In patients with cardiac stents, do not stop aspirin. If the risk of bleeding outweighs the benefit (eg, with intracranial procedures), an informed discussion involving the surgeon, cardiologist, and patient is critical to ascertain risks vs benefits.

See related editorial

In patients using aspirin for secondary prevention, the decision depends on the patient’s cardiac status and an assessment of risk vs benefit. Aspirin has no role in patients undergoing noncardiac surgery who are at low risk of a major adverse cardiac event.^1,2

Aspirin used for secondary prevention reduces rates of death from vascular causes,³ but data on the magnitude of benefit in the perioperative setting are still evolving. In patients with coronary stents, continuing aspirin is beneficial,^4,5 whereas stopping it is associated with an increased risk of acute stent thrombosis, which causes significant morbidity and mortality.⁶

SURGERY AND THROMBOTIC RISK: WHY CONSIDER ASPIRIN?

The Vascular Events in Noncardiac Surgery Patients Cohort Evaluation (VISION) study⁷ prospectively screened 15,133 patients for myocardial injury with troponin T levels daily for the first 3 consecutive postoperative days; 1,263 (8%) of the patients had a troponin elevation of 0.03 ng/mL or higher. The 30-day mortality rate in this group was 9.8%, compared with 1.1% in patients with a troponin T level of less than 0.03 ng/mL (odds ratio 10.07; 95% confidence interval [CI] 7.84–12.94; P < .001).⁸ The higher the peak troponin T concentration, the higher the risk of death within 30 days:

• 0.01 ng/mL or less, risk 1.0%
• 0.02 ng/mL, risk 4.0%
• 0.03 to 0.29 ng/mL, risk 9.3%
• 0.30 ng/mL or greater, risk 16.9%.⁷

Myocardial injury is a common postoperative vascular complication.⁷ Myocardial infarction (MI) or injury perioperatively increases the risk of death: 1 in 10 patients dies within 30 days after surgery.⁸

Surgery creates substantial physiologic stress through factors such as fasting, anesthesia, intubation, surgical trauma, extubation, and pain. It promotes coagulation⁹ and inflammation with activation of platelets,¹⁰ potentially leading to thrombosis.¹¹ Coronary thrombosis secondary to plaque rupture^11,12 can result in perioperative MI. Perioperative hemodynamic variability, anemia, and hypoxia can lead to demand-supply mismatch and also cause cardiac ischemia.

Aspirin is an antiplatelet agent that irreversibly inhibits platelet aggregation by blocking the formation of cyclooxygenase. It has been used for several decades as an antithrombotic agent in primary and secondary prevention. However, its benefit in primary prevention is uncertain, and the magnitude of antithrombotic benefit must be balanced against the risk of bleeding.

The Antithrombotic Trialists’ Collaboration¹³ performed a systematic review of 6 primary prevention trials involving 95,000 patients and found that aspirin therapy was associated with a 12% reduction in serious vascular events, which occurred in 0.51% of patients taking aspirin per year vs 0.57% of controls (P = .0001). However, aspirin also increased the risk of major bleeding, at a rate of 0.10% vs 0.07% per year (P < .0001), with 2 bleeding events for every avoided vascular event.¹³

WILL ASPIRIN PROTECT PATIENTS AT CARDIAC RISK?

The second Perioperative Ischemic Evaluation trial (POISE 2),¹ in patients with atherosclerotic disease or at risk for it, found that giving aspirin in the perioperative period did not reduce the rate of death or nonfatal MI, but increased the risk of a major bleeding event.

The trial included 10,010 patients undergoing noncardiac surgery who were randomly assigned to receive aspirin or placebo. The aspirin arm included 2 groups: patients who were not on aspirin (initiation arm), and patients on aspirin at the time of randomization (continuation arm).

Death or nonfatal MI (the primary outcome) occurred in 7.0% of patients on aspirin vs 7.1% of patients receiving placebo (hazard ratio [HR] 0.99, 95% CI 0.86–1.15, P = .92). The risk of major bleeding was 4.6% in the aspirin group vs 3.8% in the placebo group (HR 1.23, 95% CI 1.01–1.49, ^P = .04).¹

George et al,¹⁴ in a prospective observational study in a single tertiary care center, found that fewer patients with myocardial injury in noncardiac surgery died if they took aspirin or clopidogrel postoperatively. Conversely, lack of antithrombotic therapy was an independent predictor of death (P < .001). The mortality rate in patients with myocardial injury who were on antithrombotic therapy postoperatively was 6.7%, compared with 12.1% in those without postoperative antithrombotic therapy (estimated number needed to treat, 19).¹⁴

Percutaneous coronary intervention (PCI) accounts for 3.6% of all operating-room procedures in the United States,¹⁵ and 20% to 35% of patients who undergo PCI undergo noncardiac surgery within 2 years of stent implantation.^16,17

Antiplatelet therapy is discontinued in about 20% of patients with previous PCI who undergo noncardiac surgery.¹⁸

Observational data have shown that stopping antiplatelet therapy in patients with previous PCI with stent placement who undergo noncardiac surgery is the single most important predictor of stent thrombosis and death.^19–21 The risk increases if the interval between stent implantation and surgery is shorter, especially within 180 days.^16,17 Patients who have stent thrombosis are at significantly higher risk of death.

Graham et al⁴ conducted a subgroup analysis of the POISE 2 trial comparing aspirin and placebo in 470 patients who had undergone PCI (427 had stent placement, and the rest had angioplasty or an unspecified type of PCI); 234 patients received aspirin and 236 placebo. The median time from stent implantation to surgery was 5.3 years.

Of the patients in the aspirin arm, 14 (6%) had the primary outcome of death or nonfatal MI compared with 27 patients (11.5%) in the placebo arm (absolute risk reduction 5.5%, 95% CI 0.4%–10.5%). The result, which differed from that in the primary trial,¹ was due to reduction in MI in the PCI subgroup on aspirin. PCI patients who were on aspirin did not have increased bleeding risk. This subgroup analysis, albeit small and limited, suggests that continuing low-dose aspirin in patients with previous PCI, irrespective of the type of stent or the time from stent implantations, minimizes the risk of perioperative MI.

GUIDELINES AND RECOMMENDATIONS

Routine perioperative use of aspirin increases the risk of bleeding without a reduction in ischemic events.¹ Patients with prior PCI are at increased risk of acute stent thrombosis when antiplatelet medications are discontinued.^20,21 Available data, although limited, support continuing low-dose aspirin without interruption in the perioperative period in PCI patients,⁴ as do the guidelines from the American College of Cardiology.⁵

We propose a management algorithm for patients undergoing noncardiac surgery on antiplatelet therapy that takes into consideration whether the surgery is urgent, elective, or time-sensitive (Figure 1). It is imperative to involve the cardiologist, surgeon, anesthesiologist, and the patient in the decision-making process.

In the perioperative setting for patients undergoing noncardiac surgery:

• Discontinue aspirin in patients without coronary heart disease, as bleeding risk outweighs benefit.
• Consider aspirin in patients at high risk for a major adverse cardiac event if benefits outweigh risk.
• Continue low-dose aspirin without interruption in patients with a coronary stent, irrespective of the type of stent.
• If a patient has had PCI with stent placement but is not currently on aspirin, talk with the patient and the treating cardiologist to find out why, and initiate aspirin if no contraindications exist.

In patients with cardiac stents, do not stop aspirin. If the risk of bleeding outweighs the benefit (eg, with intracranial procedures), an informed discussion involving the surgeon, cardiologist, and patient is critical to ascertain risks vs benefits.

See related editorial

In patients using aspirin for secondary prevention, the decision depends on the patient’s cardiac status and an assessment of risk vs benefit. Aspirin has no role in patients undergoing noncardiac surgery who are at low risk of a major adverse cardiac event.^1,2

Aspirin used for secondary prevention reduces rates of death from vascular causes,³ but data on the magnitude of benefit in the perioperative setting are still evolving. In patients with coronary stents, continuing aspirin is beneficial,^4,5 whereas stopping it is associated with an increased risk of acute stent thrombosis, which causes significant morbidity and mortality.⁶

SURGERY AND THROMBOTIC RISK: WHY CONSIDER ASPIRIN?

The Vascular Events in Noncardiac Surgery Patients Cohort Evaluation (VISION) study⁷ prospectively screened 15,133 patients for myocardial injury with troponin T levels daily for the first 3 consecutive postoperative days; 1,263 (8%) of the patients had a troponin elevation of 0.03 ng/mL or higher. The 30-day mortality rate in this group was 9.8%, compared with 1.1% in patients with a troponin T level of less than 0.03 ng/mL (odds ratio 10.07; 95% confidence interval [CI] 7.84–12.94; P < .001).⁸ The higher the peak troponin T concentration, the higher the risk of death within 30 days:

• 0.01 ng/mL or less, risk 1.0%
• 0.02 ng/mL, risk 4.0%
• 0.03 to 0.29 ng/mL, risk 9.3%
• 0.30 ng/mL or greater, risk 16.9%.⁷

Myocardial injury is a common postoperative vascular complication.⁷ Myocardial infarction (MI) or injury perioperatively increases the risk of death: 1 in 10 patients dies within 30 days after surgery.⁸

Surgery creates substantial physiologic stress through factors such as fasting, anesthesia, intubation, surgical trauma, extubation, and pain. It promotes coagulation⁹ and inflammation with activation of platelets,¹⁰ potentially leading to thrombosis.¹¹ Coronary thrombosis secondary to plaque rupture^11,12 can result in perioperative MI. Perioperative hemodynamic variability, anemia, and hypoxia can lead to demand-supply mismatch and also cause cardiac ischemia.

Aspirin is an antiplatelet agent that irreversibly inhibits platelet aggregation by blocking the formation of cyclooxygenase. It has been used for several decades as an antithrombotic agent in primary and secondary prevention. However, its benefit in primary prevention is uncertain, and the magnitude of antithrombotic benefit must be balanced against the risk of bleeding.

The Antithrombotic Trialists’ Collaboration¹³ performed a systematic review of 6 primary prevention trials involving 95,000 patients and found that aspirin therapy was associated with a 12% reduction in serious vascular events, which occurred in 0.51% of patients taking aspirin per year vs 0.57% of controls (P = .0001). However, aspirin also increased the risk of major bleeding, at a rate of 0.10% vs 0.07% per year (P < .0001), with 2 bleeding events for every avoided vascular event.¹³

WILL ASPIRIN PROTECT PATIENTS AT CARDIAC RISK?

The second Perioperative Ischemic Evaluation trial (POISE 2),¹ in patients with atherosclerotic disease or at risk for it, found that giving aspirin in the perioperative period did not reduce the rate of death or nonfatal MI, but increased the risk of a major bleeding event.

The trial included 10,010 patients undergoing noncardiac surgery who were randomly assigned to receive aspirin or placebo. The aspirin arm included 2 groups: patients who were not on aspirin (initiation arm), and patients on aspirin at the time of randomization (continuation arm).

Death or nonfatal MI (the primary outcome) occurred in 7.0% of patients on aspirin vs 7.1% of patients receiving placebo (hazard ratio [HR] 0.99, 95% CI 0.86–1.15, P = .92). The risk of major bleeding was 4.6% in the aspirin group vs 3.8% in the placebo group (HR 1.23, 95% CI 1.01–1.49, ^P = .04).¹

George et al,¹⁴ in a prospective observational study in a single tertiary care center, found that fewer patients with myocardial injury in noncardiac surgery died if they took aspirin or clopidogrel postoperatively. Conversely, lack of antithrombotic therapy was an independent predictor of death (P < .001). The mortality rate in patients with myocardial injury who were on antithrombotic therapy postoperatively was 6.7%, compared with 12.1% in those without postoperative antithrombotic therapy (estimated number needed to treat, 19).¹⁴

Percutaneous coronary intervention (PCI) accounts for 3.6% of all operating-room procedures in the United States,¹⁵ and 20% to 35% of patients who undergo PCI undergo noncardiac surgery within 2 years of stent implantation.^16,17

Antiplatelet therapy is discontinued in about 20% of patients with previous PCI who undergo noncardiac surgery.¹⁸

Observational data have shown that stopping antiplatelet therapy in patients with previous PCI with stent placement who undergo noncardiac surgery is the single most important predictor of stent thrombosis and death.^19–21 The risk increases if the interval between stent implantation and surgery is shorter, especially within 180 days.^16,17 Patients who have stent thrombosis are at significantly higher risk of death.

Graham et al⁴ conducted a subgroup analysis of the POISE 2 trial comparing aspirin and placebo in 470 patients who had undergone PCI (427 had stent placement, and the rest had angioplasty or an unspecified type of PCI); 234 patients received aspirin and 236 placebo. The median time from stent implantation to surgery was 5.3 years.

Of the patients in the aspirin arm, 14 (6%) had the primary outcome of death or nonfatal MI compared with 27 patients (11.5%) in the placebo arm (absolute risk reduction 5.5%, 95% CI 0.4%–10.5%). The result, which differed from that in the primary trial,¹ was due to reduction in MI in the PCI subgroup on aspirin. PCI patients who were on aspirin did not have increased bleeding risk. This subgroup analysis, albeit small and limited, suggests that continuing low-dose aspirin in patients with previous PCI, irrespective of the type of stent or the time from stent implantations, minimizes the risk of perioperative MI.

GUIDELINES AND RECOMMENDATIONS

Routine perioperative use of aspirin increases the risk of bleeding without a reduction in ischemic events.¹ Patients with prior PCI are at increased risk of acute stent thrombosis when antiplatelet medications are discontinued.^20,21 Available data, although limited, support continuing low-dose aspirin without interruption in the perioperative period in PCI patients,⁴ as do the guidelines from the American College of Cardiology.⁵

We propose a management algorithm for patients undergoing noncardiac surgery on antiplatelet therapy that takes into consideration whether the surgery is urgent, elective, or time-sensitive (Figure 1). It is imperative to involve the cardiologist, surgeon, anesthesiologist, and the patient in the decision-making process.

In the perioperative setting for patients undergoing noncardiac surgery:

• Discontinue aspirin in patients without coronary heart disease, as bleeding risk outweighs benefit.
• Consider aspirin in patients at high risk for a major adverse cardiac event if benefits outweigh risk.
• Continue low-dose aspirin without interruption in patients with a coronary stent, irrespective of the type of stent.
• If a patient has had PCI with stent placement but is not currently on aspirin, talk with the patient and the treating cardiologist to find out why, and initiate aspirin if no contraindications exist.

In patients with cardiac stents, do not stop aspirin. If the risk of bleeding outweighs the benefit (eg, with intracranial procedures), an informed discussion involving the surgeon, cardiologist, and patient is critical to ascertain risks vs benefits.

See related editorial

In patients using aspirin for secondary prevention, the decision depends on the patient’s cardiac status and an assessment of risk vs benefit. Aspirin has no role in patients undergoing noncardiac surgery who are at low risk of a major adverse cardiac event.^1,2

Aspirin used for secondary prevention reduces rates of death from vascular causes,³ but data on the magnitude of benefit in the perioperative setting are still evolving. In patients with coronary stents, continuing aspirin is beneficial,^4,5 whereas stopping it is associated with an increased risk of acute stent thrombosis, which causes significant morbidity and mortality.⁶

SURGERY AND THROMBOTIC RISK: WHY CONSIDER ASPIRIN?

The Vascular Events in Noncardiac Surgery Patients Cohort Evaluation (VISION) study⁷ prospectively screened 15,133 patients for myocardial injury with troponin T levels daily for the first 3 consecutive postoperative days; 1,263 (8%) of the patients had a troponin elevation of 0.03 ng/mL or higher. The 30-day mortality rate in this group was 9.8%, compared with 1.1% in patients with a troponin T level of less than 0.03 ng/mL (odds ratio 10.07; 95% confidence interval [CI] 7.84–12.94; P < .001).⁸ The higher the peak troponin T concentration, the higher the risk of death within 30 days:

• 0.01 ng/mL or less, risk 1.0%
• 0.02 ng/mL, risk 4.0%
• 0.03 to 0.29 ng/mL, risk 9.3%
• 0.30 ng/mL or greater, risk 16.9%.⁷

Myocardial injury is a common postoperative vascular complication.⁷ Myocardial infarction (MI) or injury perioperatively increases the risk of death: 1 in 10 patients dies within 30 days after surgery.⁸

Surgery creates substantial physiologic stress through factors such as fasting, anesthesia, intubation, surgical trauma, extubation, and pain. It promotes coagulation⁹ and inflammation with activation of platelets,¹⁰ potentially leading to thrombosis.¹¹ Coronary thrombosis secondary to plaque rupture^11,12 can result in perioperative MI. Perioperative hemodynamic variability, anemia, and hypoxia can lead to demand-supply mismatch and also cause cardiac ischemia.

Aspirin is an antiplatelet agent that irreversibly inhibits platelet aggregation by blocking the formation of cyclooxygenase. It has been used for several decades as an antithrombotic agent in primary and secondary prevention. However, its benefit in primary prevention is uncertain, and the magnitude of antithrombotic benefit must be balanced against the risk of bleeding.

The Antithrombotic Trialists’ Collaboration¹³ performed a systematic review of 6 primary prevention trials involving 95,000 patients and found that aspirin therapy was associated with a 12% reduction in serious vascular events, which occurred in 0.51% of patients taking aspirin per year vs 0.57% of controls (P = .0001). However, aspirin also increased the risk of major bleeding, at a rate of 0.10% vs 0.07% per year (P < .0001), with 2 bleeding events for every avoided vascular event.¹³

WILL ASPIRIN PROTECT PATIENTS AT CARDIAC RISK?

The second Perioperative Ischemic Evaluation trial (POISE 2),¹ in patients with atherosclerotic disease or at risk for it, found that giving aspirin in the perioperative period did not reduce the rate of death or nonfatal MI, but increased the risk of a major bleeding event.

The trial included 10,010 patients undergoing noncardiac surgery who were randomly assigned to receive aspirin or placebo. The aspirin arm included 2 groups: patients who were not on aspirin (initiation arm), and patients on aspirin at the time of randomization (continuation arm).

Death or nonfatal MI (the primary outcome) occurred in 7.0% of patients on aspirin vs 7.1% of patients receiving placebo (hazard ratio [HR] 0.99, 95% CI 0.86–1.15, P = .92). The risk of major bleeding was 4.6% in the aspirin group vs 3.8% in the placebo group (HR 1.23, 95% CI 1.01–1.49, ^P = .04).¹

George et al,¹⁴ in a prospective observational study in a single tertiary care center, found that fewer patients with myocardial injury in noncardiac surgery died if they took aspirin or clopidogrel postoperatively. Conversely, lack of antithrombotic therapy was an independent predictor of death (P < .001). The mortality rate in patients with myocardial injury who were on antithrombotic therapy postoperatively was 6.7%, compared with 12.1% in those without postoperative antithrombotic therapy (estimated number needed to treat, 19).¹⁴

Percutaneous coronary intervention (PCI) accounts for 3.6% of all operating-room procedures in the United States,¹⁵ and 20% to 35% of patients who undergo PCI undergo noncardiac surgery within 2 years of stent implantation.^16,17

Antiplatelet therapy is discontinued in about 20% of patients with previous PCI who undergo noncardiac surgery.¹⁸

Observational data have shown that stopping antiplatelet therapy in patients with previous PCI with stent placement who undergo noncardiac surgery is the single most important predictor of stent thrombosis and death.^19–21 The risk increases if the interval between stent implantation and surgery is shorter, especially within 180 days.^16,17 Patients who have stent thrombosis are at significantly higher risk of death.

Graham et al⁴ conducted a subgroup analysis of the POISE 2 trial comparing aspirin and placebo in 470 patients who had undergone PCI (427 had stent placement, and the rest had angioplasty or an unspecified type of PCI); 234 patients received aspirin and 236 placebo. The median time from stent implantation to surgery was 5.3 years.

Of the patients in the aspirin arm, 14 (6%) had the primary outcome of death or nonfatal MI compared with 27 patients (11.5%) in the placebo arm (absolute risk reduction 5.5%, 95% CI 0.4%–10.5%). The result, which differed from that in the primary trial,¹ was due to reduction in MI in the PCI subgroup on aspirin. PCI patients who were on aspirin did not have increased bleeding risk. This subgroup analysis, albeit small and limited, suggests that continuing low-dose aspirin in patients with previous PCI, irrespective of the type of stent or the time from stent implantations, minimizes the risk of perioperative MI.

GUIDELINES AND RECOMMENDATIONS

Routine perioperative use of aspirin increases the risk of bleeding without a reduction in ischemic events.¹ Patients with prior PCI are at increased risk of acute stent thrombosis when antiplatelet medications are discontinued.^20,21 Available data, although limited, support continuing low-dose aspirin without interruption in the perioperative period in PCI patients,⁴ as do the guidelines from the American College of Cardiology.⁵

We propose a management algorithm for patients undergoing noncardiac surgery on antiplatelet therapy that takes into consideration whether the surgery is urgent, elective, or time-sensitive (Figure 1). It is imperative to involve the cardiologist, surgeon, anesthesiologist, and the patient in the decision-making process.

In the perioperative setting for patients undergoing noncardiac surgery:

• Discontinue aspirin in patients without coronary heart disease, as bleeding risk outweighs benefit.
• Consider aspirin in patients at high risk for a major adverse cardiac event if benefits outweigh risk.
• Continue low-dose aspirin without interruption in patients with a coronary stent, irrespective of the type of stent.
• If a patient has had PCI with stent placement but is not currently on aspirin, talk with the patient and the treating cardiologist to find out why, and initiate aspirin if no contraindications exist.

A 61-year-old woman presented to our hematology clinic for evaluation of multiple episodes of bruising. The first episode occurred 8 months earlier, when she developed a large bruise after water skiing. Two months before coming to us, she went to her local emergency room because of new bruising and was found to have a prolonged activated partial thromboplastin time (aPTT) of 60 seconds (reference range 23.3–34.9), but she underwent no further testing at that time.

At presentation to our clinic, she reported having no fevers, night sweats, unintentional weight loss, swollen lymph nodes, joint pain, rashes, mouth sores, nosebleeds, or blood in the urine or stool. Her history was notable only for hypothyroidism, which was diagnosed in the previous year. Her medications included levothyroxine, vitamin D₃, and vitamin C. She had been taking a baby aspirin daily for the past 10 years but had stopped 1 month earlier because of the bruising.

Ten years earlier she had been evaluated for a possible transient ischemic attack; laboratory results at that time included a normal aPTT of 25.1 seconds and a normal factor VIII level of 153% (reference range 50%–173%).

EVALUATION FOR AN ISOLATED PROLONGED aPTT

1. What is the appropriate next test to evaluate this patient’s prolonged aPTT?

• Lupus anticoagulant panel
• Coagulation factor levels
• Mixing studies
• Bethesda assay

Mixing studies

Once a prolonged aPTT is confirmed, the appropriate next step is a mixing study. This involves mixing the patient’s plasma with pooled normal plasma in a 1-to-1 ratio, then repeating the aPTT test immediately, and again after 1 hour of incubation at 37°C. If the patient does not have enough of one of the coagulation factors, the aPTT immediately returns to the normal range when plasma is mixed with the pooled plasma because the pooled plasma contains the factor that is lacking. If this happens, then factor assays should be performed to identify the deficient factor.¹

Various antibodies that inhibit coagulation factors can also affect the aPTT. There are 2 general types: immediate-acting and delayed.

With an immediate-acting inhibitor, the aPTT does not correct into the normal range with initial mixing. Immediate-acting inhibitors are often seen together with lupus anticoagulants, which are nonspecific phospholipid antibodies. If an immediate-acting inhibitor is detected, further testing should focus on evaluation for lupus anticoagulant, including phospholipid-dependency studies.

With a delayed inhibitor, the aPTT initially comes down, but subsequently goes back up after incubation. Acquired factor VIII inhibitor is a classic delayed-type inhibitor and is also the most common factor inhibitor.¹ If a delayed-acting inhibitor is found, specific intrinsic factor levels should be measured (factors VIII, IX, XI, and XII),² and testing should also be done for lupus anticoagulant, as these inhibitors may occur together.

Bethesda assay

Case continued: Results of mixing and Bethesda studies

2. What is the most likely underlying condition associated with this patient’s factor VIII inhibitor?

• Autoimmune disease
• Malignancy
• A medication
• Unknown (idiopathic)

Acquired hemophilia A (AHA) is a rare disorder caused by autoantibodies against factor VIII. Its estimated incidence is about 1 person per million per year.⁴ It usually presents as unexplained bruising or bleeding and is only rarely diagnosed by an incidentally noted prolonged aPTT. The severity of bleeding is variable and can include subcutaneous, soft-tissue, retroperitoneal, gastrointestinal, and intracranial hemorrhage.⁵

AHA is considered idiopathic in more than half of cases. A study based on a European registry⁵ of 501 patients with AHA and a UK study⁶ of 172 patients found no underlying disease in 52% and 65% of patients, respectively. For patients with an identified cause, the most common causes were malignancy (12%⁵ and 15%⁶) and autoimmune disease (12%⁵ and 17%⁶).

