Extremes of hospital utilization by patients with sickle cell disease (SCD) are problematic for patients, clinicians, and policymakers.110 Although patients manage their pain at home most of the time, even acute crises,11 a small minority of SCD patients accounts for a remarkable amount of hospital resource utilization.1, 3, 4, 6, 1114 Where it is quite unusual for a patient with SCD to be hospitalized more than twice per year,1, 11 in prior work with payer datasets our group identified some patients who were hospitalized more frequently than once per month. In rare cases, admission rates exceeding once per week were identified.1 High‐utilizing SCD patients, and particularly the very high‐utilizing subset, account for the majority of costs of care for the population.13, 14

In previous work by our group describing hospital utilization among members of a regional Medicaid MCO, results suggested that high utilization was a relatively transient phenomenon for most patients, likely resulting from short‐term increases in hospitalization rates among previously moderate utilizers.1 However, high‐utilizing members whose inpatient admission rate did not quickly moderate were progressively less likely to resume a more typical utilization pattern.

The present study used the State Inpatient Databases for years 2004 to 2007 from the Agency for Healthcare Research and Quality to replicate prior findings and to investigate questions not addressed in our prior work. Specifically, hospital discharge data from all hospitals in the state of California were examined to identify first‐year adolescent and adult high utilizers and to follow their hospital utilization over time. The objectives of the study were as follows:

• 1

  To identify historical predictors of a period of high utilization by comparing diagnoses between 20042006 in patients who were new high utilizers in 2007 with those who were never high utilizers.

• 2

  To identify predictors of a persistent rather than moderating course by following patients who were new high utilizers in 2005 over the succeeding 2 years.

• 3

  To replicate prior findings on the course of high hospital utilization.

Patients and Methods

Construction of the Study Subset

The study data set was constructed as follows (Fig. 1):

• 1

  Patient identifiers associated with a diagnosis of sickle cell disease were selected by identifying admissions with ICD‐9 diagnosis codes for sickle cell disease appearing in the first 10 diagnoses for calendar years 2004 to 2007 (these included ICD‐9 codes 282.60 to 282.64, 282.68, 282.69, 282.41, and 282.42). Of this group, patients who had a record of at least 1 admission for sickle cell crisis were identified. An admission for crisis was operationalized as a hospitalization with 1 discharge diagnosis coded as 282.42, 282.62, 282.64, or 282.69. This yielded a data set of 34,363 admissions among 3169 patients.

• 2

  Admissions with missing patient identification numbers were excluded (n = 2365 of 34,363 admissions, 6.88%).

• 3

  Hospitalizations were tabulated for each unique patient identifier.

• 4

  Patients with a known age of 13 years or more in 2004 were selected. There were 481 patients excluded due to age below 13 years, and 814 excluded for having an uncertain age. The final sample consisted of 1874 unique patient identifiers representing 10,704 hospital admissions.

 

Figure 1

As patients who were hospitalized more often were more likely to have inconsistent data, the exclusion for unknown and inconsistent age likely biased the findings by excluding more frequently hospitalized patients. Further post‐hoc analyses were conducted to gauge the extent of this bias, reported in Results, below.

Results

Course of New High Utilizers

Patients who were high utilizers in 2005 but not in 2004 were identified, and their hospital utilization from 2005 to 2007 was plotted (Fig. 2). The results are shown in Figure 1. Fifty‐five of the original 91 (60.44%) new high utilizers moderated in the following year, and 6.59% were known to have died in the hospital. Of the surviving 30 (32.97%) who did not moderate in the second year, 19 (65.3%) continued the high‐utilizing pattern into the third year, while 9 (16.36%) of those who moderated in year 2 returned to the high‐utilizing pattern in year 3. During this 3‐year period, 10 members (10.99%) of the initial group died in the hospital.

Figure 2

Discussion

Extremes of hospital utilization by patients with sickle cell disease (SCD) are problematic for patients, clinicians, and policymakers.110 Although patients manage their pain at home most of the time, even acute crises,11 a small minority of SCD patients accounts for a remarkable amount of hospital resource utilization.1, 3, 4, 6, 1114 Where it is quite unusual for a patient with SCD to be hospitalized more than twice per year,1, 11 in prior work with payer datasets our group identified some patients who were hospitalized more frequently than once per month. In rare cases, admission rates exceeding once per week were identified.1 High‐utilizing SCD patients, and particularly the very high‐utilizing subset, account for the majority of costs of care for the population.13, 14

In previous work by our group describing hospital utilization among members of a regional Medicaid MCO, results suggested that high utilization was a relatively transient phenomenon for most patients, likely resulting from short‐term increases in hospitalization rates among previously moderate utilizers.1 However, high‐utilizing members whose inpatient admission rate did not quickly moderate were progressively less likely to resume a more typical utilization pattern.

The present study used the State Inpatient Databases for years 2004 to 2007 from the Agency for Healthcare Research and Quality to replicate prior findings and to investigate questions not addressed in our prior work. Specifically, hospital discharge data from all hospitals in the state of California were examined to identify first‐year adolescent and adult high utilizers and to follow their hospital utilization over time. The objectives of the study were as follows:

• 1

  To identify historical predictors of a period of high utilization by comparing diagnoses between 20042006 in patients who were new high utilizers in 2007 with those who were never high utilizers.

• 2

  To identify predictors of a persistent rather than moderating course by following patients who were new high utilizers in 2005 over the succeeding 2 years.

• 3

  To replicate prior findings on the course of high hospital utilization.

Patients and Methods

Construction of the Study Subset

The study data set was constructed as follows (Fig. 1):

• 1

  Patient identifiers associated with a diagnosis of sickle cell disease were selected by identifying admissions with ICD‐9 diagnosis codes for sickle cell disease appearing in the first 10 diagnoses for calendar years 2004 to 2007 (these included ICD‐9 codes 282.60 to 282.64, 282.68, 282.69, 282.41, and 282.42). Of this group, patients who had a record of at least 1 admission for sickle cell crisis were identified. An admission for crisis was operationalized as a hospitalization with 1 discharge diagnosis coded as 282.42, 282.62, 282.64, or 282.69. This yielded a data set of 34,363 admissions among 3169 patients.

• 2

  Admissions with missing patient identification numbers were excluded (n = 2365 of 34,363 admissions, 6.88%).

