREAST BIOPSY

Benign breast disease is diagnosed when a patient with a palpable or radiographic abnormality undergoes breast biopsy with benign findings.^28,29 It can be largely grouped into 3 categories: nonproliferative, proliferative without atypia, and proliferative with atypia (Table 4).^28,29

If core-needle biopsy study results are benign, the next step is to establish radiologic-pathologic and clinical-pathologic concordance. If the findings on clinical examination or imaging are not consistent with those on pathologic study, excisional biopsy should be performed, as imaging-directed biopsy may not have adequately sampled the lesion.³⁰

Nonproliferative lesions account for about 65% of findings on core-needle biopsy and include simple cysts, fibroadenomas, columnar cell changes, apocrine metaplasia, and mild ductal hyperplasia of the usual type. These lesions do not significantly increase the risk of breast cancer; the relative risk is 1.2 to 1.4.^28,29 Additionally, the risk of “upstaging” after excisional biopsy—ie, to a higher-risk lesion or to malignancy—is minimal. Therefore, no additional action is necessary when these findings alone are noted on core-needle biopsy.

Proliferative lesions without atypia account for about 30% of biopsy results and include usual ductal hyperplasia, sclerosing adenosis, columnar hyperplasia, papilloma, and radial scar. Generally, there is a slightly increased risk of subsequent breast cancer, with a relative risk of 1.7 to 2.1.²⁸ Usual ductal hyperplasia and columnar hyperplasia have little risk of upstaging with excision, and therefore, surgical consultation is not recommended.

Previously, surgical excision was recommended for any intraductal papilloma due to risk of upgrade in pathologic diagnosis at the time of excision. However, more recent data suggest that the upgrade rate is about 2.2% for a solitary papilloma that is less than 1 cm in diameter and without associated mass lesion (either clinically or radiographically), is concordant with radiographic findings, and has no associated atypical cells on biopsy.³¹ In this case, observation and short-interval clinical follow-up are reasonable. If there are multiple papillomas, the patient has symptoms such as persistent bloody nipple discharge, or any of the above criteria are not met, surgical excision is recommended.²⁸

Similarly, radial scars and complex sclerosing lesions are increasingly likely to be associated with malignancy based on size. Upstaging ranges from 0% to 12%. It is again important when evaluating radial scars that there is pathologic concordance and that there were no associated high-risk lesions on pathology. If this is the case, it is reasonable to clinically monitor patients with small radial scars, particularly in those who do not have an elevated risk of developing breast cancer.³⁰

For all patients who have undergone biopsy and whose pathology study results are benign, a thorough risk evaluation should be performed, including calculation of their lifetime risk of breast cancer. This can be done with the National Cancer Institute Breast Cancer Risk Assessment Tool, the International Breast Cancer Intervention Study (IBIS) risk calculator, or other model using family history as a basis for calculations. Patients found to have a lifetime risk of breast cancer of greater than 20% to 25% should be offered annual screening with magnetic resonance imaging in addition to mammography.

ATYPICAL HYPERPLASIA: INCREASED RISK

When biopsy study shows atypical ductal hyperplasia or atypical lobular hyperplasia, there is an increased risk of breast cancer.^28,32 The absolute overall risk of developing breast cancer in 25 years is 30%, and that risk is further stratified based on the number of foci of atypia noted in the specimen.²⁹

When core-needle biopsy study reveals atypical ductal hyperplasia in the tissue, there is a 15% to 30% risk of finding breast cancer with surgical excision.²⁸ Surgical excision is therefore recommended for atypical ductal hyperplasia noted on core-needle biopsy.²⁸

In contrast, when atypical lobular hyperplasia alone is noted, the risk of upstagingto malignancy varies widely—from 0% to 67%—although recent studies have noted risks of 1% to 3%.^33,34 Thus, the decision for surgical excision is more variable. Generally, if the atypical lobular hyperplasia is noted incidentally, is not associated with a higher grade lesion, and is concordant with imaging, it is reasonable to closely monitor with serial imaging and physical examination. Excision is unnecessary.³⁵

Patients found to have atypical hyperplasia on breast biopsy should receive counseling about risk-reducing medications. Selective estrogen receptor modulators such as tamoxifen and raloxifene have been shown to reduce the risk of breast cancer by as much as 86% in patients with atypical hyperplasia.³⁶ Similarly, aromatase inhibitors such as exemestane and anastrozole reduce breast cancer risk by approximately 65%.³⁷

Breast concerns account for approximately 3% of all female visits to a primary care practice.¹ The most common symptoms are breast lumps and breast pain.