Drugs have rarely been associated with factor VIII inhibitors. Such occurrences have been reported with interferon, blood thinners, antibiotics, and psychiatric medications, but no study yet has indicated causation. However, patients with congenital hemophilia A treated with factor VIII preparations have about a 15% chance of developing factor VIII inhibitors. In this setting, inhibitors develop in response to recombinant factor VIII exposure, unlike the autoimmune phenomena seen in AHA.

TREATMENT OF ACQUIRED HEMOPHILIA A

3. What is the most appropriate treatment for AHA?

• Desmopressin and prednisone
• Recombinant porcine factor VIII and prednisone plus cyclophosphamide
• Recombinant factor VIIa and rituximab
• Any of the above

Any of the above regimens can be used. In general, treatment of AHA has two purposes: to stop acute hemorrhage, and to reduce the level of factor VIII inhibitor. No standard treatment guidelines are available; evidence of the effectiveness of different drugs is based largely on data on congenital hemophilia A.³

Acute treatment to stop bleeding

Initial treatment of AHA often focuses on stopping an acute hemorrhage by either raising circulating levels of factor VIII or bypassing it in the coagulation cascade.

Desmopressin can temporarily raise factor VIII levels, but it is often ineffective in AHA unless the patient has very low inhibitor titers.³

Factor VIII concentrate (human or recombinant porcine factor VIII) may be effective in patients with low inhibitor titers (< 5 BU). Higher doses are often required than those used in congenital hemophilia A. Factor VIII concentrate is usually combined with immunosuppressive treatment to lower the factor VIII inhibitor level (described below).³

If these methods are ineffective or the patient has high inhibitor titers (> 5 BU), activated prothrombin complex concentrates, known as FEIBA (factor eight inhibitor bypassing activity), or recombinant factor VIIa is available. These agents bypass factor VIII in the clotting cascade.

Immunosuppression to reduce factor VIII inhibitor

Immunosuppressive agents are the mainstay of AHA treatment to lower the inhibitor level.

Regimens vary. A 2003 meta-analysis⁴ including 249 patients found that prednisone alone resulted in complete response in about 30% of patients, and the addition of cyclophosphamide increased the response rate to 60% to 100%. High-dose intravenous immunoglobulin led to conflicting results. Conclusions were limited by the variability of dosing and duration in treatment regimens among the 20 different studies included.

An analysis of 331 patients in the European Acquired Hemophilia Registry (EACH2)⁷ found that steroids alone produced remission in 48% of patients, while steroids combined with cyclophosphamide raised the rate to 70%. Rituximab-based regimens were successful in 59% but required twice as long to achieve remission as steroid or cyclophosphamide-based regimens. No benefit was noted from intravenous immunoglobulin.

Risks of disease and treatment

AHA is associated with significant risk of morbidity and death related to bleeding, complications of treatment, and underlying disease.

In EACH2, 16 of the 331 patients died of bleeding, 16 died of causes related to immunosuppression, and 45 died of causes related to the underlying condition.⁵ In the UK registry of 172 patients, 13 patients died of bleeding, and 12 died of sepsis related to immunosuppression.⁶

The factor VIII level and inhibitor titer are not necessarily useful in stratifying bleeding risk, as severe and fatal bleeding can occur at variable levels and patients remain at risk of bleeding as long as the inhibitor persists.^6,7

The patient was started on 60 mg daily of prednisone, resulting in a decrease in her aPTT, increase in factor VIII level, and lower Bethesda titer. On a return visit, her absolute lymphocyte count was 7.04 × 10⁹/L (reference range 1.0–4.0). She reported no fevers, chills, or recent infections.  

EVALUATING LYMPHOCYTOSIS

Lymphocytosis is defined in most laboratories as an absolute lymphocyte count greater than 4.0 × 10⁹/L for adults. Normally, T cells (CD3+) make up 60% to 80% of lymphocytes, B cells (CD20+) 10% to 20%, and natural killer (NK) cells (CD3–, CD56+) 5% to 10%. Lymphocytosis is usually caused by infection, but it can have other causes, including malignancy.

Peripheral blood smear. If there is no clear cause of lymphocytosis, a peripheral blood smear can be used to assess lymphocyte morphology, providing clues to the underlying etiology. For example, atypical lymphocytes are often seen in infectious mononucleosis, while “smudge” lymphocytes are characteristic of chronic lymphocytic leukemia. If a peripheral smear shows abnormal morphology, further workup should include establishing whether the lymphocytes are polyclonal or clonal.⁸

CASE CONTINUED: LARGE GRANULAR LYMPHOCYTES

Image provided by Karl Theil, MD, Clinical Pathology, Cleveland Clinic.

4. What is the next step to evaluate the patient’s lymphocytosis?

• Bone marrow biopsy
• Karyotype analysis
• Flow cytometry
• Fluorescence in situ hybridization

Flow cytometry with V-beta analysis is the best first test to determine the cause of lymphocytosis after review of the peripheral smear. For persistent lymphocytosis, flow cytometry should be done even if a peripheral smear shows normal lymphocyte morphology.

Most T cells possess receptors composed of alpha and beta chains, each encoded by variable (V), diversity (D), joining (J), and constant (C) gene segments. The V, D, and J segments undergo rearrangement during T-cell development in the thymus based on antigen exposure, producing a diverse T-cell receptor population.

In a polyclonal population of lymphocytes, the T-cell receptors have a variety of gene segment arrangements, indicating normal T-cell development. But in a clonal population of lymphocytes, the T-cell receptors have a single identical gene segment arrangement, indicating they all originated from a single clone.⁹ Lymphocytosis in response to an infection is typically polyclonal, while malignant lymphocytosis is clonal. 

Monoclonal antibodies against many of the variable regions of the beta chain (V-beta) of T-cell receptors have been developed, enabling flow cytometry to establish clonality.

T-cell receptor gene rearrangement studies can also be performed using polymerase chain reaction and Southern blot techniques.⁹

Karyotype analysis is usually not performed for the finding of LGLs, because most leukemias (eg, T-cell and NK-cell leukemias) have cells with a normal karyotype. 

Bone marrow biopsy is invasive and usually not required to evaluate LGLs. It can be especially risky for a patient with a bleeding disorder such as a factor VIII inhibitor.¹⁰

Case continued: Flow cytometry confirms clonality

Subsequent flow cytometry found that more than 50% of the patient’s lymphocytes were LGLs that co-expressed CD3+, CD8+, CD56+, and CD57+, with aberrantly decreased CD7 expression. T-cell V-beta analysis demonstrated an expansion of the V-beta 17 family, and T-cell receptor gene analysis with polymerase chain reaction confirmed the presence of a clonal rearrangement.

LGL LEUKEMIA: CLASSIFICATION AND MANAGEMENT

LGLs normally account for 10% to 15% of peripheral mononuclear cells.¹¹ LGL leukemia is caused by a clonal population of cytotoxic T cells or NK cells and involves an increased number of LGLs (usually > 2 × 10⁹/L).¹⁰

LGL leukemia is divided into 3 categories according to the most recent World Health Organization classification^10,12:

T-cell LGL leukemia (about 85% of cases) is considered indolent but can cause significant cytopenias and is often associated with autoimmune disease.¹³ Cells usually express a CD3+, CD8+, CD16+, and CD57+ phenotype. Survival is about 70% at 10 years.

Chronic NK-cell lymphocytosis (about 10%) also tends to have an indolent course with cytopenia and an autoimmune association, and with a similar prognosis to T-cell LGL leukemia. Cells express a CD3–, CD16+, and CD56+ phenotype.

Aggressive NK-cell LGL leukemia (about 5%) is associated with Epstein-Barr virus infection and occurs in younger patients. It is characterized by severe cytopenias, “B symptoms” (ie, fever, night sweats, weight loss), and has a very poor prognosis. Like chronic NK-cell lymphocytosis, cells express a CD3–, CD16+, and CD56+ phenotype. Fas (CD95) and Fas-ligand (CD178) are strongly expressed.^10,13

Most cases of LGL leukemia can be diagnosed on the basis of classic morphology on peripheral blood smear and evidence of clonality on flow cytometry or gene rearrangement studies. T-cell receptor gene studies cannot be used to establish clonality in the NK subtypes, as NK cells do not express T-cell receptors.¹¹

Case continued: Diagnosis, continued course 

In our patient, T-cell LGL leukemia was diagnosed on the basis of the peripheral smear, flow cytometry results, and positive T-cell receptor gene studies for clonal rearrangement in the T-cell receptor beta region.

While her corticosteroid therapy was being tapered, her factor III inhibitor level increased, and she had a small episode of bleeding, prompting the start of cyclophosphamide 50 mg daily with lower doses of prednisone.

Patients with LGL leukemia commonly have or develop autoimmune conditions. Immune-mediated cytopenias including pure red cell aplasia, aplastic anemia, and autoimmune hemolytic anemias can occur. Neutropenia, the most common cytopenia in LGL leukemia, is thought to be at least partly autoimmune, as the degree of neutropenia is often worse than would be expected solely from bone-marrow infiltration of LGL cells.^10,14,15

Rheumatoid arthritis is the most common autoimmune condition associated with LGL leukemia, with a reported incidence between 11% and 36%.^13–15

Felty syndrome (rheumatoid arthritis, splenomegaly, and neutropenia) is often associated with LGL leukemia and is thought by some to be part of the same disease process.¹⁵

Treat with immunosuppressives if needed

Indications for treating LGL leukemia include the development of cytopenias and associated autoimmune diseases. Immunosuppressive agents, such as methotrexate, cyclophosphamide, and cyclosporine, are commonly used.^10,11,14 Most evidence of treatment efficacy is from retrospective studies and case reports, with widely variable response rates that overall are around 50%.¹⁰

ACQUIRED HEMOPHILIA A AND HEMATOLOGIC MALIGNANCY

A systematic review found 30 cases of AHA associated with hematologic malignancies.¹⁶ The largest case series¹⁷ in this analysis had 8 patients, and included diagnoses of chronic lymphocytic leukemia, erythroleukemia, myelofibrosis, multiple myeloma, and myelodysplastic syndrome. In 3 of these patients, the appearance of the inhibitor preceded the diagnosis of the underlying malignancy by an average of 3.5 months. In 1 patient with erythroleukemia and another with multiple myeloma, the activity of the inhibitor could be clearly correlated with the underlying malignancy. In the other 6 patients, no association between the two could be made.

In the same series, complete resolution of the inhibitor was related only to the level of Bethesda titer present at diagnosis, with those who achieved resolution having lower mean Bethesda titers.¹⁷ Similarly, in EACH2, lower inhibitor Bethesda titers and higher factor VIII levels at presentation were associated with faster inhibitor eradication and normalization of factor VIII levels.⁷

Murphy et al¹⁸ described a 62-year-old woman with Felty syndrome who developed a factor VIII inhibitor and was subsequently given a diagnosis of LGL leukemia. Treatment with immunosuppressive agents, including cyclophosphamide, azathioprine, and rituximab, successfully eradicated her factor VIII inhibitor, although the LGL leukemia persisted.

Case conclusion: Eradication of factor VIII inhibitor

Our patient, similar to the patient described by Murphy et al¹⁸ above, had eradication of the factor VIII inhibitor despite persistence of LGL leukemia. Between the time of diagnosis at our clinic, when she had 54% LGLs, and eradication of the inhibitor 3 months later, the LGL percentage ranged from 45% to 89%. No clear direct correlation between LGL and factor VIII inhibitor levels could be detected.

Given the strong association of LGL leukemia with autoimmune disease, it is tempting to believe that her factor VIII inhibitor was somehow related to her malignancy, although the exact mechanism remained unclear. The average age at diagnosis is 60 for LGL leukemia¹¹ and over 70 for AHA,^5,6 so advanced age may be the common denominator. Whether or not our patient will have recurrence of her factor VIII inhibitor or the development of other autoimmune diseases with the persistence of her LGL leukemia remains to be seen.

At last follow-up, our patient was off all therapy and continued to have normal aPTT and factor VIII levels. Repeat flow cytometry after treatment of her factor VIII inhibitor showed persistence of a clonal T-cell population, although reduced from 72% to 60%. It may be that the 2 entities were unrelated, and the clonal T-cell population was simply fluctuating over time. This can be determined only with further observation. As the patient had no symptoms from her LGL leukemia, she continued to be observed without treatment.

TAKE-HOME POINTS

• The coagulation assay is key to initially assessing a bleeding abnormality; whether the prothrombin time and aPTT are normal or prolonged narrows the differential diagnosis and determines next steps in evaluation.
• Mixing studies can help pinpoint the responsible deficient factor.
• Acquired factor VIII deficiency, also known as AHA, may be caused by autoimmune disease, malignancy, or medications, but it is usually idiopathic.
• AHA treatment is focused on achieving hemostasis and reducing factor VIII inhibitor.
• Lymphocytosis should be evaluated with a peripheral blood smear and flow cytometry to determine if the population is polyclonal (associated with infection) or clonal (associated with malignancy).
• LGL leukemia is usually a chronic, indolent disease, although an uncommon subtype has an aggressive course.
• The association between AHA and LGL leukemia is unclear, and both conditions must be monitored and managed.

A 61-year-old woman presented to our hematology clinic for evaluation of multiple episodes of bruising. The first episode occurred 8 months earlier, when she developed a large bruise after water skiing. Two months before coming to us, she went to her local emergency room because of new bruising and was found to have a prolonged activated partial thromboplastin time (aPTT) of 60 seconds (reference range 23.3–34.9), but she underwent no further testing at that time.

At presentation to our clinic, she reported having no fevers, night sweats, unintentional weight loss, swollen lymph nodes, joint pain, rashes, mouth sores, nosebleeds, or blood in the urine or stool. Her history was notable only for hypothyroidism, which was diagnosed in the previous year. Her medications included levothyroxine, vitamin D₃, and vitamin C. She had been taking a baby aspirin daily for the past 10 years but had stopped 1 month earlier because of the bruising.

Ten years earlier she had been evaluated for a possible transient ischemic attack; laboratory results at that time included a normal aPTT of 25.1 seconds and a normal factor VIII level of 153% (reference range 50%–173%).

EVALUATION FOR AN ISOLATED PROLONGED aPTT

1. What is the appropriate next test to evaluate this patient’s prolonged aPTT?

• Lupus anticoagulant panel
• Coagulation factor levels
• Mixing studies
• Bethesda assay

Mixing studies

Once a prolonged aPTT is confirmed, the appropriate next step is a mixing study. This involves mixing the patient’s plasma with pooled normal plasma in a 1-to-1 ratio, then repeating the aPTT test immediately, and again after 1 hour of incubation at 37°C. If the patient does not have enough of one of the coagulation factors, the aPTT immediately returns to the normal range when plasma is mixed with the pooled plasma because the pooled plasma contains the factor that is lacking. If this happens, then factor assays should be performed to identify the deficient factor.¹

Various antibodies that inhibit coagulation factors can also affect the aPTT. There are 2 general types: immediate-acting and delayed.

With an immediate-acting inhibitor, the aPTT does not correct into the normal range with initial mixing. Immediate-acting inhibitors are often seen together with lupus anticoagulants, which are nonspecific phospholipid antibodies. If an immediate-acting inhibitor is detected, further testing should focus on evaluation for lupus anticoagulant, including phospholipid-dependency studies.

With a delayed inhibitor, the aPTT initially comes down, but subsequently goes back up after incubation. Acquired factor VIII inhibitor is a classic delayed-type inhibitor and is also the most common factor inhibitor.¹ If a delayed-acting inhibitor is found, specific intrinsic factor levels should be measured (factors VIII, IX, XI, and XII),² and testing should also be done for lupus anticoagulant, as these inhibitors may occur together.

Bethesda assay

Case continued: Results of mixing and Bethesda studies

2. What is the most likely underlying condition associated with this patient’s factor VIII inhibitor?

• Autoimmune disease
• Malignancy
• A medication
• Unknown (idiopathic)

Acquired hemophilia A (AHA) is a rare disorder caused by autoantibodies against factor VIII. Its estimated incidence is about 1 person per million per year.⁴ It usually presents as unexplained bruising or bleeding and is only rarely diagnosed by an incidentally noted prolonged aPTT. The severity of bleeding is variable and can include subcutaneous, soft-tissue, retroperitoneal, gastrointestinal, and intracranial hemorrhage.⁵

AHA is considered idiopathic in more than half of cases. A study based on a European registry⁵ of 501 patients with AHA and a UK study⁶ of 172 patients found no underlying disease in 52% and 65% of patients, respectively. For patients with an identified cause, the most common causes were malignancy (12%⁵ and 15%⁶) and autoimmune disease (12%⁵ and 17%⁶).

Drugs have rarely been associated with factor VIII inhibitors. Such occurrences have been reported with interferon, blood thinners, antibiotics, and psychiatric medications, but no study yet has indicated causation. However, patients with congenital hemophilia A treated with factor VIII preparations have about a 15% chance of developing factor VIII inhibitors. In this setting, inhibitors develop in response to recombinant factor VIII exposure, unlike the autoimmune phenomena seen in AHA.

TREATMENT OF ACQUIRED HEMOPHILIA A

3. What is the most appropriate treatment for AHA?

• Desmopressin and prednisone
• Recombinant porcine factor VIII and prednisone plus cyclophosphamide
• Recombinant factor VIIa and rituximab
• Any of the above

Any of the above regimens can be used. In general, treatment of AHA has two purposes: to stop acute hemorrhage, and to reduce the level of factor VIII inhibitor. No standard treatment guidelines are available; evidence of the effectiveness of different drugs is based largely on data on congenital hemophilia A.³

Acute treatment to stop bleeding

Initial treatment of AHA often focuses on stopping an acute hemorrhage by either raising circulating levels of factor VIII or bypassing it in the coagulation cascade.

Desmopressin can temporarily raise factor VIII levels, but it is often ineffective in AHA unless the patient has very low inhibitor titers.³

Factor VIII concentrate (human or recombinant porcine factor VIII) may be effective in patients with low inhibitor titers (< 5 BU). Higher doses are often required than those used in congenital hemophilia A. Factor VIII concentrate is usually combined with immunosuppressive treatment to lower the factor VIII inhibitor level (described below).³

If these methods are ineffective or the patient has high inhibitor titers (> 5 BU), activated prothrombin complex concentrates, known as FEIBA (factor eight inhibitor bypassing activity), or recombinant factor VIIa is available. These agents bypass factor VIII in the clotting cascade.

Immunosuppression to reduce factor VIII inhibitor

Immunosuppressive agents are the mainstay of AHA treatment to lower the inhibitor level.

Regimens vary. A 2003 meta-analysis⁴ including 249 patients found that prednisone alone resulted in complete response in about 30% of patients, and the addition of cyclophosphamide increased the response rate to 60% to 100%. High-dose intravenous immunoglobulin led to conflicting results. Conclusions were limited by the variability of dosing and duration in treatment regimens among the 20 different studies included.

An analysis of 331 patients in the European Acquired Hemophilia Registry (EACH2)⁷ found that steroids alone produced remission in 48% of patients, while steroids combined with cyclophosphamide raised the rate to 70%. Rituximab-based regimens were successful in 59% but required twice as long to achieve remission as steroid or cyclophosphamide-based regimens. No benefit was noted from intravenous immunoglobulin.

Risks of disease and treatment

AHA is associated with significant risk of morbidity and death related to bleeding, complications of treatment, and underlying disease.

In EACH2, 16 of the 331 patients died of bleeding, 16 died of causes related to immunosuppression, and 45 died of causes related to the underlying condition.⁵ In the UK registry of 172 patients, 13 patients died of bleeding, and 12 died of sepsis related to immunosuppression.⁶

The factor VIII level and inhibitor titer are not necessarily useful in stratifying bleeding risk, as severe and fatal bleeding can occur at variable levels and patients remain at risk of bleeding as long as the inhibitor persists.^6,7

The patient was started on 60 mg daily of prednisone, resulting in a decrease in her aPTT, increase in factor VIII level, and lower Bethesda titer. On a return visit, her absolute lymphocyte count was 7.04 × 10⁹/L (reference range 1.0–4.0). She reported no fevers, chills, or recent infections.  

EVALUATING LYMPHOCYTOSIS

Lymphocytosis is defined in most laboratories as an absolute lymphocyte count greater than 4.0 × 10⁹/L for adults. Normally, T cells (CD3+) make up 60% to 80% of lymphocytes, B cells (CD20+) 10% to 20%, and natural killer (NK) cells (CD3–, CD56+) 5% to 10%. Lymphocytosis is usually caused by infection, but it can have other causes, including malignancy.

Peripheral blood smear. If there is no clear cause of lymphocytosis, a peripheral blood smear can be used to assess lymphocyte morphology, providing clues to the underlying etiology. For example, atypical lymphocytes are often seen in infectious mononucleosis, while “smudge” lymphocytes are characteristic of chronic lymphocytic leukemia. If a peripheral smear shows abnormal morphology, further workup should include establishing whether the lymphocytes are polyclonal or clonal.⁸

CASE CONTINUED: LARGE GRANULAR LYMPHOCYTES

Image provided by Karl Theil, MD, Clinical Pathology, Cleveland Clinic.

4. What is the next step to evaluate the patient’s lymphocytosis?

• Bone marrow biopsy
• Karyotype analysis
• Flow cytometry
• Fluorescence in situ hybridization

Flow cytometry with V-beta analysis is the best first test to determine the cause of lymphocytosis after review of the peripheral smear. For persistent lymphocytosis, flow cytometry should be done even if a peripheral smear shows normal lymphocyte morphology.

Most T cells possess receptors composed of alpha and beta chains, each encoded by variable (V), diversity (D), joining (J), and constant (C) gene segments. The V, D, and J segments undergo rearrangement during T-cell development in the thymus based on antigen exposure, producing a diverse T-cell receptor population.

In a polyclonal population of lymphocytes, the T-cell receptors have a variety of gene segment arrangements, indicating normal T-cell development. But in a clonal population of lymphocytes, the T-cell receptors have a single identical gene segment arrangement, indicating they all originated from a single clone.⁹ Lymphocytosis in response to an infection is typically polyclonal, while malignant lymphocytosis is clonal. 

Monoclonal antibodies against many of the variable regions of the beta chain (V-beta) of T-cell receptors have been developed, enabling flow cytometry to establish clonality.

T-cell receptor gene rearrangement studies can also be performed using polymerase chain reaction and Southern blot techniques.⁹

Karyotype analysis is usually not performed for the finding of LGLs, because most leukemias (eg, T-cell and NK-cell leukemias) have cells with a normal karyotype. 

Bone marrow biopsy is invasive and usually not required to evaluate LGLs. It can be especially risky for a patient with a bleeding disorder such as a factor VIII inhibitor.¹⁰

Case continued: Flow cytometry confirms clonality

Subsequent flow cytometry found that more than 50% of the patient’s lymphocytes were LGLs that co-expressed CD3+, CD8+, CD56+, and CD57+, with aberrantly decreased CD7 expression. T-cell V-beta analysis demonstrated an expansion of the V-beta 17 family, and T-cell receptor gene analysis with polymerase chain reaction confirmed the presence of a clonal rearrangement.

LGL LEUKEMIA: CLASSIFICATION AND MANAGEMENT

LGLs normally account for 10% to 15% of peripheral mononuclear cells.¹¹ LGL leukemia is caused by a clonal population of cytotoxic T cells or NK cells and involves an increased number of LGLs (usually > 2 × 10⁹/L).¹⁰

LGL leukemia is divided into 3 categories according to the most recent World Health Organization classification^10,12:

T-cell LGL leukemia (about 85% of cases) is considered indolent but can cause significant cytopenias and is often associated with autoimmune disease.¹³ Cells usually express a CD3+, CD8+, CD16+, and CD57+ phenotype. Survival is about 70% at 10 years.

Chronic NK-cell lymphocytosis (about 10%) also tends to have an indolent course with cytopenia and an autoimmune association, and with a similar prognosis to T-cell LGL leukemia. Cells express a CD3–, CD16+, and CD56+ phenotype.

Aggressive NK-cell LGL leukemia (about 5%) is associated with Epstein-Barr virus infection and occurs in younger patients. It is characterized by severe cytopenias, “B symptoms” (ie, fever, night sweats, weight loss), and has a very poor prognosis. Like chronic NK-cell lymphocytosis, cells express a CD3–, CD16+, and CD56+ phenotype. Fas (CD95) and Fas-ligand (CD178) are strongly expressed.^10,13

Most cases of LGL leukemia can be diagnosed on the basis of classic morphology on peripheral blood smear and evidence of clonality on flow cytometry or gene rearrangement studies. T-cell receptor gene studies cannot be used to establish clonality in the NK subtypes, as NK cells do not express T-cell receptors.¹¹

Case continued: Diagnosis, continued course 

In our patient, T-cell LGL leukemia was diagnosed on the basis of the peripheral smear, flow cytometry results, and positive T-cell receptor gene studies for clonal rearrangement in the T-cell receptor beta region.

While her corticosteroid therapy was being tapered, her factor III inhibitor level increased, and she had a small episode of bleeding, prompting the start of cyclophosphamide 50 mg daily with lower doses of prednisone.