• 3

  Hospitalizations were tabulated for each unique patient identifier.

• 4

  Patients with a known age of 13 years or more in 2004 were selected. There were 481 patients excluded due to age below 13 years, and 814 excluded for having an uncertain age. The final sample consisted of 1874 unique patient identifiers representing 10,704 hospital admissions.

 

Figure 1

As patients who were hospitalized more often were more likely to have inconsistent data, the exclusion for unknown and inconsistent age likely biased the findings by excluding more frequently hospitalized patients. Further post‐hoc analyses were conducted to gauge the extent of this bias, reported in Results, below.

Results

Course of New High Utilizers

Patients who were high utilizers in 2005 but not in 2004 were identified, and their hospital utilization from 2005 to 2007 was plotted (Fig. 2). The results are shown in Figure 1. Fifty‐five of the original 91 (60.44%) new high utilizers moderated in the following year, and 6.59% were known to have died in the hospital. Of the surviving 30 (32.97%) who did not moderate in the second year, 19 (65.3%) continued the high‐utilizing pattern into the third year, while 9 (16.36%) of those who moderated in year 2 returned to the high‐utilizing pattern in year 3. During this 3‐year period, 10 members (10.99%) of the initial group died in the hospital.

Figure 2

Discussion

Extremes of hospital utilization by patients with sickle cell disease (SCD) are problematic for patients, clinicians, and policymakers.110 Although patients manage their pain at home most of the time, even acute crises,11 a small minority of SCD patients accounts for a remarkable amount of hospital resource utilization.1, 3, 4, 6, 1114 Where it is quite unusual for a patient with SCD to be hospitalized more than twice per year,1, 11 in prior work with payer datasets our group identified some patients who were hospitalized more frequently than once per month. In rare cases, admission rates exceeding once per week were identified.1 High‐utilizing SCD patients, and particularly the very high‐utilizing subset, account for the majority of costs of care for the population.13, 14

In previous work by our group describing hospital utilization among members of a regional Medicaid MCO, results suggested that high utilization was a relatively transient phenomenon for most patients, likely resulting from short‐term increases in hospitalization rates among previously moderate utilizers.1 However, high‐utilizing members whose inpatient admission rate did not quickly moderate were progressively less likely to resume a more typical utilization pattern.

The present study used the State Inpatient Databases for years 2004 to 2007 from the Agency for Healthcare Research and Quality to replicate prior findings and to investigate questions not addressed in our prior work. Specifically, hospital discharge data from all hospitals in the state of California were examined to identify first‐year adolescent and adult high utilizers and to follow their hospital utilization over time. The objectives of the study were as follows:

• 1

  To identify historical predictors of a period of high utilization by comparing diagnoses between 20042006 in patients who were new high utilizers in 2007 with those who were never high utilizers.

• 2

  To identify predictors of a persistent rather than moderating course by following patients who were new high utilizers in 2005 over the succeeding 2 years.

• 3

  To replicate prior findings on the course of high hospital utilization.

Patients and Methods

Construction of the Study Subset

The study data set was constructed as follows (Fig. 1):

• 1

  Patient identifiers associated with a diagnosis of sickle cell disease were selected by identifying admissions with ICD‐9 diagnosis codes for sickle cell disease appearing in the first 10 diagnoses for calendar years 2004 to 2007 (these included ICD‐9 codes 282.60 to 282.64, 282.68, 282.69, 282.41, and 282.42). Of this group, patients who had a record of at least 1 admission for sickle cell crisis were identified. An admission for crisis was operationalized as a hospitalization with 1 discharge diagnosis coded as 282.42, 282.62, 282.64, or 282.69. This yielded a data set of 34,363 admissions among 3169 patients.

• 2

  Admissions with missing patient identification numbers were excluded (n = 2365 of 34,363 admissions, 6.88%).

• 3

  Hospitalizations were tabulated for each unique patient identifier.

• 4

  Patients with a known age of 13 years or more in 2004 were selected. There were 481 patients excluded due to age below 13 years, and 814 excluded for having an uncertain age. The final sample consisted of 1874 unique patient identifiers representing 10,704 hospital admissions.

 

Figure 1

As patients who were hospitalized more often were more likely to have inconsistent data, the exclusion for unknown and inconsistent age likely biased the findings by excluding more frequently hospitalized patients. Further post‐hoc analyses were conducted to gauge the extent of this bias, reported in Results, below.

Results

Course of New High Utilizers

Patients who were high utilizers in 2005 but not in 2004 were identified, and their hospital utilization from 2005 to 2007 was plotted (Fig. 2). The results are shown in Figure 1. Fifty‐five of the original 91 (60.44%) new high utilizers moderated in the following year, and 6.59% were known to have died in the hospital. Of the surviving 30 (32.97%) who did not moderate in the second year, 19 (65.3%) continued the high‐utilizing pattern into the third year, while 9 (16.36%) of those who moderated in year 2 returned to the high‐utilizing pattern in year 3. During this 3‐year period, 10 members (10.99%) of the initial group died in the hospital.

Figure 2

Discussion

A 66‐year‐old man presented to the emergency department with 3 weeks of progressive exertional dyspnea. He also reported a single episode of chest pain 1 day prior to admission.

Cardiac and pulmonary causes of dyspnea are the most common. Other causes include anemia or a neuromuscular process. Given the recent episode of chest pain, coronary ischemia, congestive heart failure, chronic obstructive pulmonary disease (COPD), pulmonary embolism, and pericardial effusion must be considered.

Up until 3 weeks ago, he had no exercise intolerance, and had been relatively active. He began noticing progressive dyspnea to the point where he had considerable difficulty walking up stairs, and performing minor household chores. He also complained of orthopnea and paroxysmal nocturnal dyspnea for the last 3 weeks.He denied chest pain at presentation, but 24 hours prior, he experienced one episode of sharp, left‐sided, nonradiating, nonpositional chest pain that occurred at rest. It lasted approximately 20 minutes and was not associated with diaphoresis, nausea, vomiting, or palpitations. He had never experienced chest discomfort prior to this episode. He denied fever, chills, cough, or wheezing.