Because breast cancer is the most common malignancy in women in the United States, affecting nearly 1 in 8 women in their lifetime, women with breast problems often fear the worst. However, only about 3.5% of women reporting a concern have cancer; most problems are benign (Table 1).¹

Here, we present an evidence-based review of common breast problems in primary care practice and discuss how to evaluate and manage them.

GENERAL APPROACH

The evaluation of a breast concern requires a systematic approach, beginning with a history that documents the onset, severity, and frequency of symptoms. If the concern is a lump or mass, ask whether it becomes more tender or increases in size at any point during the menstrual cycle.

Focus the physical examination on the cervical, supraclavicular, infraclavicular, and axillary lymph nodes and on the breast itself. Assess breast symmetry, note any skin changes such as dimpling, and check the nipples for discharge and inversion. Palpate the breasts for masses.

PALPABLE BREAST MASS: IMAGING NEEDED

If a mass is present, it is more likely to be malignant if any of the following is true:

• Firm to hard texture or indistinct margins
• Attached to the underlying deep fascia or skin
• Associated nipple inversion or skin dimpling.²

Breast masses are more likely benign if they have discrete, well-defined margins, are mobile with a soft to rubbery consistency, and change with the menstrual cycle. However, clinical features are unreliable indicators of cause, and thus additional investigation with breast imaging is warranted.

Mammography remains the diagnostic test of choice for all women age 30 or older who have a palpable breast mass. It is less effective in younger women because they are more likely to have extremely dense fibroglandular tissue that will limit its sensitivity to imaging.

Order diagnostic mammography, which includes additional views focused on the area of concern, rather than screening mammography, which includes only standard cranio­caudal and mediolateral oblique views. A skin marker should be applied over the palpable lump to aid imaging. Because a breast that contains a mass may be denser than the opposite breast or may show asymmetry, both breasts should be imaged. The sensitivity of diagnostic mammography varies from 85% to 90%, so a negative mammogram does not rule out malignancy.^2,3

Targeted ultrasonography of the palpable mass helps identify solid masses such as fibroadenomas or malignant tumors, classifies the margins (lobulated, smooth, or irregular), and assesses vascularity. Ultrasonography is particularly useful for characterizing cystic lesions (eg, simple, septated, or clustered cysts) and cysts with internal echoes. It can also identify lipomas or sebaceous cysts.

If the findings on both mammography and ultrasonography are benign, the likelihood of cancer is very low, with an estimated negative predictive value of 97% to 100%.^2,3 Additionally, the likelihood of nonmalignant findings on biopsy after benign imaging is approximately 99%.³

Although radiologic imaging can define palpable masses, it is intended as a clinical aid. Suspicious findings on clinical examination should never be ignored even if findings on imaging are reassuring, as studies have documented that about 5% of breast cancers may be detected on clinical breast examination alone.⁴

Other imaging tests such as magnetic resonance imaging may be considered occasionally if clinical suspicion remains high after negative mammography and ultrasonography, but they cannot confirm a diagnosis of malignancy. In that case, refer the patient to a surgeon for consideration of excisional biopsy.

Patients with an indeterminate lesion can return in 3 to 12 weeks for a follow-up examination and repeat imaging, which helps assess interval clinical stability. The latter option is especially helpful for patients with masses that are of low suspicion or for patients who prefer to avoid invasive tissue biopsy.

Patients with clinical and radiologic findings that suggest a benign cause can return for short-term follow-up in 6 months or in 12 months for their regular mammogram.

More than 50% of women experience breast pain at some point in their life.⁵ Of these, 35% report that the pain adversely affects their sleep, and 41% note that the pain detrimentally affects their sexual quality of life. Up to 66% of breast pain correlates directly with the patient’s menstrual cycle.⁵ Breast pain is rarely associated with malignancy.