Patients with LGL leukemia commonly have or develop autoimmune conditions. Immune-mediated cytopenias including pure red cell aplasia, aplastic anemia, and autoimmune hemolytic anemias can occur. Neutropenia, the most common cytopenia in LGL leukemia, is thought to be at least partly autoimmune, as the degree of neutropenia is often worse than would be expected solely from bone-marrow infiltration of LGL cells.^10,14,15

Rheumatoid arthritis is the most common autoimmune condition associated with LGL leukemia, with a reported incidence between 11% and 36%.^13–15

Felty syndrome (rheumatoid arthritis, splenomegaly, and neutropenia) is often associated with LGL leukemia and is thought by some to be part of the same disease process.¹⁵

Treat with immunosuppressives if needed

Indications for treating LGL leukemia include the development of cytopenias and associated autoimmune diseases. Immunosuppressive agents, such as methotrexate, cyclophosphamide, and cyclosporine, are commonly used.^10,11,14 Most evidence of treatment efficacy is from retrospective studies and case reports, with widely variable response rates that overall are around 50%.¹⁰

ACQUIRED HEMOPHILIA A AND HEMATOLOGIC MALIGNANCY

A systematic review found 30 cases of AHA associated with hematologic malignancies.¹⁶ The largest case series¹⁷ in this analysis had 8 patients, and included diagnoses of chronic lymphocytic leukemia, erythroleukemia, myelofibrosis, multiple myeloma, and myelodysplastic syndrome. In 3 of these patients, the appearance of the inhibitor preceded the diagnosis of the underlying malignancy by an average of 3.5 months. In 1 patient with erythroleukemia and another with multiple myeloma, the activity of the inhibitor could be clearly correlated with the underlying malignancy. In the other 6 patients, no association between the two could be made.

In the same series, complete resolution of the inhibitor was related only to the level of Bethesda titer present at diagnosis, with those who achieved resolution having lower mean Bethesda titers.¹⁷ Similarly, in EACH2, lower inhibitor Bethesda titers and higher factor VIII levels at presentation were associated with faster inhibitor eradication and normalization of factor VIII levels.⁷

Murphy et al¹⁸ described a 62-year-old woman with Felty syndrome who developed a factor VIII inhibitor and was subsequently given a diagnosis of LGL leukemia. Treatment with immunosuppressive agents, including cyclophosphamide, azathioprine, and rituximab, successfully eradicated her factor VIII inhibitor, although the LGL leukemia persisted.

Case conclusion: Eradication of factor VIII inhibitor

Our patient, similar to the patient described by Murphy et al¹⁸ above, had eradication of the factor VIII inhibitor despite persistence of LGL leukemia. Between the time of diagnosis at our clinic, when she had 54% LGLs, and eradication of the inhibitor 3 months later, the LGL percentage ranged from 45% to 89%. No clear direct correlation between LGL and factor VIII inhibitor levels could be detected.

Given the strong association of LGL leukemia with autoimmune disease, it is tempting to believe that her factor VIII inhibitor was somehow related to her malignancy, although the exact mechanism remained unclear. The average age at diagnosis is 60 for LGL leukemia¹¹ and over 70 for AHA,^5,6 so advanced age may be the common denominator. Whether or not our patient will have recurrence of her factor VIII inhibitor or the development of other autoimmune diseases with the persistence of her LGL leukemia remains to be seen.

At last follow-up, our patient was off all therapy and continued to have normal aPTT and factor VIII levels. Repeat flow cytometry after treatment of her factor VIII inhibitor showed persistence of a clonal T-cell population, although reduced from 72% to 60%. It may be that the 2 entities were unrelated, and the clonal T-cell population was simply fluctuating over time. This can be determined only with further observation. As the patient had no symptoms from her LGL leukemia, she continued to be observed without treatment.

TAKE-HOME POINTS

• The coagulation assay is key to initially assessing a bleeding abnormality; whether the prothrombin time and aPTT are normal or prolonged narrows the differential diagnosis and determines next steps in evaluation.
• Mixing studies can help pinpoint the responsible deficient factor.
• Acquired factor VIII deficiency, also known as AHA, may be caused by autoimmune disease, malignancy, or medications, but it is usually idiopathic.
• AHA treatment is focused on achieving hemostasis and reducing factor VIII inhibitor.
• Lymphocytosis should be evaluated with a peripheral blood smear and flow cytometry to determine if the population is polyclonal (associated with infection) or clonal (associated with malignancy).
• LGL leukemia is usually a chronic, indolent disease, although an uncommon subtype has an aggressive course.
• The association between AHA and LGL leukemia is unclear, and both conditions must be monitored and managed.

A 61-year-old woman presented to our hematology clinic for evaluation of multiple episodes of bruising. The first episode occurred 8 months earlier, when she developed a large bruise after water skiing. Two months before coming to us, she went to her local emergency room because of new bruising and was found to have a prolonged activated partial thromboplastin time (aPTT) of 60 seconds (reference range 23.3–34.9), but she underwent no further testing at that time.

At presentation to our clinic, she reported having no fevers, night sweats, unintentional weight loss, swollen lymph nodes, joint pain, rashes, mouth sores, nosebleeds, or blood in the urine or stool. Her history was notable only for hypothyroidism, which was diagnosed in the previous year. Her medications included levothyroxine, vitamin D₃, and vitamin C. She had been taking a baby aspirin daily for the past 10 years but had stopped 1 month earlier because of the bruising.

Ten years earlier she had been evaluated for a possible transient ischemic attack; laboratory results at that time included a normal aPTT of 25.1 seconds and a normal factor VIII level of 153% (reference range 50%–173%).

EVALUATION FOR AN ISOLATED PROLONGED aPTT

1. What is the appropriate next test to evaluate this patient’s prolonged aPTT?

• Lupus anticoagulant panel
• Coagulation factor levels
• Mixing studies
• Bethesda assay

Mixing studies

Once a prolonged aPTT is confirmed, the appropriate next step is a mixing study. This involves mixing the patient’s plasma with pooled normal plasma in a 1-to-1 ratio, then repeating the aPTT test immediately, and again after 1 hour of incubation at 37°C. If the patient does not have enough of one of the coagulation factors, the aPTT immediately returns to the normal range when plasma is mixed with the pooled plasma because the pooled plasma contains the factor that is lacking. If this happens, then factor assays should be performed to identify the deficient factor.¹

Various antibodies that inhibit coagulation factors can also affect the aPTT. There are 2 general types: immediate-acting and delayed.

With an immediate-acting inhibitor, the aPTT does not correct into the normal range with initial mixing. Immediate-acting inhibitors are often seen together with lupus anticoagulants, which are nonspecific phospholipid antibodies. If an immediate-acting inhibitor is detected, further testing should focus on evaluation for lupus anticoagulant, including phospholipid-dependency studies.

With a delayed inhibitor, the aPTT initially comes down, but subsequently goes back up after incubation. Acquired factor VIII inhibitor is a classic delayed-type inhibitor and is also the most common factor inhibitor.¹ If a delayed-acting inhibitor is found, specific intrinsic factor levels should be measured (factors VIII, IX, XI, and XII),² and testing should also be done for lupus anticoagulant, as these inhibitors may occur together.

Bethesda assay

Case continued: Results of mixing and Bethesda studies

2. What is the most likely underlying condition associated with this patient’s factor VIII inhibitor?

• Autoimmune disease
• Malignancy
• A medication
• Unknown (idiopathic)

Acquired hemophilia A (AHA) is a rare disorder caused by autoantibodies against factor VIII. Its estimated incidence is about 1 person per million per year.⁴ It usually presents as unexplained bruising or bleeding and is only rarely diagnosed by an incidentally noted prolonged aPTT. The severity of bleeding is variable and can include subcutaneous, soft-tissue, retroperitoneal, gastrointestinal, and intracranial hemorrhage.⁵

AHA is considered idiopathic in more than half of cases. A study based on a European registry⁵ of 501 patients with AHA and a UK study⁶ of 172 patients found no underlying disease in 52% and 65% of patients, respectively. For patients with an identified cause, the most common causes were malignancy (12%⁵ and 15%⁶) and autoimmune disease (12%⁵ and 17%⁶).

Drugs have rarely been associated with factor VIII inhibitors. Such occurrences have been reported with interferon, blood thinners, antibiotics, and psychiatric medications, but no study yet has indicated causation. However, patients with congenital hemophilia A treated with factor VIII preparations have about a 15% chance of developing factor VIII inhibitors. In this setting, inhibitors develop in response to recombinant factor VIII exposure, unlike the autoimmune phenomena seen in AHA.

TREATMENT OF ACQUIRED HEMOPHILIA A

3. What is the most appropriate treatment for AHA?

• Desmopressin and prednisone
• Recombinant porcine factor VIII and prednisone plus cyclophosphamide
• Recombinant factor VIIa and rituximab
• Any of the above

Any of the above regimens can be used. In general, treatment of AHA has two purposes: to stop acute hemorrhage, and to reduce the level of factor VIII inhibitor. No standard treatment guidelines are available; evidence of the effectiveness of different drugs is based largely on data on congenital hemophilia A.³

Acute treatment to stop bleeding

Initial treatment of AHA often focuses on stopping an acute hemorrhage by either raising circulating levels of factor VIII or bypassing it in the coagulation cascade.

Desmopressin can temporarily raise factor VIII levels, but it is often ineffective in AHA unless the patient has very low inhibitor titers.³

Factor VIII concentrate (human or recombinant porcine factor VIII) may be effective in patients with low inhibitor titers (< 5 BU). Higher doses are often required than those used in congenital hemophilia A. Factor VIII concentrate is usually combined with immunosuppressive treatment to lower the factor VIII inhibitor level (described below).³

If these methods are ineffective or the patient has high inhibitor titers (> 5 BU), activated prothrombin complex concentrates, known as FEIBA (factor eight inhibitor bypassing activity), or recombinant factor VIIa is available. These agents bypass factor VIII in the clotting cascade.

Immunosuppression to reduce factor VIII inhibitor

Immunosuppressive agents are the mainstay of AHA treatment to lower the inhibitor level.

Regimens vary. A 2003 meta-analysis⁴ including 249 patients found that prednisone alone resulted in complete response in about 30% of patients, and the addition of cyclophosphamide increased the response rate to 60% to 100%. High-dose intravenous immunoglobulin led to conflicting results. Conclusions were limited by the variability of dosing and duration in treatment regimens among the 20 different studies included.

An analysis of 331 patients in the European Acquired Hemophilia Registry (EACH2)⁷ found that steroids alone produced remission in 48% of patients, while steroids combined with cyclophosphamide raised the rate to 70%. Rituximab-based regimens were successful in 59% but required twice as long to achieve remission as steroid or cyclophosphamide-based regimens. No benefit was noted from intravenous immunoglobulin.

Risks of disease and treatment

AHA is associated with significant risk of morbidity and death related to bleeding, complications of treatment, and underlying disease.

In EACH2, 16 of the 331 patients died of bleeding, 16 died of causes related to immunosuppression, and 45 died of causes related to the underlying condition.⁵ In the UK registry of 172 patients, 13 patients died of bleeding, and 12 died of sepsis related to immunosuppression.⁶

The factor VIII level and inhibitor titer are not necessarily useful in stratifying bleeding risk, as severe and fatal bleeding can occur at variable levels and patients remain at risk of bleeding as long as the inhibitor persists.^6,7

The patient was started on 60 mg daily of prednisone, resulting in a decrease in her aPTT, increase in factor VIII level, and lower Bethesda titer. On a return visit, her absolute lymphocyte count was 7.04 × 10⁹/L (reference range 1.0–4.0). She reported no fevers, chills, or recent infections.  

EVALUATING LYMPHOCYTOSIS

Lymphocytosis is defined in most laboratories as an absolute lymphocyte count greater than 4.0 × 10⁹/L for adults. Normally, T cells (CD3+) make up 60% to 80% of lymphocytes, B cells (CD20+) 10% to 20%, and natural killer (NK) cells (CD3–, CD56+) 5% to 10%. Lymphocytosis is usually caused by infection, but it can have other causes, including malignancy.

Peripheral blood smear. If there is no clear cause of lymphocytosis, a peripheral blood smear can be used to assess lymphocyte morphology, providing clues to the underlying etiology. For example, atypical lymphocytes are often seen in infectious mononucleosis, while “smudge” lymphocytes are characteristic of chronic lymphocytic leukemia. If a peripheral smear shows abnormal morphology, further workup should include establishing whether the lymphocytes are polyclonal or clonal.⁸

CASE CONTINUED: LARGE GRANULAR LYMPHOCYTES

Image provided by Karl Theil, MD, Clinical Pathology, Cleveland Clinic.

4. What is the next step to evaluate the patient’s lymphocytosis?

• Bone marrow biopsy
• Karyotype analysis
• Flow cytometry
• Fluorescence in situ hybridization

Flow cytometry with V-beta analysis is the best first test to determine the cause of lymphocytosis after review of the peripheral smear. For persistent lymphocytosis, flow cytometry should be done even if a peripheral smear shows normal lymphocyte morphology.

Most T cells possess receptors composed of alpha and beta chains, each encoded by variable (V), diversity (D), joining (J), and constant (C) gene segments. The V, D, and J segments undergo rearrangement during T-cell development in the thymus based on antigen exposure, producing a diverse T-cell receptor population.

In a polyclonal population of lymphocytes, the T-cell receptors have a variety of gene segment arrangements, indicating normal T-cell development. But in a clonal population of lymphocytes, the T-cell receptors have a single identical gene segment arrangement, indicating they all originated from a single clone.⁹ Lymphocytosis in response to an infection is typically polyclonal, while malignant lymphocytosis is clonal. 

Monoclonal antibodies against many of the variable regions of the beta chain (V-beta) of T-cell receptors have been developed, enabling flow cytometry to establish clonality.

T-cell receptor gene rearrangement studies can also be performed using polymerase chain reaction and Southern blot techniques.⁹

Karyotype analysis is usually not performed for the finding of LGLs, because most leukemias (eg, T-cell and NK-cell leukemias) have cells with a normal karyotype. 

Bone marrow biopsy is invasive and usually not required to evaluate LGLs. It can be especially risky for a patient with a bleeding disorder such as a factor VIII inhibitor.¹⁰

Case continued: Flow cytometry confirms clonality

Subsequent flow cytometry found that more than 50% of the patient’s lymphocytes were LGLs that co-expressed CD3+, CD8+, CD56+, and CD57+, with aberrantly decreased CD7 expression. T-cell V-beta analysis demonstrated an expansion of the V-beta 17 family, and T-cell receptor gene analysis with polymerase chain reaction confirmed the presence of a clonal rearrangement.

LGL LEUKEMIA: CLASSIFICATION AND MANAGEMENT

LGLs normally account for 10% to 15% of peripheral mononuclear cells.¹¹ LGL leukemia is caused by a clonal population of cytotoxic T cells or NK cells and involves an increased number of LGLs (usually > 2 × 10⁹/L).¹⁰

LGL leukemia is divided into 3 categories according to the most recent World Health Organization classification^10,12:

T-cell LGL leukemia (about 85% of cases) is considered indolent but can cause significant cytopenias and is often associated with autoimmune disease.¹³ Cells usually express a CD3+, CD8+, CD16+, and CD57+ phenotype. Survival is about 70% at 10 years.

Chronic NK-cell lymphocytosis (about 10%) also tends to have an indolent course with cytopenia and an autoimmune association, and with a similar prognosis to T-cell LGL leukemia. Cells express a CD3–, CD16+, and CD56+ phenotype.

Aggressive NK-cell LGL leukemia (about 5%) is associated with Epstein-Barr virus infection and occurs in younger patients. It is characterized by severe cytopenias, “B symptoms” (ie, fever, night sweats, weight loss), and has a very poor prognosis. Like chronic NK-cell lymphocytosis, cells express a CD3–, CD16+, and CD56+ phenotype. Fas (CD95) and Fas-ligand (CD178) are strongly expressed.^10,13

Most cases of LGL leukemia can be diagnosed on the basis of classic morphology on peripheral blood smear and evidence of clonality on flow cytometry or gene rearrangement studies. T-cell receptor gene studies cannot be used to establish clonality in the NK subtypes, as NK cells do not express T-cell receptors.¹¹

Case continued: Diagnosis, continued course 

In our patient, T-cell LGL leukemia was diagnosed on the basis of the peripheral smear, flow cytometry results, and positive T-cell receptor gene studies for clonal rearrangement in the T-cell receptor beta region.

While her corticosteroid therapy was being tapered, her factor III inhibitor level increased, and she had a small episode of bleeding, prompting the start of cyclophosphamide 50 mg daily with lower doses of prednisone.

Patients with LGL leukemia commonly have or develop autoimmune conditions. Immune-mediated cytopenias including pure red cell aplasia, aplastic anemia, and autoimmune hemolytic anemias can occur. Neutropenia, the most common cytopenia in LGL leukemia, is thought to be at least partly autoimmune, as the degree of neutropenia is often worse than would be expected solely from bone-marrow infiltration of LGL cells.^10,14,15

Rheumatoid arthritis is the most common autoimmune condition associated with LGL leukemia, with a reported incidence between 11% and 36%.^13–15

Felty syndrome (rheumatoid arthritis, splenomegaly, and neutropenia) is often associated with LGL leukemia and is thought by some to be part of the same disease process.¹⁵

Treat with immunosuppressives if needed

Indications for treating LGL leukemia include the development of cytopenias and associated autoimmune diseases. Immunosuppressive agents, such as methotrexate, cyclophosphamide, and cyclosporine, are commonly used.^10,11,14 Most evidence of treatment efficacy is from retrospective studies and case reports, with widely variable response rates that overall are around 50%.¹⁰

ACQUIRED HEMOPHILIA A AND HEMATOLOGIC MALIGNANCY

A systematic review found 30 cases of AHA associated with hematologic malignancies.¹⁶ The largest case series¹⁷ in this analysis had 8 patients, and included diagnoses of chronic lymphocytic leukemia, erythroleukemia, myelofibrosis, multiple myeloma, and myelodysplastic syndrome. In 3 of these patients, the appearance of the inhibitor preceded the diagnosis of the underlying malignancy by an average of 3.5 months. In 1 patient with erythroleukemia and another with multiple myeloma, the activity of the inhibitor could be clearly correlated with the underlying malignancy. In the other 6 patients, no association between the two could be made.

In the same series, complete resolution of the inhibitor was related only to the level of Bethesda titer present at diagnosis, with those who achieved resolution having lower mean Bethesda titers.¹⁷ Similarly, in EACH2, lower inhibitor Bethesda titers and higher factor VIII levels at presentation were associated with faster inhibitor eradication and normalization of factor VIII levels.⁷

Murphy et al¹⁸ described a 62-year-old woman with Felty syndrome who developed a factor VIII inhibitor and was subsequently given a diagnosis of LGL leukemia. Treatment with immunosuppressive agents, including cyclophosphamide, azathioprine, and rituximab, successfully eradicated her factor VIII inhibitor, although the LGL leukemia persisted.

Case conclusion: Eradication of factor VIII inhibitor

Our patient, similar to the patient described by Murphy et al¹⁸ above, had eradication of the factor VIII inhibitor despite persistence of LGL leukemia. Between the time of diagnosis at our clinic, when she had 54% LGLs, and eradication of the inhibitor 3 months later, the LGL percentage ranged from 45% to 89%. No clear direct correlation between LGL and factor VIII inhibitor levels could be detected.

Given the strong association of LGL leukemia with autoimmune disease, it is tempting to believe that her factor VIII inhibitor was somehow related to her malignancy, although the exact mechanism remained unclear. The average age at diagnosis is 60 for LGL leukemia¹¹ and over 70 for AHA,^5,6 so advanced age may be the common denominator. Whether or not our patient will have recurrence of her factor VIII inhibitor or the development of other autoimmune diseases with the persistence of her LGL leukemia remains to be seen.

At last follow-up, our patient was off all therapy and continued to have normal aPTT and factor VIII levels. Repeat flow cytometry after treatment of her factor VIII inhibitor showed persistence of a clonal T-cell population, although reduced from 72% to 60%. It may be that the 2 entities were unrelated, and the clonal T-cell population was simply fluctuating over time. This can be determined only with further observation. As the patient had no symptoms from her LGL leukemia, she continued to be observed without treatment.

TAKE-HOME POINTS

• The coagulation assay is key to initially assessing a bleeding abnormality; whether the prothrombin time and aPTT are normal or prolonged narrows the differential diagnosis and determines next steps in evaluation.
• Mixing studies can help pinpoint the responsible deficient factor.
• Acquired factor VIII deficiency, also known as AHA, may be caused by autoimmune disease, malignancy, or medications, but it is usually idiopathic.
• AHA treatment is focused on achieving hemostasis and reducing factor VIII inhibitor.
• Lymphocytosis should be evaluated with a peripheral blood smear and flow cytometry to determine if the population is polyclonal (associated with infection) or clonal (associated with malignancy).
• LGL leukemia is usually a chronic, indolent disease, although an uncommon subtype has an aggressive course.
• The association between AHA and LGL leukemia is unclear, and both conditions must be monitored and managed.

A previously healthy 44-year-old woman presents to the emergency department with 1 day of fever, flank pain, dysuria, and persistent nausea and vomiting. Her temperature is 38.7°C (101.7°F), heart rate 102 beats per minute, and blood pressure 120/70 mm Hg. She has costovertebral angle tenderness. Laboratory testing reveals mild leukocytosis and a normal serum creatinine level; urinalysis shows leukocytes, as well as leukocyte esterase and nitrites. She has no personal or family history of nephrolithiasis. Urine cultures are obtained, and she is started on intravenous antibiotics and intravenous hydration to treat pyelonephritis.

Is imaging indicated at this point? And if so, which study is recommended?

KEY FEATURES

Acute pyelonephritis, infection of the renal parenchyma and collecting system, most often results from an ascending infection of the lower urinary tract. It is estimated to account for 250,000 office visits and 200,000 hospital admissions each year in the United States.¹

Lower urinary tract symptoms such as urinary frequency, urgency, and dysuria accompanied by fever, nausea, vomiting, and flank pain raise suspicion for acute pyelonephritis. Flank pain is a key, nearly universal feature of upper urinary tract infection in patients without diabetes, though it may be absent in up to 50% of patients with diabetes.²

Additional findings include costovertebral angle tenderness on physical examination and leukocytosis, pyuria, and bacteriuria on laboratory studies.

PREDICTING THE NEED FOR EARLY IMAGING

Though guidelines state that imaging is inappropriate in most patients with pyeloneph­ritis,^2–4 it is nevertheless often done for diagnosis or identification of complications, which have been reported in more than two-thirds of patients.^2–4

Acute pyelonephritis is generally classified as complicated or uncomplicated, though different definitions exist with regard to these classifications. The American College of Radiology’s Appropriateness Criteria2 consider patients with diabetes, immune compromise, a history of urolithiasis, or anatomic abnormality to be at highest risk for complications, and therefore recommend early imaging to assess for hydronephrosis, pyonephrosis, emphysematous pyelonephritis, and intrinsic or perinephric abscess.²

A clinical rule for predicting the need for imaging in acute pyelonephritis was developed and validated in an emergency department population in the Netherlands.³ The study suggested that restricting early imaging to patients with a history of urolithiasis, a urine pH of 7.0 or higher, or renal insufficiency—defined as a glomerular filtration rate (GFR) of 40 mL/min/1.73m² or lower as estimated by the Modification of Diet in Renal Disease formula—would provide a negative predictive value of 94% to 100% for detection of an urgent urologic disorder (pyonephrosis, renal abscess, or urolithiasis). This high negative predictive value highlights that an absence of these signs and symptoms can safely identify patients who do not need renal imaging.

The positive predictive value was less useful, as only 5% to 23% of patients who had at least 1 risk factor went on to have urgent urologic risk factors.³

Implementation of this prediction rule would have resulted in a relative reduction in imaging of 40% and an absolute reduction of 28%. Of note, use of reduced GFR in this prediction rule is not clearly validated for patients with chronic kidney disease, as the previous GFR for most patients in this study was unknown.³

Based on these data, initial imaging is recommended in patients with diabetes, immune compromise, a history of urolithiasis, anatomic abnormality, a urine pH 7.0 or higher, or a GFR 40 mL/min or lower in a patient with no history of significant renal dysfunction. Early imaging would also be reasonable in patients with a complex clinical presentation, early recurrence of symptoms after treatment, clinical decompensation, or critical illness.

In a retrospective review of 62 patients hospitalized for acute renal infection, Soulen et al5 found that the most reliable indicator of complicated acute pyelonephritis was the persistence of fever and leukocytosis at 72 hours. And another small prospective study of patients with uncomplicated pyelonephritis reported a time to defervescence of no more than 4 days.⁶

In accordance with the Appropriateness Criteria2 and based on the best available evidence, imaging is recommended in all patients who remain febrile or have persistent leukocytosis after 72 hours of antibiotic therapy. In such cases, there should be high suspicion for a complication requiring treatment.