Progressive dyspnea on exertion with associated orthopnea and paroxysmal nocturnal dyspnea is classically seen in patients with heart failure and is typically associated with left ventricular failure. However, paroxysmal nocturnal dyspnea and orthopnea are only moderately specific for heart failure. Orthopnea can also be seen in pericardial disease, and in numerous pulmonary diseases, including asthma, COPD, pulmonary hypertension, diaphragmatic weakness, pleural effusion, pulmonary embolism, and any apical lung process including lung cancer or pneumonia. Paroxysmal nocturnal dyspnea can be seen in many of the same disorders and can also be reported in obstructive sleep apnea.

His past medical history was remarkable for two episodes of syncope, occurring 5 and 3 years ago, both while working outside in warm weather. Neither was associated with chest pain, diaphoresis, palpitations, or post‐ictal symptoms. He was diagnosed with prostate cancer 8 years ago, and underwent 2 years of androgen‐deprivation therapy with goserelin along with local radiation therapy. Medications included subcutaneous goserelin every 3 months and daily omeprazole. He denied any other prescription, over‐the‐counter, or herbal medications. He reported a 50‐pack‐year history of smoking, but denied alcohol or illicit drug abuse. He denied any travel history or recent immobilization. He had no children, and there was no known history of heart disease in his family.

The past medical history of two episodes of likely exertional syncope is interesting, but the episodes were sporadic and in the distant past, arguing against a serious and ongoing process. Nonetheless, this history still raises the possibility of cardiac causes of syncope, especially causes such as hypertrophic obstructive cardiomyopathy or aortic stenosis which are classically associated with exertional syncope. Either of these two conditions can result in heart failure if untreated. The history of goserelin therapy does make the possibility of heart failure higher, as there has been an association reported between use of this drug and heart failure. His history of tobacco use is a risk factor for coronary artery disease (CAD) and COPD. An active cancer history is also a risk factor for thromboembolic disease, which remains a consideration.

On admission, his temperature was 36.9C, heart rate 94 bpm, respiratory rate 22 breaths per minute, blood pressure 200/108 mmHg, and oxygen saturation 93% breathing ambient air. He was a thin man in no acute distress. Cardiovascular examination was significant for normal first and second heart sounds, with a soft left‐sided S3; the point of maximal impulse was diffuse, but displaced laterally. His jugular venous pressure was estimated at 9 cm of H₂O while positioned at a 45‐degree angle. Rales were heard at the lung bases bilaterally. Abdominal exam was normal. His lower extremities were without edema. There were no focal neurological deficits appreciated. Skin examination was unremarkable.

His combination of physical exam findings strongly suggests heart failure, most likely related to a dilated cardiomyopathy and left ventricular dysfunction. The presence of a left‐sided S3 and rales, and the lack of markedly elevated central venous pressure and peripheral edema, suggest heart failure predominantly due to left ventricular dysfunction. Of note, he is very hypertensive. This would not be the typical finding with severely decompensated heart failure. It would be important to determine whether his elevated blood pressure is due to an acute, reversible cause (e.g., pain, dyspnea, anxiety) or whether cocaine use, psychotropic agents, rare causes such as catecholamine‐producing tumors, other neuroendocrine tumors or thyroid toxic states are at play. In addition, one might see hypertension early in the course of heart failure, from a left ventricular outflow obstructive etiology such as severe aortic stenosis or hypertrophic obstructive cardiomyopathy.

Laboratory evaluation revealed a white blood cell count of 8900/mm³, with a normal differential; hemoglobin was 13.9 g/dL; platelet count was 264,000/mm³. Serum electrolytes and liver enzymes were unremarkable, with serum creatinine 1.1 mg/dL and blood urea nitrogen 7 mg/dL. Serial cardiac troponin‐I levels drawn 8 hours apart were 0.04, 0.07, 0.08, and 0.04 ng/mL (normal <0.04). Brain natriuretic peptide was 1420 pg/mL (normal <100). Thyroid stimulating hormone was 1.19 uIU/mL (normal 0.34‐5.60). Chest radiography revealed mild cardiomegaly, with peripheral interstitial opacities in the mid and lower lobes bilaterally, with fluid within the minor fissure. A 12‐lead electrocardiogram (ECG) revealed normal sinus rhythm at 95 bpm with left anterior fascicular block; intraventricular conduction delay was present (QRS width 106 ms) and QS complexes were present in V1‐V3. In addition, there was a left atrial abnormality and voltage criteria for left ventricular hypertrophy with secondary T‐wave inversions laterally (Figure 1). No previous ECGs were available for comparison. A chest computed tomography scan with contrast showed no evidence of pulmonary embolus. It did show interlobular septal thickening and small bilateral pleural effusions, consistent with left ventricular dysfunction.

Figure 1

The patient's initial lab, imaging, and diagnostic work‐up continues to be consistent with the diagnosis of heart failure. The patient appears to have cardiomegaly and mild pulmonary edema by imaging. The etiology of heart failure remains unknown, but ischemia remains in the differential, given the mildly elevated troponins initially and the ECG findings of left anterior fascicular block and T‐wave inversions in the lateral leads. Left anterior fascicular block can be seen with ischemic heart disease (especially involving the left anterior descending coronary artery), hypertensive heart disease, valvular disease, and some infiltrative cardiac processes. The lateral T‐wave inversions are likely secondary to left ventricular hypertrophy (a so‐called strain pattern), rather than ischemia. Left ventricular hypertrophy is consistent with his hypertension, suggesting that it is chronic; his presentation may be due to hypertensive heart disease with new onset heart failure.

He was admitted to the hospital, and metoprolol, lisinopril, and intravenous furosemide were given. Transthoracic echocardiography demonstrated severe global hypokinesis with a left ventricular ejection fraction of 10%. There was no evidence of ventricular thrombus or valvular disease; however, prominent left ventricular trabeculation with deep recesses was noted (see Figure 2).

Figure 2

The echocardiographic findings of deep recesses and prominent left ventricular trabeculation are seen in only a few disorders. Sometimes these findings are thought to be due to hypertrophic obstructive cardiomyopathy. The deep trabeculations can be seen in patients with some forms of congenital heart disease associated with ventricular pressure overload during fetal development. The other cause is left ventricular noncompaction, a genetic cardiomyopathy which is becoming increasingly recognized. The disorder, along with causing heart failure, is associated with a high risk of ventricular thrombus and thromboembolic events, and a high risk of arrhythmias and sudden death. The overall prognosis appears to be poor, compared to some other cardiomyopathies. The imaging findings of left ventricular noncompaction are nearly pathognomonic, and experienced echocardiographers can usually make the diagnosis. Finally, left heart catheterization or noninvasive stress testing should be part of the workup to definitively exclude an ischemic cardiomyopathy, even in the setting of noncompaction, and especially given his recent history of chest pain.