Regardless of its severity and the low likelihood of malignancy, breast pain can be a significant source of distress for the patient, primarily because of concerns about underlying malignancy. If the patient has a focal area of pain on examination, order mammography in combination with targeted ultrasonography. The sensitivity and negative predictive value of benign findings on combination mammography and ultrasonography in this setting are as high as 100%. The incidence of underlying cancer in patients with focal breast pain and no palpable mass is approximately 1.2%.⁶

The long-term prognosis in women with diffuse, often bilateral breast pain (in the absence of additional clinical findings) is excellent. In one study, the incidence of a breast cancer diagnosis was 1.8% after a median of 51 months of follow-up.⁷ Therefore, patients presenting with diffuse pain, no palpable abnormalities, and benign imaging can be safely reassured. Magnetic resonance imaging is rarely indicated in patients with breast pain unless other clinical findings, such as a mass or skin changes, are noted and the results of mammography and ultrasonography are negative.

Treating breast pain

Treating breast pain remains a challenge. The first step is to reassure the patient about her prognosis and help her make appropriate lifestyle modifications.

A well-fitting bra. Suggest getting a professional bra fitting. Wearing a well-fitted bra that offers lift, support, and compression and reduces excess motion can help improve benign breast pain. A bra fitting is especially important for women with large breasts because it can be difficult for these women to get an accurate size. Wearing a lightly fitted bra at night may also provide comfort if there is nighttime pain with breast tissue movement.

Reducing daily caffeine intake is often advised for pain management, but strong evidence of its efficacy is lacking.

Anti-inflammatory drugs can be beneficial if used short-term, especially if costochondritis is suspected.

Danazol improves pain in more than 70% of patients with cyclical symptoms and in up to 48% of those with noncyclical symptoms.

Bromocriptine is effective in up to 54% of those with cyclical symptoms and in up to 33% of those with noncyclical symptoms.⁸ However, the US Food and Drug Administration (FDA) withdrew approval for this indication because of adverse effects.

Tamoxifen, in contrast, provides relief in 94% of those with cyclical symptoms and in 56% of those with noncyclical symptoms.⁹

Adverse effects, however, limit the use of danazol, bromocriptine, and tamoxifen, and they should be prescribed only for short-term use (3 to 6 months) and only in women with chronic debilitating pain.

A few small studies have evaluated alternative options.

Toremifene is a triphenylethylene derivative similar to tamoxifen that is also used in the adjuvant treatment of postmenopausal breast cancer (but with fewer adverse effects). It has been documented to have a significant effect on premenstrual breast pain, with a 64% reduction in breast pain scores compared with a 26% reduction with placebo.¹⁰ However, the FDA has not approved it for this indication, and it can be cost-prohibitive.

Over-the-counter medications that may provide relief for cyclic breast pain include vitamin E or B₆, products containing oil of Vitex agnus castus (chaste tree or chasteberry), and flaxseed.^11,12

Acupuncture has been evaluated in patients with noncyclic breast pain and was found to reduce pain by 56% to 67% in one study,¹³ although it did not affect quality of life.

NIPPLE DISCHARGE

From 5% to 7% of women seek medical attention for nipple discharge.^14,15 Breast cancer is found in 5% to 15% of women who undergo surgery for nipple discharge.^16,17

Review the patient’s current medications and inquire about health conditions such as thyroid dysfunction or visual field changes that suggest a pituitary mass (which can lead to nipple discharge by causing hormonal dysregulation or hyperprolactinemia).

Palpate the breasts for an underlying mass, look for lesions on the nipple, and assess the color of the fluid. Also note whether there is discharge from one or both breasts, whether it is spontaneous or expressive, and whether it occurs from a single or multiple ducts. Nipple lesions may require further testing with punch biopsy.

Nonlactational nipple discharge is classified as physiologic or pathologic. Physiologic nipple discharge is typically bilateral, involving multiple ducts, and is often clear or straw-colored but may also be green, gray, or brown.

White, opaque fluid is often related to galactorrhea as a result of hyperprolactinemia, hypothyroidism, or medications such as antipsychotic drugs (eg, haloperidol and fluphenazine) and gastrointestinal motility agents such as metoclopramide. Discharge also commonly results from benign underlying ductal abnormalities such as intraductal papilloma, periductal mastitis, and duct ectasia.

Pathologic nipple discharge is often unilateral and persistent, occurring spontaneously from a solitary duct, and may be bloody or serous.

For women with pathologic nipple discharge who are 30 or older, diagnostic imaging with mammography and subareolar ultrasonography is recommended. If the patient is younger than 30, ultrasonography of the subareolar region alone can be used. Targeted ultrasonography of any palpable area is also advised.