OPTIONS FOR IMAGING

Computed tomography

Computed tomography (CT) of the abdomen and pelvis with contrast is considered the study of choice in complicated acute pyelonephritis. CT can detect focal parenchymal abnormalities, emphysematous changes, and anatomic anomalies, and can also define the extent of disease. It can also detect perinephric fluid collections and abscesses that necessitate a change in management.^2,5

A retrospective study in 2017 found that contrast-enhanced CT done without the usual noncontrast and excretory phases had an accuracy of 90% to 92% for pyelonephritis and 96% to 99% for urolithiasis, suggesting that reduction in radiation exposure through use of only the contrast-enhanced phase of CT imaging may be reasonable.⁷

Magnetic resonance imaging

Magnetic resonance imaging (MRI) is increasingly acknowledged as effective in the evaluation of renal pathology, including the diagnosis of pyelonephritis; but it lacks the level of evidence that CT provides for detecting renal abscesses, calculi, and emphysematous pyelonephritis.^2,8,9

Though it is more costly and time-consuming than CT with contrast enhancement, MRI is nevertheless the imaging study of choice if iodinated contrast or ionizing radiation must be avoided.

MRI typically involves a precontrast phase and a gadolinium contrast-enhanced phase, though there are data to support diffusion-weighted MRI when exposure to gadolinium poses a risk to the patient, such as in pregnancy or renal impairment (particularly when the estimated GFR is < 30 mL/min/1.73 m²).¹⁰

Ultrasonography

Conventional ultrasonography is appealing due to its relatively low cost, its availability and portability, and the lack of radiation and contrast exposure. It is most helpful in detecting hydronephrosis and pyonephrosis rather than intrarenal or perinephric abscess.^2,9

Color and power Doppler ultrasonography may improve testing characteristics but not to the level of CT; in one study, sensitivity for detection of pyelonephritis was 33.3% with ultrasonography vs 81.0% with CT.¹¹

Recent studies of ultrasonography with contrast enhancement show promising results,² and it may ultimately prove to have a similar efficacy with lower risk for patients, but this has not been validated in large studies, and its availability remains limited.

Ultrasonography should be considered for patients in whom obstruction (with resulting hydronephrosis or pyonephrosis) is a primary concern, particularly when contrast exposure or radiation is contraindicated and MRI is unavailable.²

Abdominal radiography

While emphysematous pyelonephritis or a large staghorn calculus may be seen on abdominal radiography, it is not recommended for the assessment of complications in acute pyelonephritis because it lacks sensitivity.²

RETURN TO THE CASE SCENARIO

The patient in our case scenario meets the clinical criteria for uncomplicated pyelo­nephritis and is therefore not a candidate for imaging. Intravenous antibiotics should be started and should lead to rapid improvement in her condition.

Acknowledgment: The authors would like to thank Dr. Lisa Blacklock for her review of the radiology section of this paper.

A previously healthy 44-year-old woman presents to the emergency department with 1 day of fever, flank pain, dysuria, and persistent nausea and vomiting. Her temperature is 38.7°C (101.7°F), heart rate 102 beats per minute, and blood pressure 120/70 mm Hg. She has costovertebral angle tenderness. Laboratory testing reveals mild leukocytosis and a normal serum creatinine level; urinalysis shows leukocytes, as well as leukocyte esterase and nitrites. She has no personal or family history of nephrolithiasis. Urine cultures are obtained, and she is started on intravenous antibiotics and intravenous hydration to treat pyelonephritis.

Is imaging indicated at this point? And if so, which study is recommended?

KEY FEATURES

Acute pyelonephritis, infection of the renal parenchyma and collecting system, most often results from an ascending infection of the lower urinary tract. It is estimated to account for 250,000 office visits and 200,000 hospital admissions each year in the United States.¹

Lower urinary tract symptoms such as urinary frequency, urgency, and dysuria accompanied by fever, nausea, vomiting, and flank pain raise suspicion for acute pyelonephritis. Flank pain is a key, nearly universal feature of upper urinary tract infection in patients without diabetes, though it may be absent in up to 50% of patients with diabetes.²

Additional findings include costovertebral angle tenderness on physical examination and leukocytosis, pyuria, and bacteriuria on laboratory studies.

PREDICTING THE NEED FOR EARLY IMAGING

Though guidelines state that imaging is inappropriate in most patients with pyeloneph­ritis,^2–4 it is nevertheless often done for diagnosis or identification of complications, which have been reported in more than two-thirds of patients.^2–4

Acute pyelonephritis is generally classified as complicated or uncomplicated, though different definitions exist with regard to these classifications. The American College of Radiology’s Appropriateness Criteria2 consider patients with diabetes, immune compromise, a history of urolithiasis, or anatomic abnormality to be at highest risk for complications, and therefore recommend early imaging to assess for hydronephrosis, pyonephrosis, emphysematous pyelonephritis, and intrinsic or perinephric abscess.²

A clinical rule for predicting the need for imaging in acute pyelonephritis was developed and validated in an emergency department population in the Netherlands.³ The study suggested that restricting early imaging to patients with a history of urolithiasis, a urine pH of 7.0 or higher, or renal insufficiency—defined as a glomerular filtration rate (GFR) of 40 mL/min/1.73m² or lower as estimated by the Modification of Diet in Renal Disease formula—would provide a negative predictive value of 94% to 100% for detection of an urgent urologic disorder (pyonephrosis, renal abscess, or urolithiasis). This high negative predictive value highlights that an absence of these signs and symptoms can safely identify patients who do not need renal imaging.

The positive predictive value was less useful, as only 5% to 23% of patients who had at least 1 risk factor went on to have urgent urologic risk factors.³

Implementation of this prediction rule would have resulted in a relative reduction in imaging of 40% and an absolute reduction of 28%. Of note, use of reduced GFR in this prediction rule is not clearly validated for patients with chronic kidney disease, as the previous GFR for most patients in this study was unknown.³

Based on these data, initial imaging is recommended in patients with diabetes, immune compromise, a history of urolithiasis, anatomic abnormality, a urine pH 7.0 or higher, or a GFR 40 mL/min or lower in a patient with no history of significant renal dysfunction. Early imaging would also be reasonable in patients with a complex clinical presentation, early recurrence of symptoms after treatment, clinical decompensation, or critical illness.

In a retrospective review of 62 patients hospitalized for acute renal infection, Soulen et al5 found that the most reliable indicator of complicated acute pyelonephritis was the persistence of fever and leukocytosis at 72 hours. And another small prospective study of patients with uncomplicated pyelonephritis reported a time to defervescence of no more than 4 days.⁶

In accordance with the Appropriateness Criteria2 and based on the best available evidence, imaging is recommended in all patients who remain febrile or have persistent leukocytosis after 72 hours of antibiotic therapy. In such cases, there should be high suspicion for a complication requiring treatment.

OPTIONS FOR IMAGING

Computed tomography

Computed tomography (CT) of the abdomen and pelvis with contrast is considered the study of choice in complicated acute pyelonephritis. CT can detect focal parenchymal abnormalities, emphysematous changes, and anatomic anomalies, and can also define the extent of disease. It can also detect perinephric fluid collections and abscesses that necessitate a change in management.^2,5

A retrospective study in 2017 found that contrast-enhanced CT done without the usual noncontrast and excretory phases had an accuracy of 90% to 92% for pyelonephritis and 96% to 99% for urolithiasis, suggesting that reduction in radiation exposure through use of only the contrast-enhanced phase of CT imaging may be reasonable.⁷

Magnetic resonance imaging

Magnetic resonance imaging (MRI) is increasingly acknowledged as effective in the evaluation of renal pathology, including the diagnosis of pyelonephritis; but it lacks the level of evidence that CT provides for detecting renal abscesses, calculi, and emphysematous pyelonephritis.^2,8,9

Though it is more costly and time-consuming than CT with contrast enhancement, MRI is nevertheless the imaging study of choice if iodinated contrast or ionizing radiation must be avoided.

MRI typically involves a precontrast phase and a gadolinium contrast-enhanced phase, though there are data to support diffusion-weighted MRI when exposure to gadolinium poses a risk to the patient, such as in pregnancy or renal impairment (particularly when the estimated GFR is < 30 mL/min/1.73 m²).¹⁰

Ultrasonography

Conventional ultrasonography is appealing due to its relatively low cost, its availability and portability, and the lack of radiation and contrast exposure. It is most helpful in detecting hydronephrosis and pyonephrosis rather than intrarenal or perinephric abscess.^2,9

Color and power Doppler ultrasonography may improve testing characteristics but not to the level of CT; in one study, sensitivity for detection of pyelonephritis was 33.3% with ultrasonography vs 81.0% with CT.¹¹

Recent studies of ultrasonography with contrast enhancement show promising results,² and it may ultimately prove to have a similar efficacy with lower risk for patients, but this has not been validated in large studies, and its availability remains limited.

Ultrasonography should be considered for patients in whom obstruction (with resulting hydronephrosis or pyonephrosis) is a primary concern, particularly when contrast exposure or radiation is contraindicated and MRI is unavailable.²

Abdominal radiography

While emphysematous pyelonephritis or a large staghorn calculus may be seen on abdominal radiography, it is not recommended for the assessment of complications in acute pyelonephritis because it lacks sensitivity.²

RETURN TO THE CASE SCENARIO

The patient in our case scenario meets the clinical criteria for uncomplicated pyelo­nephritis and is therefore not a candidate for imaging. Intravenous antibiotics should be started and should lead to rapid improvement in her condition.

Acknowledgment: The authors would like to thank Dr. Lisa Blacklock for her review of the radiology section of this paper.

A previously healthy 44-year-old woman presents to the emergency department with 1 day of fever, flank pain, dysuria, and persistent nausea and vomiting. Her temperature is 38.7°C (101.7°F), heart rate 102 beats per minute, and blood pressure 120/70 mm Hg. She has costovertebral angle tenderness. Laboratory testing reveals mild leukocytosis and a normal serum creatinine level; urinalysis shows leukocytes, as well as leukocyte esterase and nitrites. She has no personal or family history of nephrolithiasis. Urine cultures are obtained, and she is started on intravenous antibiotics and intravenous hydration to treat pyelonephritis.

Is imaging indicated at this point? And if so, which study is recommended?

KEY FEATURES

Acute pyelonephritis, infection of the renal parenchyma and collecting system, most often results from an ascending infection of the lower urinary tract. It is estimated to account for 250,000 office visits and 200,000 hospital admissions each year in the United States.¹

Lower urinary tract symptoms such as urinary frequency, urgency, and dysuria accompanied by fever, nausea, vomiting, and flank pain raise suspicion for acute pyelonephritis. Flank pain is a key, nearly universal feature of upper urinary tract infection in patients without diabetes, though it may be absent in up to 50% of patients with diabetes.²

Additional findings include costovertebral angle tenderness on physical examination and leukocytosis, pyuria, and bacteriuria on laboratory studies.

PREDICTING THE NEED FOR EARLY IMAGING

Though guidelines state that imaging is inappropriate in most patients with pyeloneph­ritis,^2–4 it is nevertheless often done for diagnosis or identification of complications, which have been reported in more than two-thirds of patients.^2–4

Acute pyelonephritis is generally classified as complicated or uncomplicated, though different definitions exist with regard to these classifications. The American College of Radiology’s Appropriateness Criteria2 consider patients with diabetes, immune compromise, a history of urolithiasis, or anatomic abnormality to be at highest risk for complications, and therefore recommend early imaging to assess for hydronephrosis, pyonephrosis, emphysematous pyelonephritis, and intrinsic or perinephric abscess.²

A clinical rule for predicting the need for imaging in acute pyelonephritis was developed and validated in an emergency department population in the Netherlands.³ The study suggested that restricting early imaging to patients with a history of urolithiasis, a urine pH of 7.0 or higher, or renal insufficiency—defined as a glomerular filtration rate (GFR) of 40 mL/min/1.73m² or lower as estimated by the Modification of Diet in Renal Disease formula—would provide a negative predictive value of 94% to 100% for detection of an urgent urologic disorder (pyonephrosis, renal abscess, or urolithiasis). This high negative predictive value highlights that an absence of these signs and symptoms can safely identify patients who do not need renal imaging.

The positive predictive value was less useful, as only 5% to 23% of patients who had at least 1 risk factor went on to have urgent urologic risk factors.³

Implementation of this prediction rule would have resulted in a relative reduction in imaging of 40% and an absolute reduction of 28%. Of note, use of reduced GFR in this prediction rule is not clearly validated for patients with chronic kidney disease, as the previous GFR for most patients in this study was unknown.³

Based on these data, initial imaging is recommended in patients with diabetes, immune compromise, a history of urolithiasis, anatomic abnormality, a urine pH 7.0 or higher, or a GFR 40 mL/min or lower in a patient with no history of significant renal dysfunction. Early imaging would also be reasonable in patients with a complex clinical presentation, early recurrence of symptoms after treatment, clinical decompensation, or critical illness.

In a retrospective review of 62 patients hospitalized for acute renal infection, Soulen et al5 found that the most reliable indicator of complicated acute pyelonephritis was the persistence of fever and leukocytosis at 72 hours. And another small prospective study of patients with uncomplicated pyelonephritis reported a time to defervescence of no more than 4 days.⁶

In accordance with the Appropriateness Criteria2 and based on the best available evidence, imaging is recommended in all patients who remain febrile or have persistent leukocytosis after 72 hours of antibiotic therapy. In such cases, there should be high suspicion for a complication requiring treatment.

OPTIONS FOR IMAGING

Computed tomography

Computed tomography (CT) of the abdomen and pelvis with contrast is considered the study of choice in complicated acute pyelonephritis. CT can detect focal parenchymal abnormalities, emphysematous changes, and anatomic anomalies, and can also define the extent of disease. It can also detect perinephric fluid collections and abscesses that necessitate a change in management.^2,5

A retrospective study in 2017 found that contrast-enhanced CT done without the usual noncontrast and excretory phases had an accuracy of 90% to 92% for pyelonephritis and 96% to 99% for urolithiasis, suggesting that reduction in radiation exposure through use of only the contrast-enhanced phase of CT imaging may be reasonable.⁷

Magnetic resonance imaging

Magnetic resonance imaging (MRI) is increasingly acknowledged as effective in the evaluation of renal pathology, including the diagnosis of pyelonephritis; but it lacks the level of evidence that CT provides for detecting renal abscesses, calculi, and emphysematous pyelonephritis.^2,8,9

Though it is more costly and time-consuming than CT with contrast enhancement, MRI is nevertheless the imaging study of choice if iodinated contrast or ionizing radiation must be avoided.

MRI typically involves a precontrast phase and a gadolinium contrast-enhanced phase, though there are data to support diffusion-weighted MRI when exposure to gadolinium poses a risk to the patient, such as in pregnancy or renal impairment (particularly when the estimated GFR is < 30 mL/min/1.73 m²).¹⁰

Ultrasonography

Conventional ultrasonography is appealing due to its relatively low cost, its availability and portability, and the lack of radiation and contrast exposure. It is most helpful in detecting hydronephrosis and pyonephrosis rather than intrarenal or perinephric abscess.^2,9

Color and power Doppler ultrasonography may improve testing characteristics but not to the level of CT; in one study, sensitivity for detection of pyelonephritis was 33.3% with ultrasonography vs 81.0% with CT.¹¹

Recent studies of ultrasonography with contrast enhancement show promising results,² and it may ultimately prove to have a similar efficacy with lower risk for patients, but this has not been validated in large studies, and its availability remains limited.

Ultrasonography should be considered for patients in whom obstruction (with resulting hydronephrosis or pyonephrosis) is a primary concern, particularly when contrast exposure or radiation is contraindicated and MRI is unavailable.²

Abdominal radiography

While emphysematous pyelonephritis or a large staghorn calculus may be seen on abdominal radiography, it is not recommended for the assessment of complications in acute pyelonephritis because it lacks sensitivity.²

RETURN TO THE CASE SCENARIO

The patient in our case scenario meets the clinical criteria for uncomplicated pyelo­nephritis and is therefore not a candidate for imaging. Intravenous antibiotics should be started and should lead to rapid improvement in her condition.

Acknowledgment: The authors would like to thank Dr. Lisa Blacklock for her review of the radiology section of this paper.

Technology has infiltrated all parts of our everyday lives, including healthcare. Patients can make and cancel appointments, send e-mails directly to their physician, and request prescription refills—all through electronic portals. Physicians and healthcare providers must adjust to these changes in care-delivery models. Primary care providers must also adapt as younger generations seek access for their health needs outside of the doctor’s office.

See related editorial

And so it is with everyday life. Online banking and bill-paying is common. Groceries can be bought online and delivered within an hour. Connecting with family or friends around the world can be done with the touch of a button. In the United States, 90% of adults own a cell phone; many do not have a land line. More than 65% of adult Americans under age 75 own a smart phone, and 50% of the public owns a tablet computer.¹

DRIVERS OF CHANGE: THE MILLENNIALS

The development and use of new technology is driven by the coming of age of the youngest adult population, ie, “Generation Y” or millennials, ie, persons born between 1981 and 1996.² They now account for 28% of the US adult population, surpassing the baby boomers (born 1946 to 1964) by 8 million.³

Millennials have grown up with the World Wide Web at their fingertips. They are accustomed to an environment full of choices and unlimited, instantly available information.⁴

Millennials are cost-conscious shoppers who desire convenience and quick access. As patients, they often forgo traditional doctor’s office visits, turning instead to the Internet for quick answers to their questions in blogs and websites.⁵ A Kaiser Family Foundation survey in 2018 indicated that only a quarter of millennials see a primary care physician for healthcare needs.⁶

The shortage of primary care physicians

There are several reasons for this. Primary care physicians are in short supply, more Americans have insurance after the passage of the Affordable Care Act, and more physicians are working part-time or retiring earlier than in previous generations. There will be a continued shortfall of 15,000 to 49,000 full-time-equivalent primary care physicians by 2030.⁷ A survey of 15 large metropolitan markets found that the average wait time for a primary care new patient appointment increased to 24.1 days—a 30% increase from 2014. In some cities, the wait time can be 3 to 4 months.⁸

Older patients of the baby-boomer generation tend to discuss medical issues with their primary care physician, often relying on their feedback to improve their health lifestyle choices.⁹ Baby boomers who are Medicare subscribers tend to see their regular doctor at least once or twice a year¹⁰; trust is built with this continuity in care.

The rise of pharmacy clinics

But the shortage of primary care physicians and the desire of younger patients for immediate access to care have fueled the growth of new options for access, such as retail clinics in large pharmacies. These clinics are mostly found in the South and Midwest and are staffed by nurse practitioners,¹¹ and 90% of their billing falls under 10 common diagnoses, including urinary tract and upper respiratory infections. More than 40% of patients seeking care at retail pharmacy clinics are 18 to 44 years of age, and less than 25% of this group have a primary care provider.¹¹ These clinics have shorter wait times and limited out-of-pocket costs, and they are more convenient. In a study of adults visiting these clinics for vaccination, 30% did so during evening, weekend, and holiday hours, when traditional doctors’ offices are closed.¹²

Telemedicine’s foothold

Telemedicine has also taken a foothold in healthcare. Initially used for episodic illnesses, there is now growing acceptance of telemedicine for management of chronic physical and mental health problems. Accessibility to a doctor via a mobile device while at home has proven to be helpful to young, elderly, and minority patients living in rural areas,¹³ although reimbursement and legal issues continue to constrain its growth.¹⁴ Telemedicine is predicted to grow by nearly 15% from now to 2025, especially in North America and Europe, where technology has kept pace and government initiatives are encouraging its advancement.¹⁵

The American College of Physicians has published recommendations on how best to use telemedicine, especially when there is already an established patient-physician relationship. Telemedicine can bridge the divide for those who lack access to care because of geographic constraints or who cannot afford a regular doctor’s office appointment.¹⁶ It can also allow healthcare “extenders” like social workers, nutritionists, pharmacists, and nurses to work collaboratively with the primary care physician to improve patient education and outcomes.¹⁷

Wearable devices

The wearable device market continues to expand, in large part due to the increased availability and utilization of mobile technology. These gadgets can record steps, sleep, and heart rate. Consumer fitness trackers can give patients insight into their activity levels and encourage them to modify their behavior, ie, get up and move around more.¹⁷ The Deloitte Center for Health Solutions survey in 2018 showed that 62% of millennials use consumer fitness trackers to help meet their wellness goals, compared with 16% of seniors and 25% of baby boomers.¹⁸ There are few studies showing that these devices improve overall health promotion or decrease healthcare costs,^17,19 but research is ongoing.

And the “generation gap” in technology’s uptake is slowly closing: 81% of American adults own a smartphone, and the rate in people over age 50 increased from 53% in 2015 to 67% in 2018.²⁰ By comparison, 92% of millennials own a smartphone.¹

Smartphone apps

A 2015 survey of more than 1,600 US adults found that 58% had downloaded an application to their smartphone to track their health needs, with 41% using more than 5 health-related apps; the most commonly downloaded apps tracked physical activity, food intake, exercise programs or weight loss progress.²¹

Users of mobile health apps are generally younger and more highly educated than nonusers.²² However, baby boomers are willing to try mobile health apps if the apps are intuitive, accessible, and effective; this is important, especially since this group accounts for more than 20% of US healthcare expenditures.²³ Engaging and empowering baby boomers to use this technology may allow them to remain independent, live healthier, and avoid unnecessary office visits, thus decreasing strain on the limited healthcare workforce.²³

Physicians and physician educators should be aware of this generational shift. Millennial-aged doctors will continue to embrace technology to achieve their work-life balance in order to avoid burnout and maintain robust primary care practices whether in the office or outside of it.

Medical school curricula

Medical schools need to adjust their curricula to prepare the next generation of physicians to engage with these new healthcare delivery models and technology. Practicing telemedicine, assessing mobile app safety and utility, and effectively integrating data from patient-specific devices represent a new skill set that is considerably different from the typical face-to-face encounters learners experience today.

Recognizing this, more than 50% of medical schools have added telemedicine and digital health to the curriculum,²⁴ with suggestions to include telemedicine-related content in the Accreditation Council for Graduate Medical Education core competencies.²⁵

Improving the electronic medical record

Maximizing the efficiency of electronic medical records will also be important because physicians currently spend more than 50% of their workday on documentation and administrative tasks; for every 1 hour of patient contact, physicians spend 2 hours in front of the electronic medical record.²⁶ End-users (doctors, nurses, pharmacists, scribes) should interact or engage with developers of electronic medical record systems to promote platforms that enhance workflow, increase connectivity to mobile apps, foster team collaboration, and provide consistency in patient safety and privacy.²⁷

Early and continuous education on use of the electronic medical record should be routine, as proficiency improves work-life balance, physician job satisfaction, and patient care by reducing after-hours note completion and in-box tasks leading to burnout.²⁸

Technology-enabled primary care

Technology-enabled healthcare is here to stay and will continue to evolve, incorporating telehealth, smartphones, mobile apps, in-home and wearable devices, and online video communication.¹⁷ Clinicians will need to be adept at working with these technologies to advance quality care in population health. It will require clinician training and professional development, advances in technology, and revised reimbursement policies.¹⁷ But despite the increased use of mobile apps, there remain concerns about the possible dangers associated with their use, including breaches in confidentiality, conflicts of interest, and lack of professional medical involvement and evidence in their design.²⁹

THE IMPORTANCE OF BEING SAVVY

There is a growing need for primary care providers to be technologically savvy and readily accessible via e-mail, healthcare portals, or in the office to keep up with the generational shifts and expectations occurring in this decade. Healthcare systems should have the right infrastructure in place, including efficient Web platforms to support telemedicine or to synchronize digital tracking devices, as well as a trained workforce to understand and implement these revolutionary changes into everyday practice. Educators will need to provide training in these changing platforms to medical students and residents. Primary care will evolve to redefine its role within the context of these emerging technologies¹⁷ and to adjust to these market demands in order to stay relevant.

Technology has infiltrated all parts of our everyday lives, including healthcare. Patients can make and cancel appointments, send e-mails directly to their physician, and request prescription refills—all through electronic portals. Physicians and healthcare providers must adjust to these changes in care-delivery models. Primary care providers must also adapt as younger generations seek access for their health needs outside of the doctor’s office.

See related editorial

And so it is with everyday life. Online banking and bill-paying is common. Groceries can be bought online and delivered within an hour. Connecting with family or friends around the world can be done with the touch of a button. In the United States, 90% of adults own a cell phone; many do not have a land line. More than 65% of adult Americans under age 75 own a smart phone, and 50% of the public owns a tablet computer.¹

DRIVERS OF CHANGE: THE MILLENNIALS

The development and use of new technology is driven by the coming of age of the youngest adult population, ie, “Generation Y” or millennials, ie, persons born between 1981 and 1996.² They now account for 28% of the US adult population, surpassing the baby boomers (born 1946 to 1964) by 8 million.³

Millennials have grown up with the World Wide Web at their fingertips. They are accustomed to an environment full of choices and unlimited, instantly available information.⁴

Millennials are cost-conscious shoppers who desire convenience and quick access. As patients, they often forgo traditional doctor’s office visits, turning instead to the Internet for quick answers to their questions in blogs and websites.⁵ A Kaiser Family Foundation survey in 2018 indicated that only a quarter of millennials see a primary care physician for healthcare needs.⁶

The shortage of primary care physicians

There are several reasons for this. Primary care physicians are in short supply, more Americans have insurance after the passage of the Affordable Care Act, and more physicians are working part-time or retiring earlier than in previous generations. There will be a continued shortfall of 15,000 to 49,000 full-time-equivalent primary care physicians by 2030.⁷ A survey of 15 large metropolitan markets found that the average wait time for a primary care new patient appointment increased to 24.1 days—a 30% increase from 2014. In some cities, the wait time can be 3 to 4 months.⁸

Older patients of the baby-boomer generation tend to discuss medical issues with their primary care physician, often relying on their feedback to improve their health lifestyle choices.⁹ Baby boomers who are Medicare subscribers tend to see their regular doctor at least once or twice a year¹⁰; trust is built with this continuity in care.