A left heart catheterization with coronary arteriography demonstrated no angiographic evidence of obstructive coronary disease. Left ventriculography revealed severe global hypokinesis. The patient was diagnosed with left ventricular noncompaction.

The initial medical management centers upon the treatment of heart failure with a beta‐blocker, ACE‐inhibitor, and diuretics for fluid management. Patients with left ventricular noncompaction are at particularly high risk of both embolic events (thought due to propensity to develop left ventricular clots within the deep recesses of the endocardium) and sudden death from arrhythmias. Thus, anticoagulation with warfarin is often indicated and would be reasonable in this patient, given the extremely low ejection fraction. The patient does meet established criteria for primary prophylaxis of sudden death with an implantable cardioverter‐defibrillator in nonischemic cardiomyopathy (left ventricular ejection fraction <35% and New York Heart Association class II failure), and this would also be appropriate therapy as well, given the high‐risk profile of this patient population.

He was discharged in stable condition with a medical regimen consisting of diuretics, metoprolol, and lisinopril. Given the risk for thromboembolism, he was started on warfarin. On subsequent follow‐up, repeat echocardiogram revealed a persistently low left ventricular ejection fraction at 10%. Despite his marked improvement in exercise tolerance and overall well‐being after 4 months of treatment, his ejection fraction did not improve. As a result, he was evaluated and counseled for placement of an implantable cardioverter‐defibrillator, and received a dual‐chamber device shortly afterward.

COMMENTARY

Left ventricular noncompaction is a form of cardiomyopathy increasingly recognized in both pediatric and adult populations. The hallmark features are a pattern of prominent trabeculations and deep recesses in the left ventricular wall. During normal gestation, the myocardium compacts and matures while deep recesses evolve into capillary precursors of the coronary circulation. Left ventricular noncompaction may result from an arrest in this process, with cardiac myofibers failing to compact from their initial spongiform architecture into a developed endocardium.1 Restrictive relaxation from persistent trabeculae predisposes to diastolic dysfunction, while systolic dysfunction may be related to subendocardial hypoperfusion and mechanical dyssynchrony between compacted and noncompacted myocardium.2

Differentiation of left ventricular noncompaction from other cardiomyopathies, based on history and physical examination alone, is essentially impossible. There is high variability and lack of specificity in both clinical profile and onset of symptoms. Electrocardiographic findings are also nonspecific, and the diagnosis typically becomes evident only with transthoracic echocardiography. Current diagnostic criteria include: 1) absence of coexisting cardiac abnormalities; 2) a two‐layer structure with >2:1 ratio of noncompacted to compacted myocardium; 3) predominant involvement of the apical segment of myocardium; and 4) deep intertrabecular recesses demonstrated on Doppler imaging.2, 3 Although echocardiography remains the standard in clinical practice, cardiac magnetic resonance imaging is being increasingly employed as well.4

With more awareness of the disease and the development of higher resolution imaging, the reported incidence has risen. In one single‐center study performed at a heart failure/transplant clinic, 3% of 960 patients referred to heart failure clinic were diagnosed with left ventricular noncompaction, a prevalence similar to hypertensive disease and hypertrophic cardiomyopathy.5 In another community‐hospitalbased study of 4929 adult patients referred for echocardiography, 3.7% of those with systolic dysfunction were diagnosed with noncompaction.6

Left ventricular noncompaction is considered a genetic cardiomyopathy; a family history of heart failure is often present.7 Despite its congenital origin and genetic involvement,2 it is unclear why symptoms may first present at an advanced age. Chest pain and shortness of breath are common complaints, and approximately 62% of patients will have congestive heart failure at presentation.8

Tachyarrhythmia and ventricular tachycardia are commonly seen, as are systemic embolic events and pulmonary embolism. Significant predictors of death include New York Heart Association class III‐IV, sustained ventricular arrhythmias, and increased left atrial size.9

Management is focused on the treatment of arrhythmias, heart failure, and thromboembolic events. The use of standard medical therapy for heart failure (including ACE‐inhibitors and beta‐blockers) is not based on large‐scale studies, yet remains the cornerstone of therapy. An implantable cardioverter‐defibrillator is indicated after hemodynamically compromising sustained ventricular tachycardia or aborted sudden cardiac death, but there are no guidelines for primary prophylaxis outside of patients with heart failure and a depressed ejection fraction.10 Cardiac resynchronization therapy has been successful in some patients with isolated left ventricular noncompaction. Long‐term oral anticoagulation is recommended, especially when impaired left ventricular function, thrombi, or atrial fibrillation have been documented. Patients with left ventricular dysfunction in concert with left ventricular noncompaction are at 10% higher risk for embolic complications when compared to those without noncompaction.11 Familial screening with echocardiography is indicated once the diagnosis has been made.2

In this Clinical Care Conundrum, we describe a rare but increasingly recognized condition, and highlight the importance of delineating the underlying cause of cardiomyopathy when possible. Treatment of heart failure in the hospital setting is sometimes more focused on initiation of diuresis and further stabilization of the patient, and less focused on elucidation of the etiology. While recognition of left ventricular failure led to early treatment with standard therapy in this case, identification of the underlying cause allowed for targeted interventions directed at cardiac arrhythmias, embolic events, and familial screening. Of note, the discussant was careful not to let the prior history of syncopal events distract him from the central issues in this case.

This case also serves as a reminder that congenital anomalies should remain on the differential diagnosis when evaluating new complaints in adult patients. The discussant approached the presentation of new‐onset left ventricular dysfunction in a thorough manner, weighing the likelihood of ischemic and nonischemic causes in the context of the history and physical examination. Careful consideration of the patient's new clinical manifestationscoupled with characteristic echocardiographic findings and normal coronary anatomysolidified the diagnosis. By developing a broad differential, the discussant and clinical team arrived at a diagnosis that for this 66‐year‐old gentleman was a lifetime in the making.

A 66‐year‐old man presented to the emergency department with 3 weeks of progressive exertional dyspnea. He also reported a single episode of chest pain 1 day prior to admission.