Cytologic assessment of the fluid is not recommended because it can often lead to a false-positive finding of atypical cells. Imaging studies such as ductography, duct lavage, ductoscopy, and magnetic resonance imaging are also generally unnecessary; instead, a persistent clinical concern should prompt a surgical referral for consideration of duct excision.

When a patient has pathologic nipple discharge with a negative physical examination and breast imaging, studies have shown that the risk of cancer is 3% or less.¹⁸

Patients with spontaneous bloody or serous single-duct discharge with negative results on mammography and ultrasonography should be reassured that they have a low risk of underlying cancer. If the patient prefers, one approachto management is follow-up mammography and ultrasonography at 6 months and clinical examination for up to 2 years or until the discharge resolves on its own.

On the other hand, if the discharge is distressing to the patient, subareolar duct excision can be performed with both a diagnostic and therapeutic purpose.

A rash on the nipple or areolar region warrants careful evaluation because it may be the first sign of Paget disease of the breast.

In the clinical breast examination, assess the extent of the rash and the presence of any underlying breast mass or nipple discharge. Dermatitis often starts on the areola and resolves quickly with topical therapy. However, Paget disease tends to start directly on the nipple itself, is unresponsive or only partially responsive to topical therapy, and progresses gradually, leading to erosions and ultimately effacement of the nipple itself.

If the clinical examination suggests mild dermatitis and the results of breast imaging are negative, treat the patient with a topical medication because benign conditions such as dermatitis and eczema are common. However, continued follow-up is mandatory until the rash completely resolves: Paget disease sometimes initially improves with topical therapy due to its inflammatory nature.

If you suspect Paget disease or the rash does not fully resolve after 2 to 3 weeks of topical therapy, refer the patient to a dermatologist for full-thickness punch biopsy to establish the diagnosis.

Paget disease of the breast may or may not be associated with underlying ductal carcinoma in situ or invasive breast cancer.¹⁹ The absence of clinical or imaging abnormalities in a patient with Paget disease does not rule out underlying malignancy.²⁰

DENSE BREASTS

Increased breast density has been shown to be a risk factor for breast cancer and may be prognostically useful when combined with the Tyrer-Cuzick model or the Gail model of breast cancer risk.²⁴

Additionally, increased density can mask cancers on mammography, significantly reducing its sensitivity. In women with heterogeneously or extremely dense breasts, the sensitivity of mammography for detecting cancer is only 25% to 50%.²¹ Due to this low sensitivity, supplemental imaging is helpful, particularly in women already at risk of breast cancer based on family history.

Supplemental screening

Digital mammography with tomosynthesis was approved by the FDA in 2011 for use in combination with standard digital mammography for breast cancer screening. Compared with traditional 2-dimensional mammography alone, adding 3-D tomosynthesis decreases the recall rate and increases the cancer detection rate.²⁵

Tomosynthesis tends to perform better in women with heterogeneously dense breasts (BI-RADS category C). There is no significant improvement in cancer detection in women with extremely dense breasts (BI-RADS category D).²⁶

Depending on the methodology, radiation exposure can be either higher or lower than with traditional mammography. However, in all forms, the very small amount of radiation is considered safe.

Whole breast ultrasonography. When whole breast ultrasonography is used to supplement mammography, the recall rate is higher than when mammography is used alone (14% vs 7%–11%).²² It also increases the cancer detection rate by 4.4 additional cancers per 1,000 examinations. However, the false-positive rate with whole breast ultrasonography is higher; the positive predictive value of combined mammography and ultrasonography is 11.2% vs 22.6% for mammography alone.²² Therefore, we do not generally recommend whole breast ultrasonography as a supplement to mammography in women with dense breast tissue unless other studies are not an option.

Molecular breast imaging is not widely available because it requires special equipment, injection of a radiopharamceutical (technetium Tc 99m sestamibi), and a radiologist who specializes in breast imaging to interpret the results. When it is available, however, it increases the cancer detection rate by 8.8 in 1,000 examinations; the positive predictive value is similar to that of screening mammography alone.²¹ It is particularly useful in patients with dense breasts who do not qualify for screening magnetic resonance imaging (lifetime risk of < 20% to 25%).

Technetium sestamibi is associated with a minimal amount of radiation exposure (2.4 mSv vs 1.2 mSV with standard mammography). However, this exposure is much less than background radiation exposure and is considered safe.²¹

Screening mammography can disclose abnormalities such as calcifications, masses, asymmetry, or architectural distortion.²⁷ Abnormalities are reported using standardized BI-RADS categories designated with the numbers 0 through 6 (Table 3).²³

A report of BI-RADS category 0 (incomplete), 4 (suspicious), or 5 (highly suspicious) requires additional workup.