The rise of pharmacy clinics

But the shortage of primary care physicians and the desire of younger patients for immediate access to care have fueled the growth of new options for access, such as retail clinics in large pharmacies. These clinics are mostly found in the South and Midwest and are staffed by nurse practitioners,¹¹ and 90% of their billing falls under 10 common diagnoses, including urinary tract and upper respiratory infections. More than 40% of patients seeking care at retail pharmacy clinics are 18 to 44 years of age, and less than 25% of this group have a primary care provider.¹¹ These clinics have shorter wait times and limited out-of-pocket costs, and they are more convenient. In a study of adults visiting these clinics for vaccination, 30% did so during evening, weekend, and holiday hours, when traditional doctors’ offices are closed.¹²

Telemedicine’s foothold

Telemedicine has also taken a foothold in healthcare. Initially used for episodic illnesses, there is now growing acceptance of telemedicine for management of chronic physical and mental health problems. Accessibility to a doctor via a mobile device while at home has proven to be helpful to young, elderly, and minority patients living in rural areas,¹³ although reimbursement and legal issues continue to constrain its growth.¹⁴ Telemedicine is predicted to grow by nearly 15% from now to 2025, especially in North America and Europe, where technology has kept pace and government initiatives are encouraging its advancement.¹⁵

The American College of Physicians has published recommendations on how best to use telemedicine, especially when there is already an established patient-physician relationship. Telemedicine can bridge the divide for those who lack access to care because of geographic constraints or who cannot afford a regular doctor’s office appointment.¹⁶ It can also allow healthcare “extenders” like social workers, nutritionists, pharmacists, and nurses to work collaboratively with the primary care physician to improve patient education and outcomes.¹⁷

Wearable devices

The wearable device market continues to expand, in large part due to the increased availability and utilization of mobile technology. These gadgets can record steps, sleep, and heart rate. Consumer fitness trackers can give patients insight into their activity levels and encourage them to modify their behavior, ie, get up and move around more.¹⁷ The Deloitte Center for Health Solutions survey in 2018 showed that 62% of millennials use consumer fitness trackers to help meet their wellness goals, compared with 16% of seniors and 25% of baby boomers.¹⁸ There are few studies showing that these devices improve overall health promotion or decrease healthcare costs,^17,19 but research is ongoing.

And the “generation gap” in technology’s uptake is slowly closing: 81% of American adults own a smartphone, and the rate in people over age 50 increased from 53% in 2015 to 67% in 2018.²⁰ By comparison, 92% of millennials own a smartphone.¹

Smartphone apps

A 2015 survey of more than 1,600 US adults found that 58% had downloaded an application to their smartphone to track their health needs, with 41% using more than 5 health-related apps; the most commonly downloaded apps tracked physical activity, food intake, exercise programs or weight loss progress.²¹

Users of mobile health apps are generally younger and more highly educated than nonusers.²² However, baby boomers are willing to try mobile health apps if the apps are intuitive, accessible, and effective; this is important, especially since this group accounts for more than 20% of US healthcare expenditures.²³ Engaging and empowering baby boomers to use this technology may allow them to remain independent, live healthier, and avoid unnecessary office visits, thus decreasing strain on the limited healthcare workforce.²³

Physicians and physician educators should be aware of this generational shift. Millennial-aged doctors will continue to embrace technology to achieve their work-life balance in order to avoid burnout and maintain robust primary care practices whether in the office or outside of it.

Medical school curricula

Medical schools need to adjust their curricula to prepare the next generation of physicians to engage with these new healthcare delivery models and technology. Practicing telemedicine, assessing mobile app safety and utility, and effectively integrating data from patient-specific devices represent a new skill set that is considerably different from the typical face-to-face encounters learners experience today.

Recognizing this, more than 50% of medical schools have added telemedicine and digital health to the curriculum,²⁴ with suggestions to include telemedicine-related content in the Accreditation Council for Graduate Medical Education core competencies.²⁵

Improving the electronic medical record

Maximizing the efficiency of electronic medical records will also be important because physicians currently spend more than 50% of their workday on documentation and administrative tasks; for every 1 hour of patient contact, physicians spend 2 hours in front of the electronic medical record.²⁶ End-users (doctors, nurses, pharmacists, scribes) should interact or engage with developers of electronic medical record systems to promote platforms that enhance workflow, increase connectivity to mobile apps, foster team collaboration, and provide consistency in patient safety and privacy.²⁷

Early and continuous education on use of the electronic medical record should be routine, as proficiency improves work-life balance, physician job satisfaction, and patient care by reducing after-hours note completion and in-box tasks leading to burnout.²⁸

Technology-enabled primary care

Technology-enabled healthcare is here to stay and will continue to evolve, incorporating telehealth, smartphones, mobile apps, in-home and wearable devices, and online video communication.¹⁷ Clinicians will need to be adept at working with these technologies to advance quality care in population health. It will require clinician training and professional development, advances in technology, and revised reimbursement policies.¹⁷ But despite the increased use of mobile apps, there remain concerns about the possible dangers associated with their use, including breaches in confidentiality, conflicts of interest, and lack of professional medical involvement and evidence in their design.²⁹

THE IMPORTANCE OF BEING SAVVY

There is a growing need for primary care providers to be technologically savvy and readily accessible via e-mail, healthcare portals, or in the office to keep up with the generational shifts and expectations occurring in this decade. Healthcare systems should have the right infrastructure in place, including efficient Web platforms to support telemedicine or to synchronize digital tracking devices, as well as a trained workforce to understand and implement these revolutionary changes into everyday practice. Educators will need to provide training in these changing platforms to medical students and residents. Primary care will evolve to redefine its role within the context of these emerging technologies¹⁷ and to adjust to these market demands in order to stay relevant.

Technology has infiltrated all parts of our everyday lives, including healthcare. Patients can make and cancel appointments, send e-mails directly to their physician, and request prescription refills—all through electronic portals. Physicians and healthcare providers must adjust to these changes in care-delivery models. Primary care providers must also adapt as younger generations seek access for their health needs outside of the doctor’s office.

See related editorial

And so it is with everyday life. Online banking and bill-paying is common. Groceries can be bought online and delivered within an hour. Connecting with family or friends around the world can be done with the touch of a button. In the United States, 90% of adults own a cell phone; many do not have a land line. More than 65% of adult Americans under age 75 own a smart phone, and 50% of the public owns a tablet computer.¹

DRIVERS OF CHANGE: THE MILLENNIALS

The development and use of new technology is driven by the coming of age of the youngest adult population, ie, “Generation Y” or millennials, ie, persons born between 1981 and 1996.² They now account for 28% of the US adult population, surpassing the baby boomers (born 1946 to 1964) by 8 million.³

Millennials have grown up with the World Wide Web at their fingertips. They are accustomed to an environment full of choices and unlimited, instantly available information.⁴

Millennials are cost-conscious shoppers who desire convenience and quick access. As patients, they often forgo traditional doctor’s office visits, turning instead to the Internet for quick answers to their questions in blogs and websites.⁵ A Kaiser Family Foundation survey in 2018 indicated that only a quarter of millennials see a primary care physician for healthcare needs.⁶

The shortage of primary care physicians

There are several reasons for this. Primary care physicians are in short supply, more Americans have insurance after the passage of the Affordable Care Act, and more physicians are working part-time or retiring earlier than in previous generations. There will be a continued shortfall of 15,000 to 49,000 full-time-equivalent primary care physicians by 2030.⁷ A survey of 15 large metropolitan markets found that the average wait time for a primary care new patient appointment increased to 24.1 days—a 30% increase from 2014. In some cities, the wait time can be 3 to 4 months.⁸

Older patients of the baby-boomer generation tend to discuss medical issues with their primary care physician, often relying on their feedback to improve their health lifestyle choices.⁹ Baby boomers who are Medicare subscribers tend to see their regular doctor at least once or twice a year¹⁰; trust is built with this continuity in care.

The rise of pharmacy clinics

But the shortage of primary care physicians and the desire of younger patients for immediate access to care have fueled the growth of new options for access, such as retail clinics in large pharmacies. These clinics are mostly found in the South and Midwest and are staffed by nurse practitioners,¹¹ and 90% of their billing falls under 10 common diagnoses, including urinary tract and upper respiratory infections. More than 40% of patients seeking care at retail pharmacy clinics are 18 to 44 years of age, and less than 25% of this group have a primary care provider.¹¹ These clinics have shorter wait times and limited out-of-pocket costs, and they are more convenient. In a study of adults visiting these clinics for vaccination, 30% did so during evening, weekend, and holiday hours, when traditional doctors’ offices are closed.¹²

Telemedicine’s foothold

Telemedicine has also taken a foothold in healthcare. Initially used for episodic illnesses, there is now growing acceptance of telemedicine for management of chronic physical and mental health problems. Accessibility to a doctor via a mobile device while at home has proven to be helpful to young, elderly, and minority patients living in rural areas,¹³ although reimbursement and legal issues continue to constrain its growth.¹⁴ Telemedicine is predicted to grow by nearly 15% from now to 2025, especially in North America and Europe, where technology has kept pace and government initiatives are encouraging its advancement.¹⁵

The American College of Physicians has published recommendations on how best to use telemedicine, especially when there is already an established patient-physician relationship. Telemedicine can bridge the divide for those who lack access to care because of geographic constraints or who cannot afford a regular doctor’s office appointment.¹⁶ It can also allow healthcare “extenders” like social workers, nutritionists, pharmacists, and nurses to work collaboratively with the primary care physician to improve patient education and outcomes.¹⁷

Wearable devices

The wearable device market continues to expand, in large part due to the increased availability and utilization of mobile technology. These gadgets can record steps, sleep, and heart rate. Consumer fitness trackers can give patients insight into their activity levels and encourage them to modify their behavior, ie, get up and move around more.¹⁷ The Deloitte Center for Health Solutions survey in 2018 showed that 62% of millennials use consumer fitness trackers to help meet their wellness goals, compared with 16% of seniors and 25% of baby boomers.¹⁸ There are few studies showing that these devices improve overall health promotion or decrease healthcare costs,^17,19 but research is ongoing.

And the “generation gap” in technology’s uptake is slowly closing: 81% of American adults own a smartphone, and the rate in people over age 50 increased from 53% in 2015 to 67% in 2018.²⁰ By comparison, 92% of millennials own a smartphone.¹

Smartphone apps

A 2015 survey of more than 1,600 US adults found that 58% had downloaded an application to their smartphone to track their health needs, with 41% using more than 5 health-related apps; the most commonly downloaded apps tracked physical activity, food intake, exercise programs or weight loss progress.²¹

Users of mobile health apps are generally younger and more highly educated than nonusers.²² However, baby boomers are willing to try mobile health apps if the apps are intuitive, accessible, and effective; this is important, especially since this group accounts for more than 20% of US healthcare expenditures.²³ Engaging and empowering baby boomers to use this technology may allow them to remain independent, live healthier, and avoid unnecessary office visits, thus decreasing strain on the limited healthcare workforce.²³

Physicians and physician educators should be aware of this generational shift. Millennial-aged doctors will continue to embrace technology to achieve their work-life balance in order to avoid burnout and maintain robust primary care practices whether in the office or outside of it.

Medical school curricula

Medical schools need to adjust their curricula to prepare the next generation of physicians to engage with these new healthcare delivery models and technology. Practicing telemedicine, assessing mobile app safety and utility, and effectively integrating data from patient-specific devices represent a new skill set that is considerably different from the typical face-to-face encounters learners experience today.

Recognizing this, more than 50% of medical schools have added telemedicine and digital health to the curriculum,²⁴ with suggestions to include telemedicine-related content in the Accreditation Council for Graduate Medical Education core competencies.²⁵

Improving the electronic medical record

Maximizing the efficiency of electronic medical records will also be important because physicians currently spend more than 50% of their workday on documentation and administrative tasks; for every 1 hour of patient contact, physicians spend 2 hours in front of the electronic medical record.²⁶ End-users (doctors, nurses, pharmacists, scribes) should interact or engage with developers of electronic medical record systems to promote platforms that enhance workflow, increase connectivity to mobile apps, foster team collaboration, and provide consistency in patient safety and privacy.²⁷

Early and continuous education on use of the electronic medical record should be routine, as proficiency improves work-life balance, physician job satisfaction, and patient care by reducing after-hours note completion and in-box tasks leading to burnout.²⁸

Technology-enabled primary care

Technology-enabled healthcare is here to stay and will continue to evolve, incorporating telehealth, smartphones, mobile apps, in-home and wearable devices, and online video communication.¹⁷ Clinicians will need to be adept at working with these technologies to advance quality care in population health. It will require clinician training and professional development, advances in technology, and revised reimbursement policies.¹⁷ But despite the increased use of mobile apps, there remain concerns about the possible dangers associated with their use, including breaches in confidentiality, conflicts of interest, and lack of professional medical involvement and evidence in their design.²⁹

THE IMPORTANCE OF BEING SAVVY

There is a growing need for primary care providers to be technologically savvy and readily accessible via e-mail, healthcare portals, or in the office to keep up with the generational shifts and expectations occurring in this decade. Healthcare systems should have the right infrastructure in place, including efficient Web platforms to support telemedicine or to synchronize digital tracking devices, as well as a trained workforce to understand and implement these revolutionary changes into everyday practice. Educators will need to provide training in these changing platforms to medical students and residents. Primary care will evolve to redefine its role within the context of these emerging technologies¹⁷ and to adjust to these market demands in order to stay relevant.

Aspirin (acetylsalicylic acid) and its progenitors are valuable medications with a history spanning at least 4 millennia. An enormous number of patients take aspirin for a variety of reasons, and managing their therapy around the time of surgery can be challenging, as Drs. Prabhakaran and Whinney discuss in this issue.¹ Even after 4,000 years, we are still learning about these remarkable drugs.

See related article

LEARNING WHAT SALICYLATES ARE

LEARNING (AND IGNORING) WHAT ASPIRIN CAN DO

In the 1940s, a general practitioner in California named Lawrence Craven recognized that many of his post-tonsillectomy patients had to be hospitalized for bleeding after he started recommending they use aspirin-containing chewing gum for pain relief.⁴ Under the then-debated hypothesis that myocardial infarction (MI) involves thrombosis, he recommended that adult men should take aspirin daily. He believed that women had lower rates of MI because they were more likely to take aspirin, something that men did not view as a “masculine” thing to do.

In a series of letters in journals such as the Mississippi Valley Medical Journal,⁵ Craven reported his observations of very low rates of MI and no strokes in aspirin users. Given the nonrigorous nature of his research and the obscure journals in which he published, his findings languished for many years. Ironically, he died of an MI in 1957.

LEARNING HOW ASPIRIN WORKS (AND A FEW OTHER THINGS)

The history of aspirin research illustrates how the fields of hemostasis and inflammation are now linked.

In the late 1960s, Weiss et al⁶ reported that aspirin rapidly and irreversibly inhibits platelet aggregation. In parallel, using biological assays in work that eventually led to the Nobel Prize, Vane⁷ discovered that inflammation involves the de novo synthesis of prostaglandins and that aspirin directly inhibits this synthesis. Further work connecting these lines of investigation led us to understand that platelet aggregation is enhanced by the prostaglandin derivative thromboxane A2, produced by cyclooxygenase-1, and that aspirin irreversibly inhibits this enzyme by acetylation.

LEARNING WHEN TO USE ASPIRIN

After decades of research ranging from the Physicians’ Health Study to well-named trials such as ARRIVE, ASCEND, and ASPREE, we now know that taking daily low doses of aspirin for primary prevention can reduce the risk of cardiovascular events and may reduce the risk of colorectal cancer—but at the cost of an increased risk of bleeding.⁸

Which patients will gain the most benefit and incur the least risk is still debated. What is certain, however, is that aspirin has an important role in acute coronary syndromes, secondary prevention of MI and stroke, and prevention of thrombosis after coronary stent placement. In the perioperative setting, we are learning that aspirin may benefit patients with myocardial injury after noncardiac surgery, a recently described clinical entity associated with surprisingly high mortality rates.^9,10

The perioperative period is a dangerous time—surgical stress, hypercoagulability, inflammation, pain, and hemodynamic changes predispose to plaque rupture and supply-demand imbalance. It is therefore logical to hope aspirin would provide protection for at-risk patients in this context.

Unfortunately, results from the second Perioperative Ischemic Evaluation trial have dampened enthusiasm.¹¹ Aspirin has now joined clonidine and beta-blockers on the list of interventions that probably do not reduce perioperative cardiovascular mortality rates. Other than protecting against stent thrombosis, aspirin’s main perioperative effect is to increase bleeding. Consequently, some surgical procedures mandate withdrawal of aspirin.

WHAT WE STILL NEED TO LEARN

Over the years, we have learned the broad outlines of using aspirin to prevent and treat cardiovascular disease, to relieve pain and inflammation (its original purpose), and to prevent stent thrombosis. 

However, many details remain to be filled in. We need to better define groups who should and should not take aspirin for primary prevention. We also need to understand aspirin’s role in cancer chemoprevention, to find better ways to mitigate its undesirable effects, and to study its role in treating myocardial injury after noncardiac surgery.

Finally, we need to determine which (if any) patients without coronary stents will benefit from continuing their aspirin perioperatively or even initiating aspirin therapy preoperatively.

Will humanity still be using salicylates 4,000 years from now? Probably not. But what we have learned and will continue to learn from this remarkable group of medications will certainly inform new and better therapies in the years to come.

Aspirin (acetylsalicylic acid) and its progenitors are valuable medications with a history spanning at least 4 millennia. An enormous number of patients take aspirin for a variety of reasons, and managing their therapy around the time of surgery can be challenging, as Drs. Prabhakaran and Whinney discuss in this issue.¹ Even after 4,000 years, we are still learning about these remarkable drugs.

See related article

LEARNING WHAT SALICYLATES ARE

LEARNING (AND IGNORING) WHAT ASPIRIN CAN DO

In the 1940s, a general practitioner in California named Lawrence Craven recognized that many of his post-tonsillectomy patients had to be hospitalized for bleeding after he started recommending they use aspirin-containing chewing gum for pain relief.⁴ Under the then-debated hypothesis that myocardial infarction (MI) involves thrombosis, he recommended that adult men should take aspirin daily. He believed that women had lower rates of MI because they were more likely to take aspirin, something that men did not view as a “masculine” thing to do.

In a series of letters in journals such as the Mississippi Valley Medical Journal,⁵ Craven reported his observations of very low rates of MI and no strokes in aspirin users. Given the nonrigorous nature of his research and the obscure journals in which he published, his findings languished for many years. Ironically, he died of an MI in 1957.

LEARNING HOW ASPIRIN WORKS (AND A FEW OTHER THINGS)

The history of aspirin research illustrates how the fields of hemostasis and inflammation are now linked.

In the late 1960s, Weiss et al⁶ reported that aspirin rapidly and irreversibly inhibits platelet aggregation. In parallel, using biological assays in work that eventually led to the Nobel Prize, Vane⁷ discovered that inflammation involves the de novo synthesis of prostaglandins and that aspirin directly inhibits this synthesis. Further work connecting these lines of investigation led us to understand that platelet aggregation is enhanced by the prostaglandin derivative thromboxane A2, produced by cyclooxygenase-1, and that aspirin irreversibly inhibits this enzyme by acetylation.

LEARNING WHEN TO USE ASPIRIN

After decades of research ranging from the Physicians’ Health Study to well-named trials such as ARRIVE, ASCEND, and ASPREE, we now know that taking daily low doses of aspirin for primary prevention can reduce the risk of cardiovascular events and may reduce the risk of colorectal cancer—but at the cost of an increased risk of bleeding.⁸

Which patients will gain the most benefit and incur the least risk is still debated. What is certain, however, is that aspirin has an important role in acute coronary syndromes, secondary prevention of MI and stroke, and prevention of thrombosis after coronary stent placement. In the perioperative setting, we are learning that aspirin may benefit patients with myocardial injury after noncardiac surgery, a recently described clinical entity associated with surprisingly high mortality rates.^9,10

The perioperative period is a dangerous time—surgical stress, hypercoagulability, inflammation, pain, and hemodynamic changes predispose to plaque rupture and supply-demand imbalance. It is therefore logical to hope aspirin would provide protection for at-risk patients in this context.

Unfortunately, results from the second Perioperative Ischemic Evaluation trial have dampened enthusiasm.¹¹ Aspirin has now joined clonidine and beta-blockers on the list of interventions that probably do not reduce perioperative cardiovascular mortality rates. Other than protecting against stent thrombosis, aspirin’s main perioperative effect is to increase bleeding. Consequently, some surgical procedures mandate withdrawal of aspirin.

WHAT WE STILL NEED TO LEARN

Over the years, we have learned the broad outlines of using aspirin to prevent and treat cardiovascular disease, to relieve pain and inflammation (its original purpose), and to prevent stent thrombosis. 

However, many details remain to be filled in. We need to better define groups who should and should not take aspirin for primary prevention. We also need to understand aspirin’s role in cancer chemoprevention, to find better ways to mitigate its undesirable effects, and to study its role in treating myocardial injury after noncardiac surgery.

Finally, we need to determine which (if any) patients without coronary stents will benefit from continuing their aspirin perioperatively or even initiating aspirin therapy preoperatively.

Will humanity still be using salicylates 4,000 years from now? Probably not. But what we have learned and will continue to learn from this remarkable group of medications will certainly inform new and better therapies in the years to come.

Aspirin (acetylsalicylic acid) and its progenitors are valuable medications with a history spanning at least 4 millennia. An enormous number of patients take aspirin for a variety of reasons, and managing their therapy around the time of surgery can be challenging, as Drs. Prabhakaran and Whinney discuss in this issue.¹ Even after 4,000 years, we are still learning about these remarkable drugs.

See related article

LEARNING WHAT SALICYLATES ARE

LEARNING (AND IGNORING) WHAT ASPIRIN CAN DO

In the 1940s, a general practitioner in California named Lawrence Craven recognized that many of his post-tonsillectomy patients had to be hospitalized for bleeding after he started recommending they use aspirin-containing chewing gum for pain relief.⁴ Under the then-debated hypothesis that myocardial infarction (MI) involves thrombosis, he recommended that adult men should take aspirin daily. He believed that women had lower rates of MI because they were more likely to take aspirin, something that men did not view as a “masculine” thing to do.

In a series of letters in journals such as the Mississippi Valley Medical Journal,⁵ Craven reported his observations of very low rates of MI and no strokes in aspirin users. Given the nonrigorous nature of his research and the obscure journals in which he published, his findings languished for many years. Ironically, he died of an MI in 1957.

LEARNING HOW ASPIRIN WORKS (AND A FEW OTHER THINGS)

The history of aspirin research illustrates how the fields of hemostasis and inflammation are now linked.

In the late 1960s, Weiss et al⁶ reported that aspirin rapidly and irreversibly inhibits platelet aggregation. In parallel, using biological assays in work that eventually led to the Nobel Prize, Vane⁷ discovered that inflammation involves the de novo synthesis of prostaglandins and that aspirin directly inhibits this synthesis. Further work connecting these lines of investigation led us to understand that platelet aggregation is enhanced by the prostaglandin derivative thromboxane A2, produced by cyclooxygenase-1, and that aspirin irreversibly inhibits this enzyme by acetylation.

LEARNING WHEN TO USE ASPIRIN

After decades of research ranging from the Physicians’ Health Study to well-named trials such as ARRIVE, ASCEND, and ASPREE, we now know that taking daily low doses of aspirin for primary prevention can reduce the risk of cardiovascular events and may reduce the risk of colorectal cancer—but at the cost of an increased risk of bleeding.⁸

Which patients will gain the most benefit and incur the least risk is still debated. What is certain, however, is that aspirin has an important role in acute coronary syndromes, secondary prevention of MI and stroke, and prevention of thrombosis after coronary stent placement. In the perioperative setting, we are learning that aspirin may benefit patients with myocardial injury after noncardiac surgery, a recently described clinical entity associated with surprisingly high mortality rates.^9,10

The perioperative period is a dangerous time—surgical stress, hypercoagulability, inflammation, pain, and hemodynamic changes predispose to plaque rupture and supply-demand imbalance. It is therefore logical to hope aspirin would provide protection for at-risk patients in this context.