Cardiac and pulmonary causes of dyspnea are the most common. Other causes include anemia or a neuromuscular process. Given the recent episode of chest pain, coronary ischemia, congestive heart failure, chronic obstructive pulmonary disease (COPD), pulmonary embolism, and pericardial effusion must be considered.

Up until 3 weeks ago, he had no exercise intolerance, and had been relatively active. He began noticing progressive dyspnea to the point where he had considerable difficulty walking up stairs, and performing minor household chores. He also complained of orthopnea and paroxysmal nocturnal dyspnea for the last 3 weeks.He denied chest pain at presentation, but 24 hours prior, he experienced one episode of sharp, left‐sided, nonradiating, nonpositional chest pain that occurred at rest. It lasted approximately 20 minutes and was not associated with diaphoresis, nausea, vomiting, or palpitations. He had never experienced chest discomfort prior to this episode. He denied fever, chills, cough, or wheezing.

Progressive dyspnea on exertion with associated orthopnea and paroxysmal nocturnal dyspnea is classically seen in patients with heart failure and is typically associated with left ventricular failure. However, paroxysmal nocturnal dyspnea and orthopnea are only moderately specific for heart failure. Orthopnea can also be seen in pericardial disease, and in numerous pulmonary diseases, including asthma, COPD, pulmonary hypertension, diaphragmatic weakness, pleural effusion, pulmonary embolism, and any apical lung process including lung cancer or pneumonia. Paroxysmal nocturnal dyspnea can be seen in many of the same disorders and can also be reported in obstructive sleep apnea.

His past medical history was remarkable for two episodes of syncope, occurring 5 and 3 years ago, both while working outside in warm weather. Neither was associated with chest pain, diaphoresis, palpitations, or post‐ictal symptoms. He was diagnosed with prostate cancer 8 years ago, and underwent 2 years of androgen‐deprivation therapy with goserelin along with local radiation therapy. Medications included subcutaneous goserelin every 3 months and daily omeprazole. He denied any other prescription, over‐the‐counter, or herbal medications. He reported a 50‐pack‐year history of smoking, but denied alcohol or illicit drug abuse. He denied any travel history or recent immobilization. He had no children, and there was no known history of heart disease in his family.

The past medical history of two episodes of likely exertional syncope is interesting, but the episodes were sporadic and in the distant past, arguing against a serious and ongoing process. Nonetheless, this history still raises the possibility of cardiac causes of syncope, especially causes such as hypertrophic obstructive cardiomyopathy or aortic stenosis which are classically associated with exertional syncope. Either of these two conditions can result in heart failure if untreated. The history of goserelin therapy does make the possibility of heart failure higher, as there has been an association reported between use of this drug and heart failure. His history of tobacco use is a risk factor for coronary artery disease (CAD) and COPD. An active cancer history is also a risk factor for thromboembolic disease, which remains a consideration.

On admission, his temperature was 36.9C, heart rate 94 bpm, respiratory rate 22 breaths per minute, blood pressure 200/108 mmHg, and oxygen saturation 93% breathing ambient air. He was a thin man in no acute distress. Cardiovascular examination was significant for normal first and second heart sounds, with a soft left‐sided S3; the point of maximal impulse was diffuse, but displaced laterally. His jugular venous pressure was estimated at 9 cm of H₂O while positioned at a 45‐degree angle. Rales were heard at the lung bases bilaterally. Abdominal exam was normal. His lower extremities were without edema. There were no focal neurological deficits appreciated. Skin examination was unremarkable.

His combination of physical exam findings strongly suggests heart failure, most likely related to a dilated cardiomyopathy and left ventricular dysfunction. The presence of a left‐sided S3 and rales, and the lack of markedly elevated central venous pressure and peripheral edema, suggest heart failure predominantly due to left ventricular dysfunction. Of note, he is very hypertensive. This would not be the typical finding with severely decompensated heart failure. It would be important to determine whether his elevated blood pressure is due to an acute, reversible cause (e.g., pain, dyspnea, anxiety) or whether cocaine use, psychotropic agents, rare causes such as catecholamine‐producing tumors, other neuroendocrine tumors or thyroid toxic states are at play. In addition, one might see hypertension early in the course of heart failure, from a left ventricular outflow obstructive etiology such as severe aortic stenosis or hypertrophic obstructive cardiomyopathy.

Laboratory evaluation revealed a white blood cell count of 8900/mm³, with a normal differential; hemoglobin was 13.9 g/dL; platelet count was 264,000/mm³. Serum electrolytes and liver enzymes were unremarkable, with serum creatinine 1.1 mg/dL and blood urea nitrogen 7 mg/dL. Serial cardiac troponin‐I levels drawn 8 hours apart were 0.04, 0.07, 0.08, and 0.04 ng/mL (normal <0.04). Brain natriuretic peptide was 1420 pg/mL (normal <100). Thyroid stimulating hormone was 1.19 uIU/mL (normal 0.34‐5.60). Chest radiography revealed mild cardiomegaly, with peripheral interstitial opacities in the mid and lower lobes bilaterally, with fluid within the minor fissure. A 12‐lead electrocardiogram (ECG) revealed normal sinus rhythm at 95 bpm with left anterior fascicular block; intraventricular conduction delay was present (QRS width 106 ms) and QS complexes were present in V1‐V3. In addition, there was a left atrial abnormality and voltage criteria for left ventricular hypertrophy with secondary T‐wave inversions laterally (Figure 1). No previous ECGs were available for comparison. A chest computed tomography scan with contrast showed no evidence of pulmonary embolus. It did show interlobular septal thickening and small bilateral pleural effusions, consistent with left ventricular dysfunction.

Figure 1

The patient's initial lab, imaging, and diagnostic work‐up continues to be consistent with the diagnosis of heart failure. The patient appears to have cardiomegaly and mild pulmonary edema by imaging. The etiology of heart failure remains unknown, but ischemia remains in the differential, given the mildly elevated troponins initially and the ECG findings of left anterior fascicular block and T‐wave inversions in the lateral leads. Left anterior fascicular block can be seen with ischemic heart disease (especially involving the left anterior descending coronary artery), hypertensive heart disease, valvular disease, and some infiltrative cardiac processes. The lateral T‐wave inversions are likely secondary to left ventricular hypertrophy (a so‐called strain pattern), rather than ischemia. Left ventricular hypertrophy is consistent with his hypertension, suggesting that it is chronic; his presentation may be due to hypertensive heart disease with new onset heart failure.