Category 1 (negative) requires no further follow-up, and the patient should resume age-appropriate screening.

For patients with category 2 (benign) findings, routine screening is recommended, whereas patients with category 3 (probably benign) are advised to come back in 6 months for follow-up imaging.

Diagnostic mammography includes additional assessments for focal symptoms or areas of abnormality noted on screening imaging or clinical examination. These may include spot magnification views of areas of asymmetry, mass, architectural distortion, or calcifications. Ultrasonography of focal breast abnormalities can help determine if there is an underlying cyst or solid mass.

MANAGEMENT OF BENIGN FINDINGS ON BREAST BIOPSY

Benign breast disease is diagnosed when a patient with a palpable or radiographic abnormality undergoes breast biopsy with benign findings.^28,29 It can be largely grouped into 3 categories: nonproliferative, proliferative without atypia, and proliferative with atypia (Table 4).^28,29

If core-needle biopsy study results are benign, the next step is to establish radiologic-pathologic and clinical-pathologic concordance. If the findings on clinical examination or imaging are not consistent with those on pathologic study, excisional biopsy should be performed, as imaging-directed biopsy may not have adequately sampled the lesion.³⁰

Nonproliferative lesions account for about 65% of findings on core-needle biopsy and include simple cysts, fibroadenomas, columnar cell changes, apocrine metaplasia, and mild ductal hyperplasia of the usual type. These lesions do not significantly increase the risk of breast cancer; the relative risk is 1.2 to 1.4.^28,29 Additionally, the risk of “upstaging” after excisional biopsy—ie, to a higher-risk lesion or to malignancy—is minimal. Therefore, no additional action is necessary when these findings alone are noted on core-needle biopsy.

Proliferative lesions without atypia account for about 30% of biopsy results and include usual ductal hyperplasia, sclerosing adenosis, columnar hyperplasia, papilloma, and radial scar. Generally, there is a slightly increased risk of subsequent breast cancer, with a relative risk of 1.7 to 2.1.²⁸ Usual ductal hyperplasia and columnar hyperplasia have little risk of upstaging with excision, and therefore, surgical consultation is not recommended.

Previously, surgical excision was recommended for any intraductal papilloma due to risk of upgrade in pathologic diagnosis at the time of excision. However, more recent data suggest that the upgrade rate is about 2.2% for a solitary papilloma that is less than 1 cm in diameter and without associated mass lesion (either clinically or radiographically), is concordant with radiographic findings, and has no associated atypical cells on biopsy.³¹ In this case, observation and short-interval clinical follow-up are reasonable. If there are multiple papillomas, the patient has symptoms such as persistent bloody nipple discharge, or any of the above criteria are not met, surgical excision is recommended.²⁸

Similarly, radial scars and complex sclerosing lesions are increasingly likely to be associated with malignancy based on size. Upstaging ranges from 0% to 12%. It is again important when evaluating radial scars that there is pathologic concordance and that there were no associated high-risk lesions on pathology. If this is the case, it is reasonable to clinically monitor patients with small radial scars, particularly in those who do not have an elevated risk of developing breast cancer.³⁰

For all patients who have undergone biopsy and whose pathology study results are benign, a thorough risk evaluation should be performed, including calculation of their lifetime risk of breast cancer. This can be done with the National Cancer Institute Breast Cancer Risk Assessment Tool, the International Breast Cancer Intervention Study (IBIS) risk calculator, or other model using family history as a basis for calculations. Patients found to have a lifetime risk of breast cancer of greater than 20% to 25% should be offered annual screening with magnetic resonance imaging in addition to mammography.

ATYPICAL HYPERPLASIA: INCREASED RISK

When biopsy study shows atypical ductal hyperplasia or atypical lobular hyperplasia, there is an increased risk of breast cancer.^28,32 The absolute overall risk of developing breast cancer in 25 years is 30%, and that risk is further stratified based on the number of foci of atypia noted in the specimen.²⁹

When core-needle biopsy study reveals atypical ductal hyperplasia in the tissue, there is a 15% to 30% risk of finding breast cancer with surgical excision.²⁸ Surgical excision is therefore recommended for atypical ductal hyperplasia noted on core-needle biopsy.²⁸