Unfortunately, results from the second Perioperative Ischemic Evaluation trial have dampened enthusiasm.¹¹ Aspirin has now joined clonidine and beta-blockers on the list of interventions that probably do not reduce perioperative cardiovascular mortality rates. Other than protecting against stent thrombosis, aspirin’s main perioperative effect is to increase bleeding. Consequently, some surgical procedures mandate withdrawal of aspirin.

WHAT WE STILL NEED TO LEARN

Over the years, we have learned the broad outlines of using aspirin to prevent and treat cardiovascular disease, to relieve pain and inflammation (its original purpose), and to prevent stent thrombosis. 

However, many details remain to be filled in. We need to better define groups who should and should not take aspirin for primary prevention. We also need to understand aspirin’s role in cancer chemoprevention, to find better ways to mitigate its undesirable effects, and to study its role in treating myocardial injury after noncardiac surgery.

Finally, we need to determine which (if any) patients without coronary stents will benefit from continuing their aspirin perioperatively or even initiating aspirin therapy preoperatively.

Will humanity still be using salicylates 4,000 years from now? Probably not. But what we have learned and will continue to learn from this remarkable group of medications will certainly inform new and better therapies in the years to come.

“My dear, here we must run as fast as we can, just to stay in place. And if you wish to go anywhere you must run twice as fast as that.”
—Lewis Carroll
Alice’s Adventures in Wonderland

The future of primary care internal medicine physicians is uncertain. According to a 2018 survey of internal medicine residents conducted by the American College of Physicians, only 11% were considering primary care as a career path.¹ In 1998, that number was 54%.²

See related commentary

Possible reasons are many:

• Lower pay compared with subspecialists in a pay system that rewards procedural competency over mental effort
• Work schedules less flexible than in other specialties (eg, hospital medicine practitioners may have 1 week on and 1 week off)
• Perceived lack of respect
• Increasing regulatory and record-keeping burdens
• Tyranny of 15- to 20-minute appointments (irrespective of patient complexity)
• Scope-of-practice concerns as other providers seek primary care equivalency status (eg, pharmacists, walk-in clinics, advanced practice providers, telemedicine providers).

The result is a projected shortage of primary care physicians of 21,100 to 55,200 by 2030, according to a 2019 report by the Association of American Medical Colleges,³ despite an expected growth in advanced practice providers in primary care such as nurse practitioners and physician assistants.

A practical result of this shortage will be even less patient access to primary care physicians. A 2017 national survey found that the average wait time for a new patient-physician appointment has already increased by 30% since 2014.⁴ The wait time to see a primary care physician varied between 29 days in major metropolitan areas (up 50% from 2014) and 56 days in mid-sized markets. The longest waits by market size were 109 days for new patients in Boston, MA, and 122 days for those living in Albany, NY.

What are the implications?

In this issue, Pravia and Diaz⁵ make the case that primary care providers must adapt their practices to meet the needs of younger generations by increasing their use of technology. We agree that telemedicine, wearable medical devices, and enhanced patient communication through the electronic medical record (EMR) are here to stay and should be embraced.

However, we have seen the challenges of adopting technologic advances without first making an adjustment to the volume-driven patient schedule. For such advances to be successfully integrated into a clinical practice, it is vital to be cognizant of the current challenges encountered in primary care internal medicine.

UNIQUE BURDENS ON PRIMARY CARE

In addition to the stress of addressing multiple complex medical problems within a short time, evaluating multiple medical problems often leads to increases in results to review, forms to complete, and calls to patients. Even treatment plans initiated by specialists are often deferred to primary care providers for dosing adjustments, follow-up laboratory testing, and monitoring.

Moreover, patients often seek a second opinion from their primary care provider regarding care provided by subspecialists, as they consider their primary care provider to be the doctor who knows them best. And though it can be personally gratifying to be considered a trusted partner in the patient’s care, these requests often result in additional phone calls to the office or another thing to address within a complex visit.

A large in-box can be daunting in the setting of increased EMR demands. Whether you have difficulty putting in basic orders or are an advanced user, each upgrade can make you feel like you’re using the EMR for the first time. This is a problem for all specialties, but in primary care, one is addressing a large spectrum of concerns, so there is less opportunity to use standardized templates that can help buffer the problem.

A study of primary care providers found that nearly 75% of each patient visit was spent on activities other than face-to-face patient care, including working with the EMR.⁶ Similarly, a study using in-office observations and after-hours diaries found that physicians from various specialties spend 2 hours on administrative duties for each hour that they see patients in the office, followed by an additional 1 to 2 hours of work after clinic, mostly devoted to the EMR.⁷

Clinicians using scribes to help with record-keeping duties often need to see more patients to compensate for the cost. Adding 2 to 3 patients to a daily schedule usually means adding more medical conditions to manage, with an exponential increase in testing and in-box burden.

The additional burden this coverage creates in primary care is often not well understood by those in other specialties.

Another burden unique to primary care providers is the nearly continuous publication of guidelines that are often confusing and discrepant. Because many high-impact guidelines represent expert consensus or evidence from specialist perspectives, they may not fit the primary care model or values: eg, primary care guidelines tend to place more emphasis on harms associated with screening.

Screening for breast and prostate cancers is a prime example. Both require shared decision-making based on patient preferences and values.^8,9 Detailed discussions about preventive screening can be difficult to achieve within the context of a medical visit owing to time limitations, especially if other medical conditions being addressed are equally controversial, such as blood pressure target goals. A decade ago, one could easily declare, “It’s time for your annual PSA test,” and move on to other concerns. Given the changing evidence, an informed patient is now likely to question whether this test should be done, how often it should be done, and whether a prostate examination should also be included.

The push toward population health has raised questions about the value of a preventive wellness visit, especially in healthy patients.^10,11 Arguments against the annual physical do not account for the value of these visits, which provide the opportunity to have time-intensive shared decision-making conversations and build a trusting patient-physician relationship. The value of the annual physical is not simply to do examinations for which there is limited evidence; it is a time for us to get to know our patients, to update their preventive needs (and the medical record), and to discuss which screening tests they may safely forgo to avoid unnecessary false-positives, leading to excess cost and harm.

This trusting relationship, developed over years, is likely to save both the patient and the healthcare system significant money. For example, it enables us to reassure patients that an antibiotic is not needed for their upper respiratory infection, to encourage them to try a dietary change before proceeding with computed tomography for their abdominal pain, or to discourage them from inappropriately aggressive screening tests that may result in overtesting or overdiagnosis.

Unfortunately, it is nearly impossible to accurately quantify these substantial benefits to the healthcare system and patients. And there is a real potential that recommendations against the annual physical may eventually affect future reimbursement, which would add to the time pressures of an already overburdened primary care workforce.

DO PRIMARY CARE PHYSICIANS MAKE A DIFFERENCE?

As medicine and technology evolve, patients have more ways to access care. However, the Internet also provides patients with access to more conflicting information than ever before, making it even more important for clinicians, as trusted partners in their patients’ health, to help patients navigate the waters of information and misinformation.

Studies have shown that having a primary care physician is associated with a longer life span, higher likelihood of reporting good health, and similar clinical outcomes for common conditions such as diabetes and hypertension when compared with subspecialty care, but at a lower cost and with less resource utilization.^12,13 In a study published in 2019, Basu et al¹² found that for every 10 additional primary care physicians per 100,000 population, there was an associated 51.5-day increase in life expectancy, compared with a 19.2-day increase for specialists. Cost savings also occur. Similarly, a review by the American College of Physicians¹³ found that each additional primary care physician per 10,000 population in a US state increased the state’s health quality ranking by more than 10 spots and reduced their overall spending per Medicare beneficiary. In contrast, an increase of 1 specialist per 10,000 population was linked to a 9-spot decrease in health-quality ranking and an increase in spending.

WHY CHOOSE PRIMARY CARE?

As medical students, we fell in love with internal medicine because of the complexity and intellectual challenges of working through a diagnostic dilemma. There is a certain excitement in not knowing what type of patients will show up that day.

Primary care’s focus on continuity and developing long-standing relationships with patients and their families is largely unmatched in the subspecialty field. It is satisfying to have a general knowledge of the human body, and the central vantage point with which to weigh different subspecialty recommendations. We feel such sentiments are common to those interested in primary care, but sadly, we believe these are not enough to sustain the future of primary care internal medicine.

IS THE FUTURE BRIGHT OR BLEAK?

Primary care internists must resist the call to “run twice as fast.” Instead, we need to look for ways where our unique skill sets can benefit the health of our nation while attracting students to internal medicine primary care. The following are potential areas for moving forward.

The aging of America

The US Census Bureau projects that by the year 2035, older adults will outnumber children for the first time in US history, and by the year 2060, nearly 25% of the US population will be 65 or older.¹⁴ The rise of the geriatric patient and the need for comprehensive care will create a continued demand for primary care internists. There certainly aren’t enough geriatricians to meet this need, and primary care internists are well trained to fill this gap.

The rise of the team approach

As we are learning, complex disease management benefits from a team approach. The rise of new models of care delivery such as accountable care organizations and patient-centered medical homes echo this reality. The day of a single provider “doing it all” is fading.

The focus on population health in these models has given rise to multidisciplinary teams—including physicians, nurses, advanced practice providers, social workers, and pharmacists—whose function is to help manage and improve the physical, mental, and social care of patients, often in a capitated payment system. The primary care internist can play a key role in leading these teams, and such partnerships may help lessen reliance on the current primary care hustle of 15- to 20-minute visits. In such models, it is possible that the internist will need to see each patient only once or twice a year, in a longer appointment slot, instead of 4 to 6 times per year in rushed visits. The hope is that this will encourage the relationship-building that is so important in primary care and reduce the time and volume scheduling burdens seen in the current fee-for-service system.

The joy of digging into a diagnostic dilemma has been a hallmark of internal medicine. The rise of technology should enable primary care internists to increase their diagnostic capabilities in the office without an overreliance on subspecialists.

Examples of technology that may benefit primary care are artificial intelligence with real-time diagnostic support, precision medicine, and office-based point-of-care ultrasonography.^15–17 By increasing the diagnostic power of an office-based visit, we hope that the prestige factor of primary care medicine will increase as internists incorporate such advances into their clinics—not to mention the joy of making an appropriate diagnosis in real time.

Reimbursement and the value of time

Time is a valuable commodity for primary care internists. Unfortunately, there seems to be less of it in today’s practice. Gone are the days when we could go to the doctors’ dining room to decompress, chat, and break bread with colleagues. Today, we are more likely to be found in front of our computers over lunch answering patients’ messages. Time is also a key reason that physicians express frustration with issues such as prior authorizations for medications. These tasks routinely take time away from what is valuable—the care of our patients.

The rise of innovative practice models such as direct primary care and concierge medicine can be seen as a market response to the frustrations of increasing regulatory complexity, billing hassles, and lack of time. However, some have cautioned that such models have the potential to worsen healthcare disparities because patients pay out of pocket for some or all of their care in these practices.¹⁸

Interestingly, the Centers for Medicare and Medicaid Services recently unveiled new voluntary payment models for primary care that go into effect in 2020. These models may allow for increased practice innovation. The 2 proposed options are Primary Care First (designed for small primary care practices) and Direct Contracting (aimed at larger practices). These models are designed to provide a predictable up-front payment stream (a set payment per beneficiary) to the primary care practice. Hopefully, these options will move primary care away from the current fee-for-service, multiple-patient-visit model.

The primary care model allows practices to “assume financial risk in exchange for reduced administrative burden and performance-based payments” and “introduces new, higher payments for practices that care for complex, chronically ill patients.”¹⁹ It is too soon to know the effectiveness of such models, but any reimbursement innovation should be met with cautious optimism.

In addition, the Centers for Medicare and Medicaid Services has recently moved to reduce requirements for documentation. For example, one can fully bill with a medical student note without needing to repeat visit notes.^20,21 Such changes should decrease the time needed to document the EMR and free up more time to care for patients.

A CALL TO ACTION

The national shortage of primary care providers points to the fact that this is a difficult career, and one that remains undervalued. One step we need to take is to protect the time we have with patients. It is doubtful that seeing a greater number of sicker patients each day, in addition to the responsibilities of proactive population-based care (“panel management”), will attract younger generations of physicians to fill this void, no matter what technology we adopt.

Keys to facilitating this change include understanding the value of primary care physicians, decreasing the burden of documentation, facilitating team-care options to support them, and expanding diagnostic tools available to use within primary care. If we don’t demand change, who will be there to take care of us when we grow old?

“My dear, here we must run as fast as we can, just to stay in place. And if you wish to go anywhere you must run twice as fast as that.”
—Lewis Carroll
Alice’s Adventures in Wonderland

The future of primary care internal medicine physicians is uncertain. According to a 2018 survey of internal medicine residents conducted by the American College of Physicians, only 11% were considering primary care as a career path.¹ In 1998, that number was 54%.²

See related commentary

Possible reasons are many:

• Lower pay compared with subspecialists in a pay system that rewards procedural competency over mental effort
• Work schedules less flexible than in other specialties (eg, hospital medicine practitioners may have 1 week on and 1 week off)
• Perceived lack of respect
• Increasing regulatory and record-keeping burdens
• Tyranny of 15- to 20-minute appointments (irrespective of patient complexity)
• Scope-of-practice concerns as other providers seek primary care equivalency status (eg, pharmacists, walk-in clinics, advanced practice providers, telemedicine providers).

The result is a projected shortage of primary care physicians of 21,100 to 55,200 by 2030, according to a 2019 report by the Association of American Medical Colleges,³ despite an expected growth in advanced practice providers in primary care such as nurse practitioners and physician assistants.

A practical result of this shortage will be even less patient access to primary care physicians. A 2017 national survey found that the average wait time for a new patient-physician appointment has already increased by 30% since 2014.⁴ The wait time to see a primary care physician varied between 29 days in major metropolitan areas (up 50% from 2014) and 56 days in mid-sized markets. The longest waits by market size were 109 days for new patients in Boston, MA, and 122 days for those living in Albany, NY.

What are the implications?

In this issue, Pravia and Diaz⁵ make the case that primary care providers must adapt their practices to meet the needs of younger generations by increasing their use of technology. We agree that telemedicine, wearable medical devices, and enhanced patient communication through the electronic medical record (EMR) are here to stay and should be embraced.

However, we have seen the challenges of adopting technologic advances without first making an adjustment to the volume-driven patient schedule. For such advances to be successfully integrated into a clinical practice, it is vital to be cognizant of the current challenges encountered in primary care internal medicine.

UNIQUE BURDENS ON PRIMARY CARE

In addition to the stress of addressing multiple complex medical problems within a short time, evaluating multiple medical problems often leads to increases in results to review, forms to complete, and calls to patients. Even treatment plans initiated by specialists are often deferred to primary care providers for dosing adjustments, follow-up laboratory testing, and monitoring.

Moreover, patients often seek a second opinion from their primary care provider regarding care provided by subspecialists, as they consider their primary care provider to be the doctor who knows them best. And though it can be personally gratifying to be considered a trusted partner in the patient’s care, these requests often result in additional phone calls to the office or another thing to address within a complex visit.

A large in-box can be daunting in the setting of increased EMR demands. Whether you have difficulty putting in basic orders or are an advanced user, each upgrade can make you feel like you’re using the EMR for the first time. This is a problem for all specialties, but in primary care, one is addressing a large spectrum of concerns, so there is less opportunity to use standardized templates that can help buffer the problem.

A study of primary care providers found that nearly 75% of each patient visit was spent on activities other than face-to-face patient care, including working with the EMR.⁶ Similarly, a study using in-office observations and after-hours diaries found that physicians from various specialties spend 2 hours on administrative duties for each hour that they see patients in the office, followed by an additional 1 to 2 hours of work after clinic, mostly devoted to the EMR.⁷

Clinicians using scribes to help with record-keeping duties often need to see more patients to compensate for the cost. Adding 2 to 3 patients to a daily schedule usually means adding more medical conditions to manage, with an exponential increase in testing and in-box burden.

The additional burden this coverage creates in primary care is often not well understood by those in other specialties.

Another burden unique to primary care providers is the nearly continuous publication of guidelines that are often confusing and discrepant. Because many high-impact guidelines represent expert consensus or evidence from specialist perspectives, they may not fit the primary care model or values: eg, primary care guidelines tend to place more emphasis on harms associated with screening.

Screening for breast and prostate cancers is a prime example. Both require shared decision-making based on patient preferences and values.^8,9 Detailed discussions about preventive screening can be difficult to achieve within the context of a medical visit owing to time limitations, especially if other medical conditions being addressed are equally controversial, such as blood pressure target goals. A decade ago, one could easily declare, “It’s time for your annual PSA test,” and move on to other concerns. Given the changing evidence, an informed patient is now likely to question whether this test should be done, how often it should be done, and whether a prostate examination should also be included.

The push toward population health has raised questions about the value of a preventive wellness visit, especially in healthy patients.^10,11 Arguments against the annual physical do not account for the value of these visits, which provide the opportunity to have time-intensive shared decision-making conversations and build a trusting patient-physician relationship. The value of the annual physical is not simply to do examinations for which there is limited evidence; it is a time for us to get to know our patients, to update their preventive needs (and the medical record), and to discuss which screening tests they may safely forgo to avoid unnecessary false-positives, leading to excess cost and harm.

This trusting relationship, developed over years, is likely to save both the patient and the healthcare system significant money. For example, it enables us to reassure patients that an antibiotic is not needed for their upper respiratory infection, to encourage them to try a dietary change before proceeding with computed tomography for their abdominal pain, or to discourage them from inappropriately aggressive screening tests that may result in overtesting or overdiagnosis.

Unfortunately, it is nearly impossible to accurately quantify these substantial benefits to the healthcare system and patients. And there is a real potential that recommendations against the annual physical may eventually affect future reimbursement, which would add to the time pressures of an already overburdened primary care workforce.

DO PRIMARY CARE PHYSICIANS MAKE A DIFFERENCE?

As medicine and technology evolve, patients have more ways to access care. However, the Internet also provides patients with access to more conflicting information than ever before, making it even more important for clinicians, as trusted partners in their patients’ health, to help patients navigate the waters of information and misinformation.

Studies have shown that having a primary care physician is associated with a longer life span, higher likelihood of reporting good health, and similar clinical outcomes for common conditions such as diabetes and hypertension when compared with subspecialty care, but at a lower cost and with less resource utilization.^12,13 In a study published in 2019, Basu et al¹² found that for every 10 additional primary care physicians per 100,000 population, there was an associated 51.5-day increase in life expectancy, compared with a 19.2-day increase for specialists. Cost savings also occur. Similarly, a review by the American College of Physicians¹³ found that each additional primary care physician per 10,000 population in a US state increased the state’s health quality ranking by more than 10 spots and reduced their overall spending per Medicare beneficiary. In contrast, an increase of 1 specialist per 10,000 population was linked to a 9-spot decrease in health-quality ranking and an increase in spending.

WHY CHOOSE PRIMARY CARE?

As medical students, we fell in love with internal medicine because of the complexity and intellectual challenges of working through a diagnostic dilemma. There is a certain excitement in not knowing what type of patients will show up that day.

Primary care’s focus on continuity and developing long-standing relationships with patients and their families is largely unmatched in the subspecialty field. It is satisfying to have a general knowledge of the human body, and the central vantage point with which to weigh different subspecialty recommendations. We feel such sentiments are common to those interested in primary care, but sadly, we believe these are not enough to sustain the future of primary care internal medicine.

IS THE FUTURE BRIGHT OR BLEAK?

Primary care internists must resist the call to “run twice as fast.” Instead, we need to look for ways where our unique skill sets can benefit the health of our nation while attracting students to internal medicine primary care. The following are potential areas for moving forward.

The aging of America

The US Census Bureau projects that by the year 2035, older adults will outnumber children for the first time in US history, and by the year 2060, nearly 25% of the US population will be 65 or older.¹⁴ The rise of the geriatric patient and the need for comprehensive care will create a continued demand for primary care internists. There certainly aren’t enough geriatricians to meet this need, and primary care internists are well trained to fill this gap.

The rise of the team approach

As we are learning, complex disease management benefits from a team approach. The rise of new models of care delivery such as accountable care organizations and patient-centered medical homes echo this reality. The day of a single provider “doing it all” is fading.

The focus on population health in these models has given rise to multidisciplinary teams—including physicians, nurses, advanced practice providers, social workers, and pharmacists—whose function is to help manage and improve the physical, mental, and social care of patients, often in a capitated payment system. The primary care internist can play a key role in leading these teams, and such partnerships may help lessen reliance on the current primary care hustle of 15- to 20-minute visits. In such models, it is possible that the internist will need to see each patient only once or twice a year, in a longer appointment slot, instead of 4 to 6 times per year in rushed visits. The hope is that this will encourage the relationship-building that is so important in primary care and reduce the time and volume scheduling burdens seen in the current fee-for-service system.

The joy of digging into a diagnostic dilemma has been a hallmark of internal medicine. The rise of technology should enable primary care internists to increase their diagnostic capabilities in the office without an overreliance on subspecialists.

Examples of technology that may benefit primary care are artificial intelligence with real-time diagnostic support, precision medicine, and office-based point-of-care ultrasonography.^15–17 By increasing the diagnostic power of an office-based visit, we hope that the prestige factor of primary care medicine will increase as internists incorporate such advances into their clinics—not to mention the joy of making an appropriate diagnosis in real time.

Reimbursement and the value of time

Time is a valuable commodity for primary care internists. Unfortunately, there seems to be less of it in today’s practice. Gone are the days when we could go to the doctors’ dining room to decompress, chat, and break bread with colleagues. Today, we are more likely to be found in front of our computers over lunch answering patients’ messages. Time is also a key reason that physicians express frustration with issues such as prior authorizations for medications. These tasks routinely take time away from what is valuable—the care of our patients.

The rise of innovative practice models such as direct primary care and concierge medicine can be seen as a market response to the frustrations of increasing regulatory complexity, billing hassles, and lack of time. However, some have cautioned that such models have the potential to worsen healthcare disparities because patients pay out of pocket for some or all of their care in these practices.¹⁸

Interestingly, the Centers for Medicare and Medicaid Services recently unveiled new voluntary payment models for primary care that go into effect in 2020. These models may allow for increased practice innovation. The 2 proposed options are Primary Care First (designed for small primary care practices) and Direct Contracting (aimed at larger practices). These models are designed to provide a predictable up-front payment stream (a set payment per beneficiary) to the primary care practice. Hopefully, these options will move primary care away from the current fee-for-service, multiple-patient-visit model.

The primary care model allows practices to “assume financial risk in exchange for reduced administrative burden and performance-based payments” and “introduces new, higher payments for practices that care for complex, chronically ill patients.”¹⁹ It is too soon to know the effectiveness of such models, but any reimbursement innovation should be met with cautious optimism.

In addition, the Centers for Medicare and Medicaid Services has recently moved to reduce requirements for documentation. For example, one can fully bill with a medical student note without needing to repeat visit notes.^20,21 Such changes should decrease the time needed to document the EMR and free up more time to care for patients.

A CALL TO ACTION

The national shortage of primary care providers points to the fact that this is a difficult career, and one that remains undervalued. One step we need to take is to protect the time we have with patients. It is doubtful that seeing a greater number of sicker patients each day, in addition to the responsibilities of proactive population-based care (“panel management”), will attract younger generations of physicians to fill this void, no matter what technology we adopt.

Keys to facilitating this change include understanding the value of primary care physicians, decreasing the burden of documentation, facilitating team-care options to support them, and expanding diagnostic tools available to use within primary care. If we don’t demand change, who will be there to take care of us when we grow old?

“My dear, here we must run as fast as we can, just to stay in place. And if you wish to go anywhere you must run twice as fast as that.”
—Lewis Carroll
Alice’s Adventures in Wonderland

The future of primary care internal medicine physicians is uncertain. According to a 2018 survey of internal medicine residents conducted by the American College of Physicians, only 11% were considering primary care as a career path.¹ In 1998, that number was 54%.²

See related commentary

Possible reasons are many:

• Lower pay compared with subspecialists in a pay system that rewards procedural competency over mental effort
• Work schedules less flexible than in other specialties (eg, hospital medicine practitioners may have 1 week on and 1 week off)
• Perceived lack of respect
• Increasing regulatory and record-keeping burdens
• Tyranny of 15- to 20-minute appointments (irrespective of patient complexity)
• Scope-of-practice concerns as other providers seek primary care equivalency status (eg, pharmacists, walk-in clinics, advanced practice providers, telemedicine providers).

The result is a projected shortage of primary care physicians of 21,100 to 55,200 by 2030, according to a 2019 report by the Association of American Medical Colleges,³ despite an expected growth in advanced practice providers in primary care such as nurse practitioners and physician assistants.

A practical result of this shortage will be even less patient access to primary care physicians. A 2017 national survey found that the average wait time for a new patient-physician appointment has already increased by 30% since 2014.⁴ The wait time to see a primary care physician varied between 29 days in major metropolitan areas (up 50% from 2014) and 56 days in mid-sized markets. The longest waits by market size were 109 days for new patients in Boston, MA, and 122 days for those living in Albany, NY.

What are the implications?

In this issue, Pravia and Diaz⁵ make the case that primary care providers must adapt their practices to meet the needs of younger generations by increasing their use of technology. We agree that telemedicine, wearable medical devices, and enhanced patient communication through the electronic medical record (EMR) are here to stay and should be embraced.