He was admitted to the hospital, and metoprolol, lisinopril, and intravenous furosemide were given. Transthoracic echocardiography demonstrated severe global hypokinesis with a left ventricular ejection fraction of 10%. There was no evidence of ventricular thrombus or valvular disease; however, prominent left ventricular trabeculation with deep recesses was noted (see Figure 2).

Figure 2

The echocardiographic findings of deep recesses and prominent left ventricular trabeculation are seen in only a few disorders. Sometimes these findings are thought to be due to hypertrophic obstructive cardiomyopathy. The deep trabeculations can be seen in patients with some forms of congenital heart disease associated with ventricular pressure overload during fetal development. The other cause is left ventricular noncompaction, a genetic cardiomyopathy which is becoming increasingly recognized. The disorder, along with causing heart failure, is associated with a high risk of ventricular thrombus and thromboembolic events, and a high risk of arrhythmias and sudden death. The overall prognosis appears to be poor, compared to some other cardiomyopathies. The imaging findings of left ventricular noncompaction are nearly pathognomonic, and experienced echocardiographers can usually make the diagnosis. Finally, left heart catheterization or noninvasive stress testing should be part of the workup to definitively exclude an ischemic cardiomyopathy, even in the setting of noncompaction, and especially given his recent history of chest pain.

A left heart catheterization with coronary arteriography demonstrated no angiographic evidence of obstructive coronary disease. Left ventriculography revealed severe global hypokinesis. The patient was diagnosed with left ventricular noncompaction.

The initial medical management centers upon the treatment of heart failure with a beta‐blocker, ACE‐inhibitor, and diuretics for fluid management. Patients with left ventricular noncompaction are at particularly high risk of both embolic events (thought due to propensity to develop left ventricular clots within the deep recesses of the endocardium) and sudden death from arrhythmias. Thus, anticoagulation with warfarin is often indicated and would be reasonable in this patient, given the extremely low ejection fraction. The patient does meet established criteria for primary prophylaxis of sudden death with an implantable cardioverter‐defibrillator in nonischemic cardiomyopathy (left ventricular ejection fraction <35% and New York Heart Association class II failure), and this would also be appropriate therapy as well, given the high‐risk profile of this patient population.

He was discharged in stable condition with a medical regimen consisting of diuretics, metoprolol, and lisinopril. Given the risk for thromboembolism, he was started on warfarin. On subsequent follow‐up, repeat echocardiogram revealed a persistently low left ventricular ejection fraction at 10%. Despite his marked improvement in exercise tolerance and overall well‐being after 4 months of treatment, his ejection fraction did not improve. As a result, he was evaluated and counseled for placement of an implantable cardioverter‐defibrillator, and received a dual‐chamber device shortly afterward.

COMMENTARY

Left ventricular noncompaction is a form of cardiomyopathy increasingly recognized in both pediatric and adult populations. The hallmark features are a pattern of prominent trabeculations and deep recesses in the left ventricular wall. During normal gestation, the myocardium compacts and matures while deep recesses evolve into capillary precursors of the coronary circulation. Left ventricular noncompaction may result from an arrest in this process, with cardiac myofibers failing to compact from their initial spongiform architecture into a developed endocardium.1 Restrictive relaxation from persistent trabeculae predisposes to diastolic dysfunction, while systolic dysfunction may be related to subendocardial hypoperfusion and mechanical dyssynchrony between compacted and noncompacted myocardium.2

Differentiation of left ventricular noncompaction from other cardiomyopathies, based on history and physical examination alone, is essentially impossible. There is high variability and lack of specificity in both clinical profile and onset of symptoms. Electrocardiographic findings are also nonspecific, and the diagnosis typically becomes evident only with transthoracic echocardiography. Current diagnostic criteria include: 1) absence of coexisting cardiac abnormalities; 2) a two‐layer structure with >2:1 ratio of noncompacted to compacted myocardium; 3) predominant involvement of the apical segment of myocardium; and 4) deep intertrabecular recesses demonstrated on Doppler imaging.2, 3 Although echocardiography remains the standard in clinical practice, cardiac magnetic resonance imaging is being increasingly employed as well.4

With more awareness of the disease and the development of higher resolution imaging, the reported incidence has risen. In one single‐center study performed at a heart failure/transplant clinic, 3% of 960 patients referred to heart failure clinic were diagnosed with left ventricular noncompaction, a prevalence similar to hypertensive disease and hypertrophic cardiomyopathy.5 In another community‐hospitalbased study of 4929 adult patients referred for echocardiography, 3.7% of those with systolic dysfunction were diagnosed with noncompaction.6

Left ventricular noncompaction is considered a genetic cardiomyopathy; a family history of heart failure is often present.7 Despite its congenital origin and genetic involvement,2 it is unclear why symptoms may first present at an advanced age. Chest pain and shortness of breath are common complaints, and approximately 62% of patients will have congestive heart failure at presentation.8

Tachyarrhythmia and ventricular tachycardia are commonly seen, as are systemic embolic events and pulmonary embolism. Significant predictors of death include New York Heart Association class III‐IV, sustained ventricular arrhythmias, and increased left atrial size.9

Management is focused on the treatment of arrhythmias, heart failure, and thromboembolic events. The use of standard medical therapy for heart failure (including ACE‐inhibitors and beta‐blockers) is not based on large‐scale studies, yet remains the cornerstone of therapy. An implantable cardioverter‐defibrillator is indicated after hemodynamically compromising sustained ventricular tachycardia or aborted sudden cardiac death, but there are no guidelines for primary prophylaxis outside of patients with heart failure and a depressed ejection fraction.10 Cardiac resynchronization therapy has been successful in some patients with isolated left ventricular noncompaction. Long‐term oral anticoagulation is recommended, especially when impaired left ventricular function, thrombi, or atrial fibrillation have been documented. Patients with left ventricular dysfunction in concert with left ventricular noncompaction are at 10% higher risk for embolic complications when compared to those without noncompaction.11 Familial screening with echocardiography is indicated once the diagnosis has been made.2

In this Clinical Care Conundrum, we describe a rare but increasingly recognized condition, and highlight the importance of delineating the underlying cause of cardiomyopathy when possible. Treatment of heart failure in the hospital setting is sometimes more focused on initiation of diuresis and further stabilization of the patient, and less focused on elucidation of the etiology. While recognition of left ventricular failure led to early treatment with standard therapy in this case, identification of the underlying cause allowed for targeted interventions directed at cardiac arrhythmias, embolic events, and familial screening. Of note, the discussant was careful not to let the prior history of syncopal events distract him from the central issues in this case.