However, we have seen the challenges of adopting technologic advances without first making an adjustment to the volume-driven patient schedule. For such advances to be successfully integrated into a clinical practice, it is vital to be cognizant of the current challenges encountered in primary care internal medicine.

UNIQUE BURDENS ON PRIMARY CARE

In addition to the stress of addressing multiple complex medical problems within a short time, evaluating multiple medical problems often leads to increases in results to review, forms to complete, and calls to patients. Even treatment plans initiated by specialists are often deferred to primary care providers for dosing adjustments, follow-up laboratory testing, and monitoring.

Moreover, patients often seek a second opinion from their primary care provider regarding care provided by subspecialists, as they consider their primary care provider to be the doctor who knows them best. And though it can be personally gratifying to be considered a trusted partner in the patient’s care, these requests often result in additional phone calls to the office or another thing to address within a complex visit.

A large in-box can be daunting in the setting of increased EMR demands. Whether you have difficulty putting in basic orders or are an advanced user, each upgrade can make you feel like you’re using the EMR for the first time. This is a problem for all specialties, but in primary care, one is addressing a large spectrum of concerns, so there is less opportunity to use standardized templates that can help buffer the problem.

A study of primary care providers found that nearly 75% of each patient visit was spent on activities other than face-to-face patient care, including working with the EMR.⁶ Similarly, a study using in-office observations and after-hours diaries found that physicians from various specialties spend 2 hours on administrative duties for each hour that they see patients in the office, followed by an additional 1 to 2 hours of work after clinic, mostly devoted to the EMR.⁷

Clinicians using scribes to help with record-keeping duties often need to see more patients to compensate for the cost. Adding 2 to 3 patients to a daily schedule usually means adding more medical conditions to manage, with an exponential increase in testing and in-box burden.

The additional burden this coverage creates in primary care is often not well understood by those in other specialties.

Another burden unique to primary care providers is the nearly continuous publication of guidelines that are often confusing and discrepant. Because many high-impact guidelines represent expert consensus or evidence from specialist perspectives, they may not fit the primary care model or values: eg, primary care guidelines tend to place more emphasis on harms associated with screening.

Screening for breast and prostate cancers is a prime example. Both require shared decision-making based on patient preferences and values.^8,9 Detailed discussions about preventive screening can be difficult to achieve within the context of a medical visit owing to time limitations, especially if other medical conditions being addressed are equally controversial, such as blood pressure target goals. A decade ago, one could easily declare, “It’s time for your annual PSA test,” and move on to other concerns. Given the changing evidence, an informed patient is now likely to question whether this test should be done, how often it should be done, and whether a prostate examination should also be included.

The push toward population health has raised questions about the value of a preventive wellness visit, especially in healthy patients.^10,11 Arguments against the annual physical do not account for the value of these visits, which provide the opportunity to have time-intensive shared decision-making conversations and build a trusting patient-physician relationship. The value of the annual physical is not simply to do examinations for which there is limited evidence; it is a time for us to get to know our patients, to update their preventive needs (and the medical record), and to discuss which screening tests they may safely forgo to avoid unnecessary false-positives, leading to excess cost and harm.

This trusting relationship, developed over years, is likely to save both the patient and the healthcare system significant money. For example, it enables us to reassure patients that an antibiotic is not needed for their upper respiratory infection, to encourage them to try a dietary change before proceeding with computed tomography for their abdominal pain, or to discourage them from inappropriately aggressive screening tests that may result in overtesting or overdiagnosis.

Unfortunately, it is nearly impossible to accurately quantify these substantial benefits to the healthcare system and patients. And there is a real potential that recommendations against the annual physical may eventually affect future reimbursement, which would add to the time pressures of an already overburdened primary care workforce.

DO PRIMARY CARE PHYSICIANS MAKE A DIFFERENCE?

As medicine and technology evolve, patients have more ways to access care. However, the Internet also provides patients with access to more conflicting information than ever before, making it even more important for clinicians, as trusted partners in their patients’ health, to help patients navigate the waters of information and misinformation.

Studies have shown that having a primary care physician is associated with a longer life span, higher likelihood of reporting good health, and similar clinical outcomes for common conditions such as diabetes and hypertension when compared with subspecialty care, but at a lower cost and with less resource utilization.^12,13 In a study published in 2019, Basu et al¹² found that for every 10 additional primary care physicians per 100,000 population, there was an associated 51.5-day increase in life expectancy, compared with a 19.2-day increase for specialists. Cost savings also occur. Similarly, a review by the American College of Physicians¹³ found that each additional primary care physician per 10,000 population in a US state increased the state’s health quality ranking by more than 10 spots and reduced their overall spending per Medicare beneficiary. In contrast, an increase of 1 specialist per 10,000 population was linked to a 9-spot decrease in health-quality ranking and an increase in spending.

WHY CHOOSE PRIMARY CARE?

As medical students, we fell in love with internal medicine because of the complexity and intellectual challenges of working through a diagnostic dilemma. There is a certain excitement in not knowing what type of patients will show up that day.

Primary care’s focus on continuity and developing long-standing relationships with patients and their families is largely unmatched in the subspecialty field. It is satisfying to have a general knowledge of the human body, and the central vantage point with which to weigh different subspecialty recommendations. We feel such sentiments are common to those interested in primary care, but sadly, we believe these are not enough to sustain the future of primary care internal medicine.

IS THE FUTURE BRIGHT OR BLEAK?

Primary care internists must resist the call to “run twice as fast.” Instead, we need to look for ways where our unique skill sets can benefit the health of our nation while attracting students to internal medicine primary care. The following are potential areas for moving forward.

The aging of America

The US Census Bureau projects that by the year 2035, older adults will outnumber children for the first time in US history, and by the year 2060, nearly 25% of the US population will be 65 or older.¹⁴ The rise of the geriatric patient and the need for comprehensive care will create a continued demand for primary care internists. There certainly aren’t enough geriatricians to meet this need, and primary care internists are well trained to fill this gap.

The rise of the team approach

As we are learning, complex disease management benefits from a team approach. The rise of new models of care delivery such as accountable care organizations and patient-centered medical homes echo this reality. The day of a single provider “doing it all” is fading.

The focus on population health in these models has given rise to multidisciplinary teams—including physicians, nurses, advanced practice providers, social workers, and pharmacists—whose function is to help manage and improve the physical, mental, and social care of patients, often in a capitated payment system. The primary care internist can play a key role in leading these teams, and such partnerships may help lessen reliance on the current primary care hustle of 15- to 20-minute visits. In such models, it is possible that the internist will need to see each patient only once or twice a year, in a longer appointment slot, instead of 4 to 6 times per year in rushed visits. The hope is that this will encourage the relationship-building that is so important in primary care and reduce the time and volume scheduling burdens seen in the current fee-for-service system.

The joy of digging into a diagnostic dilemma has been a hallmark of internal medicine. The rise of technology should enable primary care internists to increase their diagnostic capabilities in the office without an overreliance on subspecialists.

Examples of technology that may benefit primary care are artificial intelligence with real-time diagnostic support, precision medicine, and office-based point-of-care ultrasonography.^15–17 By increasing the diagnostic power of an office-based visit, we hope that the prestige factor of primary care medicine will increase as internists incorporate such advances into their clinics—not to mention the joy of making an appropriate diagnosis in real time.

Reimbursement and the value of time

Time is a valuable commodity for primary care internists. Unfortunately, there seems to be less of it in today’s practice. Gone are the days when we could go to the doctors’ dining room to decompress, chat, and break bread with colleagues. Today, we are more likely to be found in front of our computers over lunch answering patients’ messages. Time is also a key reason that physicians express frustration with issues such as prior authorizations for medications. These tasks routinely take time away from what is valuable—the care of our patients.

The rise of innovative practice models such as direct primary care and concierge medicine can be seen as a market response to the frustrations of increasing regulatory complexity, billing hassles, and lack of time. However, some have cautioned that such models have the potential to worsen healthcare disparities because patients pay out of pocket for some or all of their care in these practices.¹⁸

Interestingly, the Centers for Medicare and Medicaid Services recently unveiled new voluntary payment models for primary care that go into effect in 2020. These models may allow for increased practice innovation. The 2 proposed options are Primary Care First (designed for small primary care practices) and Direct Contracting (aimed at larger practices). These models are designed to provide a predictable up-front payment stream (a set payment per beneficiary) to the primary care practice. Hopefully, these options will move primary care away from the current fee-for-service, multiple-patient-visit model.

The primary care model allows practices to “assume financial risk in exchange for reduced administrative burden and performance-based payments” and “introduces new, higher payments for practices that care for complex, chronically ill patients.”¹⁹ It is too soon to know the effectiveness of such models, but any reimbursement innovation should be met with cautious optimism.

In addition, the Centers for Medicare and Medicaid Services has recently moved to reduce requirements for documentation. For example, one can fully bill with a medical student note without needing to repeat visit notes.^20,21 Such changes should decrease the time needed to document the EMR and free up more time to care for patients.

A CALL TO ACTION

The national shortage of primary care providers points to the fact that this is a difficult career, and one that remains undervalued. One step we need to take is to protect the time we have with patients. It is doubtful that seeing a greater number of sicker patients each day, in addition to the responsibilities of proactive population-based care (“panel management”), will attract younger generations of physicians to fill this void, no matter what technology we adopt.

Keys to facilitating this change include understanding the value of primary care physicians, decreasing the burden of documentation, facilitating team-care options to support them, and expanding diagnostic tools available to use within primary care. If we don’t demand change, who will be there to take care of us when we grow old?

Autism spectrum disorder (ASD) is a neurodevelopmental condition typically diagnosed early in life: the median age at diagnosis is 52 months.¹ Because research demonstrates the benefits of early intervention,² when we think about people with ASD, we generally think about children and adolescents. 

See related article

However, autism spans the entirety of one’s life. This means that children with ASD will grow to be adults with ASD. The US Centers for Disease Control and Prevention estimated that 1 in 59 children were diagnosed with ASD during the surveillance year 2014,¹ which was nearly double the prevalence from just 8 years earlier,³ and a 15% increase since 2012.⁴ As these children grow up, this translates to an ever-growing number of adults with autism.

UNMET NEEDS

Healthcare, housing, and intellectual and developmental disability services for adults with ASD currently fall well short of meeting the needs of this exploding population. If solutions are to be realized, innovative approaches must be employed.

Swetlik et al,⁵ in this issue of the Journal, offer valuable insights into the challenges that practitioners and their adult patients with ASD encounter as a result of seismic shifts in diagnostic criteria, increasing prevalence, and changes to healthcare financial coverage. They also review behavioral and pharmacologic treatments, reproductive health, and caregiver fatigue and discuss the role of the physician and other healthcare practitioners who are likely to have only limited exposure to adult patients with ASD. These wide-ranging considerations speak to the complexity of the healthcare needs of this population.

Swetlik et al also underscore that transition planning is essential for primary care, psychiatry, behavioral health services, continuing education, skill development, and appropriate prevocational training for adolescents with ASD, and yet it is often underutilized or unavailable. There is a dearth of experienced practitioners across these disciplines to serve adults with ASD. The complexity of navigating bureaucratic processes to secure funding (typically Medicaid) supports the necessity of planning early to achieve desired outcomes for each young adult. Additionally, the number of Medicaid waivers that fund many supportive services are limited.

GROWING UP IS HARD; START PLANNING EARLY

Swetlik et al describe the stress these circumstances create for people with ASD and their families. Entering adulthood is a complicated process, fraught with emotional overtones that must include medical care, work considerations, legal and financial arrangements, and, for many, the search for an appropriate residential environment. Planning for these transitions should begin years before adulthood if the process is to work smoothly and effectively.

A transition involving a shift away from a team of familiar pediatric healthcare providers to unfamiliar adult practitioners can be distressing for any adolescent with a chronic condition. For those with ASD, who may have diminished socialization and communication skills, the transition can be especially challenging and must be handled with care.

This transition pales in comparison with the disruptive force of a permanent move out of the family home. Over the next 10 years, 500,000 youths in the United States will age out of school-based ASD services,⁶ and a great many of them will be put on long waiting lists for residential placement.⁷

For young adults with ASD, particularly those with complex needs, establishing an advantageous long-term living arrangement may mean the difference between a healthy, self-directed launch into a new phase of life, or a consequential misstep that exacerbates or worsens symptoms and creates new stressors for the young adult and his or her family. It is especially important that arrangements be made before an aging guardian starts to experience declining health.

Thoughtful and deliberate preplanning helps to reduce stress and prevent emergency placements, and promotes long-term quality of life for people with ASD.

For many years, the prevailing model for the provision of long-term care services for individuals with intellectual and developmental disabilities was institutional care. Large facilities, often located in expansive, self-contained campuses, provided around-the-clock care. Residents slept, ate, worked, and were expected to receive social and emotional fulfillment at the facility.

For some, this was an acceptable model. For many, it was not, but there were few available alternatives. At its best, this model provided a safe environment for its residents, but it did not facilitate achieving an integrated, self-directed life experience. At its worst, neglect and abuse were rampant.

Numerous legislative acts, court decisions, and advocacy efforts drove the deinstitutionalization movement for individuals with intellectual and developmental disabilities between the early 1960s and today. The 1999 case of Olmstead v LC⁸ was among the most significant. In this landmark case, in accordance with the 1990 Americans With Disabilities Act, the US Supreme Court ruled that people with disabilities have the right to receive state-funded services and support in the community rather than in institutions, as long as several criteria are met:

• Community supports are appropriate
• The individual desires to live in the community
• The accommodations to facilitate that arrangement are considered to be reasonable.

In the 20 years since the Olmstead decision, residential services for adults have shifted at an accelerated rate away from institutions toward smaller, community-based settings.^9,10 Community models include but are not limited to:

• Group homes that serve individuals with intellectual and developmental disabilities and provide 24-hour support
• Apartments or homes where individuals live and receive intermittent, less-intensive support
• Adult foster care.

DSM-5: AUTISM IS HETEROGENEOUS

In the fifth edition of the Diagnostic and Statistical Manual of Mental Disorders (DSM-5),¹¹ ASD is characterized by persistent deficits in social interaction and social communication, which begin in early development and are observed in conjunction with restricted, repetitive behaviors, interests, or activities.

DSM-5 provides more than 20 examples of how these criteria might be met. Further, DSM-5 encourages clinicians to select diagnostic specifiers to address overall symptom severity, cognitive abilities, and associated medical conditions.

ONE RESIDENTIAL MODEL DOES NOT FIT ALL

The complex matrix of potential symptom manifestations in people with ASD clearly indicates the need for numerous distinctive residential models for adults with ASD.

One person with severe symptoms of ASD may require one-to-one staffing and proactive preparatory support in order to safely leave the house for a desired social experience. The person may be unable to read, to independently access public transportation, to cope with deviations in expected routine, to initiate conversation, or to remain calm if distressed. This person would benefit from a residential model that allows for a high staffing ratio, access to appropriate transportation, sophisticated autism-informed supports, and the availability of social experiences that are easily accessed—in other words, a very controlled environment.

Another person, with less severe symptoms and fewer behavioral challenges, who possesses a driver’s license and holds a job, may struggle with isolation and loneliness resulting from social inhibitions and skill deficits. This person’s support needs would differ, with emphasis placed on maintaining the appropriate social context rather than on providing a high level of individualized behavioral support.

The shift away from a one-size-fits-all institutional model for long-term care has benefited many individuals with intellectual and developmental disabilities who have experienced opportunities for community integration.

Still, for many adults with ASD, particularly those with complex needs and complex behavior profiles, the widespread conceptual shift to new and different models that assume that all people with intellectual and developmental disabilities will benefit from smaller, scattered-site settings is ill-fitting. It is erroneous to believe that for all adults with ASD, regardless of the complexity of their symptoms, living within a broader community of neurotypically developing neighbors breeds a richer sense of inclusion and connectivity.

FINDING CARE CAN BE DAUNTING

Families of adults with severe symptoms of ASD who seek placement in more traditional community residential models often find it difficult to find capable healthcare providers to serve them. Such settings are ill-equipped to deal with significantly challenging behaviors such as physical aggression, self-injury, property destruction, and elopement (wandering). These supported housing models lack the necessary staffing resources.

Further, publicly available funding options for stand-alone group homes do not typically allow for intensive supervision and management from professionals with expertise in autism. Without specialists who can  implement autism-specific best-practice methods for assessment, service planning, staff training, data collection, and the provision of visual and technological supports for residents, it is difficult to achieve desired outcomes. For example, patients can find it challenging to visit physicians’ offices for preventive and urgent care. Lacking a caregiver who is familiar with the adult patient with ASD and who can help express his or her concerns to healthcare providers, efficient evaluation of any potentially serious medical issue is a daunting task.

A residential model that is gaining popularity across the United States among families and individuals affected by ASD is the intentional community.

Although forms and functions may vary, intentional communities are planned residential developments that promote social cohesion and strive to meet the shared needs of its members. Intentional communities for adults with ASD are designed to meet their social, communication, sensory, and behavioral needs. Every detail from the selection of land, to the construction of housing, the selected staffing model, the daily structure, and the considerations for transportation and amenities are all informed by the specific needs of individuals with autism. Safety, integration, self-direction, independence, and social connectivity are common goals.   

Successful intentional communities designed for people with intellectual and developmental disabilities often have facilities devoted to recreation, continuing education, socialization, and supportive services. Staff members who work within these communities are highly trained in the unique needs of people with these disorders. Intentional communities aspire to embody the individualized, integrated community-living approach that the Olmstead decision called for, while simultaneously offering the resource-rich, safe, and supportive experience that a campus atmosphere can offer.

Almost all recently developed models allow for residents to live among neurotypical peers and have easy access to the broader community. Communities range in size from several condominiums on a cul-de-sac to expansive developments with more than a hundred homes.

The allure of an ASD-informed intentional community that provides for the social, vocational, health, and safety needs of its residents is similar to that which leads large numbers of aging, neurotypically developing individuals to seek out retirement communities. Nationally recognized models of intentional communities include First Place (Phoenix, AZ), Sweetwater Spectrum (Sonoma, CA), Cape Cod Village (Orleans, MA), and Bittersweet Farms in Ohio.

First Place is a 55-unit apartment complex near downtown Phoenix that identifies as “community-connected” and “transit-oriented.” Although there are some individuals in the complex who do not have ASD, the development was created for those who do. The goal is to enhance the quality of life for residents through the provision of housing, jobs, social opportunities, and a supportive community.

Sweetwater Spectrum is located blocks from the Sonoma downtown plaza, on just under 3 acres of land. It includes several 4-bedroom homes, a community center with a kitchen, exercise studio, media room, and library, an expansive organic garden, and an outdoor pool.

The Autism Housing Network lists more than 75 intentional communities on its resource page. There are many exciting models in development. For example, Monarch Center for Autism in Cleveland, OH, is planning to develop an innovative intentional community. It will include mixed supported living options for adults across the autism spectrum, separate housing options for parents and family members, on-site social and recreational opportunities, green space, and retail stores intended to serve members of the surrounding community and provide employment and socialization opportunities for its residents.

Casa Familia in South Florida will soon begin constructing a large intentional community that will include innovative housing options, classrooms, social areas, an auditorium, walkways, bike paths, pools, and social enterprises.

It is critical that these ASD intentional communities continue to emerge to meet the long-term needs of the rapidly growing and aging ASD population. 

THE TIME TO ACT IS NOW

Swetlik et al synthesize important, contemporary research on adult ASD healthcare considerations, pursuant to informing the many decisions that physicians and other healthcare professionals must make to address the diverse needs of this population. Their article advocates for further research and highlights the crisis surrounding the scarcity of practitioners specializing in adult ASD.

As for current healthcare providers, parents, care coordinators, and other stakeholders who are tasked with transition planning for individuals with ASD, particularly those with severe symptoms, the time to act is now, especially in creating new intentional community models.

Most adult healthcare providers have not been routinely charged with the responsibility, nor do they have the available time and resources to meet the social and communication needs of these patients. But when faced with an ever-expanding group of patients who demonstrate inadequate social and communication skills, the healthcare system must not turn a blind eye.

The symptoms of autism do not magically resolve when a child reaches adulthood. The medical community must partner with society at large to offer transitional solutions, including intentional communities, to the rapidly growing number of adults with ASD. Current demand outweighs supply, but if we work together, we can create innovative and highly effective solutions. After all, children with autism do not disappear. They grow into adults with autism.

Autism spectrum disorder (ASD) is a neurodevelopmental condition typically diagnosed early in life: the median age at diagnosis is 52 months.¹ Because research demonstrates the benefits of early intervention,² when we think about people with ASD, we generally think about children and adolescents. 

See related article

However, autism spans the entirety of one’s life. This means that children with ASD will grow to be adults with ASD. The US Centers for Disease Control and Prevention estimated that 1 in 59 children were diagnosed with ASD during the surveillance year 2014,¹ which was nearly double the prevalence from just 8 years earlier,³ and a 15% increase since 2012.⁴ As these children grow up, this translates to an ever-growing number of adults with autism.

UNMET NEEDS

Healthcare, housing, and intellectual and developmental disability services for adults with ASD currently fall well short of meeting the needs of this exploding population. If solutions are to be realized, innovative approaches must be employed.

Swetlik et al,⁵ in this issue of the Journal, offer valuable insights into the challenges that practitioners and their adult patients with ASD encounter as a result of seismic shifts in diagnostic criteria, increasing prevalence, and changes to healthcare financial coverage. They also review behavioral and pharmacologic treatments, reproductive health, and caregiver fatigue and discuss the role of the physician and other healthcare practitioners who are likely to have only limited exposure to adult patients with ASD. These wide-ranging considerations speak to the complexity of the healthcare needs of this population.

Swetlik et al also underscore that transition planning is essential for primary care, psychiatry, behavioral health services, continuing education, skill development, and appropriate prevocational training for adolescents with ASD, and yet it is often underutilized or unavailable. There is a dearth of experienced practitioners across these disciplines to serve adults with ASD. The complexity of navigating bureaucratic processes to secure funding (typically Medicaid) supports the necessity of planning early to achieve desired outcomes for each young adult. Additionally, the number of Medicaid waivers that fund many supportive services are limited.

GROWING UP IS HARD; START PLANNING EARLY

Swetlik et al describe the stress these circumstances create for people with ASD and their families. Entering adulthood is a complicated process, fraught with emotional overtones that must include medical care, work considerations, legal and financial arrangements, and, for many, the search for an appropriate residential environment. Planning for these transitions should begin years before adulthood if the process is to work smoothly and effectively.

A transition involving a shift away from a team of familiar pediatric healthcare providers to unfamiliar adult practitioners can be distressing for any adolescent with a chronic condition. For those with ASD, who may have diminished socialization and communication skills, the transition can be especially challenging and must be handled with care.

This transition pales in comparison with the disruptive force of a permanent move out of the family home. Over the next 10 years, 500,000 youths in the United States will age out of school-based ASD services,⁶ and a great many of them will be put on long waiting lists for residential placement.⁷

For young adults with ASD, particularly those with complex needs, establishing an advantageous long-term living arrangement may mean the difference between a healthy, self-directed launch into a new phase of life, or a consequential misstep that exacerbates or worsens symptoms and creates new stressors for the young adult and his or her family. It is especially important that arrangements be made before an aging guardian starts to experience declining health.

Thoughtful and deliberate preplanning helps to reduce stress and prevent emergency placements, and promotes long-term quality of life for people with ASD.

For many years, the prevailing model for the provision of long-term care services for individuals with intellectual and developmental disabilities was institutional care. Large facilities, often located in expansive, self-contained campuses, provided around-the-clock care. Residents slept, ate, worked, and were expected to receive social and emotional fulfillment at the facility.

For some, this was an acceptable model. For many, it was not, but there were few available alternatives. At its best, this model provided a safe environment for its residents, but it did not facilitate achieving an integrated, self-directed life experience. At its worst, neglect and abuse were rampant.

Numerous legislative acts, court decisions, and advocacy efforts drove the deinstitutionalization movement for individuals with intellectual and developmental disabilities between the early 1960s and today. The 1999 case of Olmstead v LC⁸ was among the most significant. In this landmark case, in accordance with the 1990 Americans With Disabilities Act, the US Supreme Court ruled that people with disabilities have the right to receive state-funded services and support in the community rather than in institutions, as long as several criteria are met:

• Community supports are appropriate
• The individual desires to live in the community
• The accommodations to facilitate that arrangement are considered to be reasonable.

In the 20 years since the Olmstead decision, residential services for adults have shifted at an accelerated rate away from institutions toward smaller, community-based settings.^9,10 Community models include but are not limited to:

• Group homes that serve individuals with intellectual and developmental disabilities and provide 24-hour support
• Apartments or homes where individuals live and receive intermittent, less-intensive support
• Adult foster care.

DSM-5: AUTISM IS HETEROGENEOUS

In the fifth edition of the Diagnostic and Statistical Manual of Mental Disorders (DSM-5),¹¹ ASD is characterized by persistent deficits in social interaction and social communication, which begin in early development and are observed in conjunction with restricted, repetitive behaviors, interests, or activities.