This case also serves as a reminder that congenital anomalies should remain on the differential diagnosis when evaluating new complaints in adult patients. The discussant approached the presentation of new‐onset left ventricular dysfunction in a thorough manner, weighing the likelihood of ischemic and nonischemic causes in the context of the history and physical examination. Careful consideration of the patient's new clinical manifestationscoupled with characteristic echocardiographic findings and normal coronary anatomysolidified the diagnosis. By developing a broad differential, the discussant and clinical team arrived at a diagnosis that for this 66‐year‐old gentleman was a lifetime in the making.

A 66‐year‐old man presented to the emergency department with 3 weeks of progressive exertional dyspnea. He also reported a single episode of chest pain 1 day prior to admission.

Cardiac and pulmonary causes of dyspnea are the most common. Other causes include anemia or a neuromuscular process. Given the recent episode of chest pain, coronary ischemia, congestive heart failure, chronic obstructive pulmonary disease (COPD), pulmonary embolism, and pericardial effusion must be considered.

Up until 3 weeks ago, he had no exercise intolerance, and had been relatively active. He began noticing progressive dyspnea to the point where he had considerable difficulty walking up stairs, and performing minor household chores. He also complained of orthopnea and paroxysmal nocturnal dyspnea for the last 3 weeks.He denied chest pain at presentation, but 24 hours prior, he experienced one episode of sharp, left‐sided, nonradiating, nonpositional chest pain that occurred at rest. It lasted approximately 20 minutes and was not associated with diaphoresis, nausea, vomiting, or palpitations. He had never experienced chest discomfort prior to this episode. He denied fever, chills, cough, or wheezing.

Progressive dyspnea on exertion with associated orthopnea and paroxysmal nocturnal dyspnea is classically seen in patients with heart failure and is typically associated with left ventricular failure. However, paroxysmal nocturnal dyspnea and orthopnea are only moderately specific for heart failure. Orthopnea can also be seen in pericardial disease, and in numerous pulmonary diseases, including asthma, COPD, pulmonary hypertension, diaphragmatic weakness, pleural effusion, pulmonary embolism, and any apical lung process including lung cancer or pneumonia. Paroxysmal nocturnal dyspnea can be seen in many of the same disorders and can also be reported in obstructive sleep apnea.

His past medical history was remarkable for two episodes of syncope, occurring 5 and 3 years ago, both while working outside in warm weather. Neither was associated with chest pain, diaphoresis, palpitations, or post‐ictal symptoms. He was diagnosed with prostate cancer 8 years ago, and underwent 2 years of androgen‐deprivation therapy with goserelin along with local radiation therapy. Medications included subcutaneous goserelin every 3 months and daily omeprazole. He denied any other prescription, over‐the‐counter, or herbal medications. He reported a 50‐pack‐year history of smoking, but denied alcohol or illicit drug abuse. He denied any travel history or recent immobilization. He had no children, and there was no known history of heart disease in his family.

The past medical history of two episodes of likely exertional syncope is interesting, but the episodes were sporadic and in the distant past, arguing against a serious and ongoing process. Nonetheless, this history still raises the possibility of cardiac causes of syncope, especially causes such as hypertrophic obstructive cardiomyopathy or aortic stenosis which are classically associated with exertional syncope. Either of these two conditions can result in heart failure if untreated. The history of goserelin therapy does make the possibility of heart failure higher, as there has been an association reported between use of this drug and heart failure. His history of tobacco use is a risk factor for coronary artery disease (CAD) and COPD. An active cancer history is also a risk factor for thromboembolic disease, which remains a consideration.

On admission, his temperature was 36.9C, heart rate 94 bpm, respiratory rate 22 breaths per minute, blood pressure 200/108 mmHg, and oxygen saturation 93% breathing ambient air. He was a thin man in no acute distress. Cardiovascular examination was significant for normal first and second heart sounds, with a soft left‐sided S3; the point of maximal impulse was diffuse, but displaced laterally. His jugular venous pressure was estimated at 9 cm of H₂O while positioned at a 45‐degree angle. Rales were heard at the lung bases bilaterally. Abdominal exam was normal. His lower extremities were without edema. There were no focal neurological deficits appreciated. Skin examination was unremarkable.

His combination of physical exam findings strongly suggests heart failure, most likely related to a dilated cardiomyopathy and left ventricular dysfunction. The presence of a left‐sided S3 and rales, and the lack of markedly elevated central venous pressure and peripheral edema, suggest heart failure predominantly due to left ventricular dysfunction. Of note, he is very hypertensive. This would not be the typical finding with severely decompensated heart failure. It would be important to determine whether his elevated blood pressure is due to an acute, reversible cause (e.g., pain, dyspnea, anxiety) or whether cocaine use, psychotropic agents, rare causes such as catecholamine‐producing tumors, other neuroendocrine tumors or thyroid toxic states are at play. In addition, one might see hypertension early in the course of heart failure, from a left ventricular outflow obstructive etiology such as severe aortic stenosis or hypertrophic obstructive cardiomyopathy.

Laboratory evaluation revealed a white blood cell count of 8900/mm³, with a normal differential; hemoglobin was 13.9 g/dL; platelet count was 264,000/mm³. Serum electrolytes and liver enzymes were unremarkable, with serum creatinine 1.1 mg/dL and blood urea nitrogen 7 mg/dL. Serial cardiac troponin‐I levels drawn 8 hours apart were 0.04, 0.07, 0.08, and 0.04 ng/mL (normal <0.04). Brain natriuretic peptide was 1420 pg/mL (normal <100). Thyroid stimulating hormone was 1.19 uIU/mL (normal 0.34‐5.60). Chest radiography revealed mild cardiomegaly, with peripheral interstitial opacities in the mid and lower lobes bilaterally, with fluid within the minor fissure. A 12‐lead electrocardiogram (ECG) revealed normal sinus rhythm at 95 bpm with left anterior fascicular block; intraventricular conduction delay was present (QRS width 106 ms) and QS complexes were present in V1‐V3. In addition, there was a left atrial abnormality and voltage criteria for left ventricular hypertrophy with secondary T‐wave inversions laterally (Figure 1). No previous ECGs were available for comparison. A chest computed tomography scan with contrast showed no evidence of pulmonary embolus. It did show interlobular septal thickening and small bilateral pleural effusions, consistent with left ventricular dysfunction.