DSM-5 provides more than 20 examples of how these criteria might be met. Further, DSM-5 encourages clinicians to select diagnostic specifiers to address overall symptom severity, cognitive abilities, and associated medical conditions.

ONE RESIDENTIAL MODEL DOES NOT FIT ALL

The complex matrix of potential symptom manifestations in people with ASD clearly indicates the need for numerous distinctive residential models for adults with ASD.

One person with severe symptoms of ASD may require one-to-one staffing and proactive preparatory support in order to safely leave the house for a desired social experience. The person may be unable to read, to independently access public transportation, to cope with deviations in expected routine, to initiate conversation, or to remain calm if distressed. This person would benefit from a residential model that allows for a high staffing ratio, access to appropriate transportation, sophisticated autism-informed supports, and the availability of social experiences that are easily accessed—in other words, a very controlled environment.

Another person, with less severe symptoms and fewer behavioral challenges, who possesses a driver’s license and holds a job, may struggle with isolation and loneliness resulting from social inhibitions and skill deficits. This person’s support needs would differ, with emphasis placed on maintaining the appropriate social context rather than on providing a high level of individualized behavioral support.

The shift away from a one-size-fits-all institutional model for long-term care has benefited many individuals with intellectual and developmental disabilities who have experienced opportunities for community integration.

Still, for many adults with ASD, particularly those with complex needs and complex behavior profiles, the widespread conceptual shift to new and different models that assume that all people with intellectual and developmental disabilities will benefit from smaller, scattered-site settings is ill-fitting. It is erroneous to believe that for all adults with ASD, regardless of the complexity of their symptoms, living within a broader community of neurotypically developing neighbors breeds a richer sense of inclusion and connectivity.

FINDING CARE CAN BE DAUNTING

Families of adults with severe symptoms of ASD who seek placement in more traditional community residential models often find it difficult to find capable healthcare providers to serve them. Such settings are ill-equipped to deal with significantly challenging behaviors such as physical aggression, self-injury, property destruction, and elopement (wandering). These supported housing models lack the necessary staffing resources.

Further, publicly available funding options for stand-alone group homes do not typically allow for intensive supervision and management from professionals with expertise in autism. Without specialists who can  implement autism-specific best-practice methods for assessment, service planning, staff training, data collection, and the provision of visual and technological supports for residents, it is difficult to achieve desired outcomes. For example, patients can find it challenging to visit physicians’ offices for preventive and urgent care. Lacking a caregiver who is familiar with the adult patient with ASD and who can help express his or her concerns to healthcare providers, efficient evaluation of any potentially serious medical issue is a daunting task.

A residential model that is gaining popularity across the United States among families and individuals affected by ASD is the intentional community.

Although forms and functions may vary, intentional communities are planned residential developments that promote social cohesion and strive to meet the shared needs of its members. Intentional communities for adults with ASD are designed to meet their social, communication, sensory, and behavioral needs. Every detail from the selection of land, to the construction of housing, the selected staffing model, the daily structure, and the considerations for transportation and amenities are all informed by the specific needs of individuals with autism. Safety, integration, self-direction, independence, and social connectivity are common goals.   

Successful intentional communities designed for people with intellectual and developmental disabilities often have facilities devoted to recreation, continuing education, socialization, and supportive services. Staff members who work within these communities are highly trained in the unique needs of people with these disorders. Intentional communities aspire to embody the individualized, integrated community-living approach that the Olmstead decision called for, while simultaneously offering the resource-rich, safe, and supportive experience that a campus atmosphere can offer.

Almost all recently developed models allow for residents to live among neurotypical peers and have easy access to the broader community. Communities range in size from several condominiums on a cul-de-sac to expansive developments with more than a hundred homes.

The allure of an ASD-informed intentional community that provides for the social, vocational, health, and safety needs of its residents is similar to that which leads large numbers of aging, neurotypically developing individuals to seek out retirement communities. Nationally recognized models of intentional communities include First Place (Phoenix, AZ), Sweetwater Spectrum (Sonoma, CA), Cape Cod Village (Orleans, MA), and Bittersweet Farms in Ohio.

First Place is a 55-unit apartment complex near downtown Phoenix that identifies as “community-connected” and “transit-oriented.” Although there are some individuals in the complex who do not have ASD, the development was created for those who do. The goal is to enhance the quality of life for residents through the provision of housing, jobs, social opportunities, and a supportive community.

Sweetwater Spectrum is located blocks from the Sonoma downtown plaza, on just under 3 acres of land. It includes several 4-bedroom homes, a community center with a kitchen, exercise studio, media room, and library, an expansive organic garden, and an outdoor pool.

The Autism Housing Network lists more than 75 intentional communities on its resource page. There are many exciting models in development. For example, Monarch Center for Autism in Cleveland, OH, is planning to develop an innovative intentional community. It will include mixed supported living options for adults across the autism spectrum, separate housing options for parents and family members, on-site social and recreational opportunities, green space, and retail stores intended to serve members of the surrounding community and provide employment and socialization opportunities for its residents.

Casa Familia in South Florida will soon begin constructing a large intentional community that will include innovative housing options, classrooms, social areas, an auditorium, walkways, bike paths, pools, and social enterprises.

It is critical that these ASD intentional communities continue to emerge to meet the long-term needs of the rapidly growing and aging ASD population. 

THE TIME TO ACT IS NOW

Swetlik et al synthesize important, contemporary research on adult ASD healthcare considerations, pursuant to informing the many decisions that physicians and other healthcare professionals must make to address the diverse needs of this population. Their article advocates for further research and highlights the crisis surrounding the scarcity of practitioners specializing in adult ASD.

As for current healthcare providers, parents, care coordinators, and other stakeholders who are tasked with transition planning for individuals with ASD, particularly those with severe symptoms, the time to act is now, especially in creating new intentional community models.

Most adult healthcare providers have not been routinely charged with the responsibility, nor do they have the available time and resources to meet the social and communication needs of these patients. But when faced with an ever-expanding group of patients who demonstrate inadequate social and communication skills, the healthcare system must not turn a blind eye.

The symptoms of autism do not magically resolve when a child reaches adulthood. The medical community must partner with society at large to offer transitional solutions, including intentional communities, to the rapidly growing number of adults with ASD. Current demand outweighs supply, but if we work together, we can create innovative and highly effective solutions. After all, children with autism do not disappear. They grow into adults with autism.

Autism spectrum disorder (ASD) is a neurodevelopmental condition typically diagnosed early in life: the median age at diagnosis is 52 months.¹ Because research demonstrates the benefits of early intervention,² when we think about people with ASD, we generally think about children and adolescents. 

See related article

However, autism spans the entirety of one’s life. This means that children with ASD will grow to be adults with ASD. The US Centers for Disease Control and Prevention estimated that 1 in 59 children were diagnosed with ASD during the surveillance year 2014,¹ which was nearly double the prevalence from just 8 years earlier,³ and a 15% increase since 2012.⁴ As these children grow up, this translates to an ever-growing number of adults with autism.

UNMET NEEDS

Healthcare, housing, and intellectual and developmental disability services for adults with ASD currently fall well short of meeting the needs of this exploding population. If solutions are to be realized, innovative approaches must be employed.

Swetlik et al,⁵ in this issue of the Journal, offer valuable insights into the challenges that practitioners and their adult patients with ASD encounter as a result of seismic shifts in diagnostic criteria, increasing prevalence, and changes to healthcare financial coverage. They also review behavioral and pharmacologic treatments, reproductive health, and caregiver fatigue and discuss the role of the physician and other healthcare practitioners who are likely to have only limited exposure to adult patients with ASD. These wide-ranging considerations speak to the complexity of the healthcare needs of this population.

Swetlik et al also underscore that transition planning is essential for primary care, psychiatry, behavioral health services, continuing education, skill development, and appropriate prevocational training for adolescents with ASD, and yet it is often underutilized or unavailable. There is a dearth of experienced practitioners across these disciplines to serve adults with ASD. The complexity of navigating bureaucratic processes to secure funding (typically Medicaid) supports the necessity of planning early to achieve desired outcomes for each young adult. Additionally, the number of Medicaid waivers that fund many supportive services are limited.

GROWING UP IS HARD; START PLANNING EARLY

Swetlik et al describe the stress these circumstances create for people with ASD and their families. Entering adulthood is a complicated process, fraught with emotional overtones that must include medical care, work considerations, legal and financial arrangements, and, for many, the search for an appropriate residential environment. Planning for these transitions should begin years before adulthood if the process is to work smoothly and effectively.

A transition involving a shift away from a team of familiar pediatric healthcare providers to unfamiliar adult practitioners can be distressing for any adolescent with a chronic condition. For those with ASD, who may have diminished socialization and communication skills, the transition can be especially challenging and must be handled with care.

This transition pales in comparison with the disruptive force of a permanent move out of the family home. Over the next 10 years, 500,000 youths in the United States will age out of school-based ASD services,⁶ and a great many of them will be put on long waiting lists for residential placement.⁷

For young adults with ASD, particularly those with complex needs, establishing an advantageous long-term living arrangement may mean the difference between a healthy, self-directed launch into a new phase of life, or a consequential misstep that exacerbates or worsens symptoms and creates new stressors for the young adult and his or her family. It is especially important that arrangements be made before an aging guardian starts to experience declining health.

Thoughtful and deliberate preplanning helps to reduce stress and prevent emergency placements, and promotes long-term quality of life for people with ASD.

For many years, the prevailing model for the provision of long-term care services for individuals with intellectual and developmental disabilities was institutional care. Large facilities, often located in expansive, self-contained campuses, provided around-the-clock care. Residents slept, ate, worked, and were expected to receive social and emotional fulfillment at the facility.

For some, this was an acceptable model. For many, it was not, but there were few available alternatives. At its best, this model provided a safe environment for its residents, but it did not facilitate achieving an integrated, self-directed life experience. At its worst, neglect and abuse were rampant.

Numerous legislative acts, court decisions, and advocacy efforts drove the deinstitutionalization movement for individuals with intellectual and developmental disabilities between the early 1960s and today. The 1999 case of Olmstead v LC⁸ was among the most significant. In this landmark case, in accordance with the 1990 Americans With Disabilities Act, the US Supreme Court ruled that people with disabilities have the right to receive state-funded services and support in the community rather than in institutions, as long as several criteria are met:

• Community supports are appropriate
• The individual desires to live in the community
• The accommodations to facilitate that arrangement are considered to be reasonable.

In the 20 years since the Olmstead decision, residential services for adults have shifted at an accelerated rate away from institutions toward smaller, community-based settings.^9,10 Community models include but are not limited to:

• Group homes that serve individuals with intellectual and developmental disabilities and provide 24-hour support
• Apartments or homes where individuals live and receive intermittent, less-intensive support
• Adult foster care.

DSM-5: AUTISM IS HETEROGENEOUS

In the fifth edition of the Diagnostic and Statistical Manual of Mental Disorders (DSM-5),¹¹ ASD is characterized by persistent deficits in social interaction and social communication, which begin in early development and are observed in conjunction with restricted, repetitive behaviors, interests, or activities.

DSM-5 provides more than 20 examples of how these criteria might be met. Further, DSM-5 encourages clinicians to select diagnostic specifiers to address overall symptom severity, cognitive abilities, and associated medical conditions.

ONE RESIDENTIAL MODEL DOES NOT FIT ALL

The complex matrix of potential symptom manifestations in people with ASD clearly indicates the need for numerous distinctive residential models for adults with ASD.

One person with severe symptoms of ASD may require one-to-one staffing and proactive preparatory support in order to safely leave the house for a desired social experience. The person may be unable to read, to independently access public transportation, to cope with deviations in expected routine, to initiate conversation, or to remain calm if distressed. This person would benefit from a residential model that allows for a high staffing ratio, access to appropriate transportation, sophisticated autism-informed supports, and the availability of social experiences that are easily accessed—in other words, a very controlled environment.

Another person, with less severe symptoms and fewer behavioral challenges, who possesses a driver’s license and holds a job, may struggle with isolation and loneliness resulting from social inhibitions and skill deficits. This person’s support needs would differ, with emphasis placed on maintaining the appropriate social context rather than on providing a high level of individualized behavioral support.

The shift away from a one-size-fits-all institutional model for long-term care has benefited many individuals with intellectual and developmental disabilities who have experienced opportunities for community integration.

Still, for many adults with ASD, particularly those with complex needs and complex behavior profiles, the widespread conceptual shift to new and different models that assume that all people with intellectual and developmental disabilities will benefit from smaller, scattered-site settings is ill-fitting. It is erroneous to believe that for all adults with ASD, regardless of the complexity of their symptoms, living within a broader community of neurotypically developing neighbors breeds a richer sense of inclusion and connectivity.

FINDING CARE CAN BE DAUNTING

Families of adults with severe symptoms of ASD who seek placement in more traditional community residential models often find it difficult to find capable healthcare providers to serve them. Such settings are ill-equipped to deal with significantly challenging behaviors such as physical aggression, self-injury, property destruction, and elopement (wandering). These supported housing models lack the necessary staffing resources.

Further, publicly available funding options for stand-alone group homes do not typically allow for intensive supervision and management from professionals with expertise in autism. Without specialists who can  implement autism-specific best-practice methods for assessment, service planning, staff training, data collection, and the provision of visual and technological supports for residents, it is difficult to achieve desired outcomes. For example, patients can find it challenging to visit physicians’ offices for preventive and urgent care. Lacking a caregiver who is familiar with the adult patient with ASD and who can help express his or her concerns to healthcare providers, efficient evaluation of any potentially serious medical issue is a daunting task.

A residential model that is gaining popularity across the United States among families and individuals affected by ASD is the intentional community.

Although forms and functions may vary, intentional communities are planned residential developments that promote social cohesion and strive to meet the shared needs of its members. Intentional communities for adults with ASD are designed to meet their social, communication, sensory, and behavioral needs. Every detail from the selection of land, to the construction of housing, the selected staffing model, the daily structure, and the considerations for transportation and amenities are all informed by the specific needs of individuals with autism. Safety, integration, self-direction, independence, and social connectivity are common goals.   

Successful intentional communities designed for people with intellectual and developmental disabilities often have facilities devoted to recreation, continuing education, socialization, and supportive services. Staff members who work within these communities are highly trained in the unique needs of people with these disorders. Intentional communities aspire to embody the individualized, integrated community-living approach that the Olmstead decision called for, while simultaneously offering the resource-rich, safe, and supportive experience that a campus atmosphere can offer.

Almost all recently developed models allow for residents to live among neurotypical peers and have easy access to the broader community. Communities range in size from several condominiums on a cul-de-sac to expansive developments with more than a hundred homes.

The allure of an ASD-informed intentional community that provides for the social, vocational, health, and safety needs of its residents is similar to that which leads large numbers of aging, neurotypically developing individuals to seek out retirement communities. Nationally recognized models of intentional communities include First Place (Phoenix, AZ), Sweetwater Spectrum (Sonoma, CA), Cape Cod Village (Orleans, MA), and Bittersweet Farms in Ohio.

First Place is a 55-unit apartment complex near downtown Phoenix that identifies as “community-connected” and “transit-oriented.” Although there are some individuals in the complex who do not have ASD, the development was created for those who do. The goal is to enhance the quality of life for residents through the provision of housing, jobs, social opportunities, and a supportive community.

Sweetwater Spectrum is located blocks from the Sonoma downtown plaza, on just under 3 acres of land. It includes several 4-bedroom homes, a community center with a kitchen, exercise studio, media room, and library, an expansive organic garden, and an outdoor pool.

The Autism Housing Network lists more than 75 intentional communities on its resource page. There are many exciting models in development. For example, Monarch Center for Autism in Cleveland, OH, is planning to develop an innovative intentional community. It will include mixed supported living options for adults across the autism spectrum, separate housing options for parents and family members, on-site social and recreational opportunities, green space, and retail stores intended to serve members of the surrounding community and provide employment and socialization opportunities for its residents.

Casa Familia in South Florida will soon begin constructing a large intentional community that will include innovative housing options, classrooms, social areas, an auditorium, walkways, bike paths, pools, and social enterprises.

It is critical that these ASD intentional communities continue to emerge to meet the long-term needs of the rapidly growing and aging ASD population. 

THE TIME TO ACT IS NOW

Swetlik et al synthesize important, contemporary research on adult ASD healthcare considerations, pursuant to informing the many decisions that physicians and other healthcare professionals must make to address the diverse needs of this population. Their article advocates for further research and highlights the crisis surrounding the scarcity of practitioners specializing in adult ASD.

As for current healthcare providers, parents, care coordinators, and other stakeholders who are tasked with transition planning for individuals with ASD, particularly those with severe symptoms, the time to act is now, especially in creating new intentional community models.

Most adult healthcare providers have not been routinely charged with the responsibility, nor do they have the available time and resources to meet the social and communication needs of these patients. But when faced with an ever-expanding group of patients who demonstrate inadequate social and communication skills, the healthcare system must not turn a blind eye.

The symptoms of autism do not magically resolve when a child reaches adulthood. The medical community must partner with society at large to offer transitional solutions, including intentional communities, to the rapidly growing number of adults with ASD. Current demand outweighs supply, but if we work together, we can create innovative and highly effective solutions. After all, children with autism do not disappear. They grow into adults with autism.

To the Editor: We read with great pleasure the article by Sharayah et al about acute gastro­paresis in a patient with diabetic ketoacidosis.¹ However, in the case description, the authors reached a diagnosis of gastroparesis secondary to diabetic ketoacidosis without aptly ruling out some of its most common causes such as hypokalemia and other electrolyte imbalances seen in diabetic patients (in the setting of recurrent vomiting).

The authors also did not include the patient’s duration of diabetes or hemoglobin A_1c level, both of which are linked with gastroparesis in diabetic patients.² Pertinent biochemical information that can help readers formulate a rational approach and journey to making a diagnosis appears elusive in their article.

To the Editor: We read with great pleasure the article by Sharayah et al about acute gastro­paresis in a patient with diabetic ketoacidosis.¹ However, in the case description, the authors reached a diagnosis of gastroparesis secondary to diabetic ketoacidosis without aptly ruling out some of its most common causes such as hypokalemia and other electrolyte imbalances seen in diabetic patients (in the setting of recurrent vomiting).

The authors also did not include the patient’s duration of diabetes or hemoglobin A_1c level, both of which are linked with gastroparesis in diabetic patients.² Pertinent biochemical information that can help readers formulate a rational approach and journey to making a diagnosis appears elusive in their article.

To the Editor: We read with great pleasure the article by Sharayah et al about acute gastro­paresis in a patient with diabetic ketoacidosis.¹ However, in the case description, the authors reached a diagnosis of gastroparesis secondary to diabetic ketoacidosis without aptly ruling out some of its most common causes such as hypokalemia and other electrolyte imbalances seen in diabetic patients (in the setting of recurrent vomiting).

The authors also did not include the patient’s duration of diabetes or hemoglobin A_1c level, both of which are linked with gastroparesis in diabetic patients.² Pertinent biochemical information that can help readers formulate a rational approach and journey to making a diagnosis appears elusive in their article.

In Reply: We thank the readers for their letter. Our patient’s laboratory values at the time of presentation were as follows:

• Corrected sodium 142 mmol/L
• Potassium 5.5 mmol/L
• Phosphorus 6.6 mmol/L.

The rest of the electrolyte levels were within normal limits.

These reported electrolyte levels were unlikely to cause such gastroparesis. The patient’s hemoglobin A_1c was 8.7% at the time of presentation, with no previous values available. However, since abdominal computed tomography done 1 year before this presentation did not show stomach dilation and the patient was asymptomatic, his gastroparesis was presumed to be acute.

In Reply: We thank the readers for their letter. Our patient’s laboratory values at the time of presentation were as follows:

• Corrected sodium 142 mmol/L
• Potassium 5.5 mmol/L
• Phosphorus 6.6 mmol/L.

The rest of the electrolyte levels were within normal limits.

These reported electrolyte levels were unlikely to cause such gastroparesis. The patient’s hemoglobin A_1c was 8.7% at the time of presentation, with no previous values available. However, since abdominal computed tomography done 1 year before this presentation did not show stomach dilation and the patient was asymptomatic, his gastroparesis was presumed to be acute.

In Reply: We thank the readers for their letter. Our patient’s laboratory values at the time of presentation were as follows:

• Corrected sodium 142 mmol/L
• Potassium 5.5 mmol/L
• Phosphorus 6.6 mmol/L.

The rest of the electrolyte levels were within normal limits.

These reported electrolyte levels were unlikely to cause such gastroparesis. The patient’s hemoglobin A_1c was 8.7% at the time of presentation, with no previous values available. However, since abdominal computed tomography done 1 year before this presentation did not show stomach dilation and the patient was asymptomatic, his gastroparesis was presumed to be acute.

Infant mortality dropped slightly but not significantly in 2017, according to data released Aug. 1 by the National Center for Health Statistics, based on data from the National Vital Statistics System.

The rate for 2017 was 5.79 deaths per 1,000 live births, which was not statistically different from the rate of 5.87 in 2016, the National Center for Health Statistics said in a new report. Neonatal and postneonatal mortality – 3.85 and 1.94 per 1,000, respectively – both showed the same nonsignificant drop from 2016 to 2017.

About two-thirds of the infants who died in 2017 were children born preterm (less than 37 weeks’ gestation), the NCHS said, and “the mortality rate for infants born before 28 weeks of gestation [389.4 per 1,000] was 183 times the rate for term infants” born at 37-41 weeks.

Rates at the state level in 2017 ranged from a low of 3.66 deaths/1,000 live births in Massachusetts to a high of 8.73/1,000 in Mississippi. Washington (3.88) was the only other state with a rate below 4.0, while Arkansas (8.10) was the only other state above 8.0 (The District of Columbia had a rate of 8.16.). Infant mortality was significantly lower than the national rate in 11 states and significantly higher in 15 states and D.C., according to the report.

Overall, in 2017, 3,855,500 live births occurred, with 22,341 infants having died before the age of 1 year, data from the National Vital Statistics System’s linked birth/infant death file show. In 1995, the first year that the linked file was available, the corresponding numbers were 3,899,589 births and 29,505 deaths, for a rate of 7.57 deaths/1,000 live births.

[email protected]

Infant mortality dropped slightly but not significantly in 2017, according to data released Aug. 1 by the National Center for Health Statistics, based on data from the National Vital Statistics System.

The rate for 2017 was 5.79 deaths per 1,000 live births, which was not statistically different from the rate of 5.87 in 2016, the National Center for Health Statistics said in a new report. Neonatal and postneonatal mortality – 3.85 and 1.94 per 1,000, respectively – both showed the same nonsignificant drop from 2016 to 2017.

About two-thirds of the infants who died in 2017 were children born preterm (less than 37 weeks’ gestation), the NCHS said, and “the mortality rate for infants born before 28 weeks of gestation [389.4 per 1,000] was 183 times the rate for term infants” born at 37-41 weeks.

Rates at the state level in 2017 ranged from a low of 3.66 deaths/1,000 live births in Massachusetts to a high of 8.73/1,000 in Mississippi. Washington (3.88) was the only other state with a rate below 4.0, while Arkansas (8.10) was the only other state above 8.0 (The District of Columbia had a rate of 8.16.). Infant mortality was significantly lower than the national rate in 11 states and significantly higher in 15 states and D.C., according to the report.

Overall, in 2017, 3,855,500 live births occurred, with 22,341 infants having died before the age of 1 year, data from the National Vital Statistics System’s linked birth/infant death file show. In 1995, the first year that the linked file was available, the corresponding numbers were 3,899,589 births and 29,505 deaths, for a rate of 7.57 deaths/1,000 live births.

[email protected]

Infant mortality dropped slightly but not significantly in 2017, according to data released Aug. 1 by the National Center for Health Statistics, based on data from the National Vital Statistics System.

The rate for 2017 was 5.79 deaths per 1,000 live births, which was not statistically different from the rate of 5.87 in 2016, the National Center for Health Statistics said in a new report. Neonatal and postneonatal mortality – 3.85 and 1.94 per 1,000, respectively – both showed the same nonsignificant drop from 2016 to 2017.

About two-thirds of the infants who died in 2017 were children born preterm (less than 37 weeks’ gestation), the NCHS said, and “the mortality rate for infants born before 28 weeks of gestation [389.4 per 1,000] was 183 times the rate for term infants” born at 37-41 weeks.

Rates at the state level in 2017 ranged from a low of 3.66 deaths/1,000 live births in Massachusetts to a high of 8.73/1,000 in Mississippi. Washington (3.88) was the only other state with a rate below 4.0, while Arkansas (8.10) was the only other state above 8.0 (The District of Columbia had a rate of 8.16.). Infant mortality was significantly lower than the national rate in 11 states and significantly higher in 15 states and D.C., according to the report.

Overall, in 2017, 3,855,500 live births occurred, with 22,341 infants having died before the age of 1 year, data from the National Vital Statistics System’s linked birth/infant death file show. In 1995, the first year that the linked file was available, the corresponding numbers were 3,899,589 births and 29,505 deaths, for a rate of 7.57 deaths/1,000 live births.

[email protected]