Figure 1

The patient's initial lab, imaging, and diagnostic work‐up continues to be consistent with the diagnosis of heart failure. The patient appears to have cardiomegaly and mild pulmonary edema by imaging. The etiology of heart failure remains unknown, but ischemia remains in the differential, given the mildly elevated troponins initially and the ECG findings of left anterior fascicular block and T‐wave inversions in the lateral leads. Left anterior fascicular block can be seen with ischemic heart disease (especially involving the left anterior descending coronary artery), hypertensive heart disease, valvular disease, and some infiltrative cardiac processes. The lateral T‐wave inversions are likely secondary to left ventricular hypertrophy (a so‐called strain pattern), rather than ischemia. Left ventricular hypertrophy is consistent with his hypertension, suggesting that it is chronic; his presentation may be due to hypertensive heart disease with new onset heart failure.

He was admitted to the hospital, and metoprolol, lisinopril, and intravenous furosemide were given. Transthoracic echocardiography demonstrated severe global hypokinesis with a left ventricular ejection fraction of 10%. There was no evidence of ventricular thrombus or valvular disease; however, prominent left ventricular trabeculation with deep recesses was noted (see Figure 2).

Figure 2

The echocardiographic findings of deep recesses and prominent left ventricular trabeculation are seen in only a few disorders. Sometimes these findings are thought to be due to hypertrophic obstructive cardiomyopathy. The deep trabeculations can be seen in patients with some forms of congenital heart disease associated with ventricular pressure overload during fetal development. The other cause is left ventricular noncompaction, a genetic cardiomyopathy which is becoming increasingly recognized. The disorder, along with causing heart failure, is associated with a high risk of ventricular thrombus and thromboembolic events, and a high risk of arrhythmias and sudden death. The overall prognosis appears to be poor, compared to some other cardiomyopathies. The imaging findings of left ventricular noncompaction are nearly pathognomonic, and experienced echocardiographers can usually make the diagnosis. Finally, left heart catheterization or noninvasive stress testing should be part of the workup to definitively exclude an ischemic cardiomyopathy, even in the setting of noncompaction, and especially given his recent history of chest pain.

A left heart catheterization with coronary arteriography demonstrated no angiographic evidence of obstructive coronary disease. Left ventriculography revealed severe global hypokinesis. The patient was diagnosed with left ventricular noncompaction.

The initial medical management centers upon the treatment of heart failure with a beta‐blocker, ACE‐inhibitor, and diuretics for fluid management. Patients with left ventricular noncompaction are at particularly high risk of both embolic events (thought due to propensity to develop left ventricular clots within the deep recesses of the endocardium) and sudden death from arrhythmias. Thus, anticoagulation with warfarin is often indicated and would be reasonable in this patient, given the extremely low ejection fraction. The patient does meet established criteria for primary prophylaxis of sudden death with an implantable cardioverter‐defibrillator in nonischemic cardiomyopathy (left ventricular ejection fraction <35% and New York Heart Association class II failure), and this would also be appropriate therapy as well, given the high‐risk profile of this patient population.

He was discharged in stable condition with a medical regimen consisting of diuretics, metoprolol, and lisinopril. Given the risk for thromboembolism, he was started on warfarin. On subsequent follow‐up, repeat echocardiogram revealed a persistently low left ventricular ejection fraction at 10%. Despite his marked improvement in exercise tolerance and overall well‐being after 4 months of treatment, his ejection fraction did not improve. As a result, he was evaluated and counseled for placement of an implantable cardioverter‐defibrillator, and received a dual‐chamber device shortly afterward.

COMMENTARY

Left ventricular noncompaction is a form of cardiomyopathy increasingly recognized in both pediatric and adult populations. The hallmark features are a pattern of prominent trabeculations and deep recesses in the left ventricular wall. During normal gestation, the myocardium compacts and matures while deep recesses evolve into capillary precursors of the coronary circulation. Left ventricular noncompaction may result from an arrest in this process, with cardiac myofibers failing to compact from their initial spongiform architecture into a developed endocardium.1 Restrictive relaxation from persistent trabeculae predisposes to diastolic dysfunction, while systolic dysfunction may be related to subendocardial hypoperfusion and mechanical dyssynchrony between compacted and noncompacted myocardium.2

Differentiation of left ventricular noncompaction from other cardiomyopathies, based on history and physical examination alone, is essentially impossible. There is high variability and lack of specificity in both clinical profile and onset of symptoms. Electrocardiographic findings are also nonspecific, and the diagnosis typically becomes evident only with transthoracic echocardiography. Current diagnostic criteria include: 1) absence of coexisting cardiac abnormalities; 2) a two‐layer structure with >2:1 ratio of noncompacted to compacted myocardium; 3) predominant involvement of the apical segment of myocardium; and 4) deep intertrabecular recesses demonstrated on Doppler imaging.2, 3 Although echocardiography remains the standard in clinical practice, cardiac magnetic resonance imaging is being increasingly employed as well.4

With more awareness of the disease and the development of higher resolution imaging, the reported incidence has risen. In one single‐center study performed at a heart failure/transplant clinic, 3% of 960 patients referred to heart failure clinic were diagnosed with left ventricular noncompaction, a prevalence similar to hypertensive disease and hypertrophic cardiomyopathy.5 In another community‐hospitalbased study of 4929 adult patients referred for echocardiography, 3.7% of those with systolic dysfunction were diagnosed with noncompaction.6

Left ventricular noncompaction is considered a genetic cardiomyopathy; a family history of heart failure is often present.7 Despite its congenital origin and genetic involvement,2 it is unclear why symptoms may first present at an advanced age. Chest pain and shortness of breath are common complaints, and approximately 62% of patients will have congestive heart failure at presentation.8

Tachyarrhythmia and ventricular tachycardia are commonly seen, as are systemic embolic events and pulmonary embolism. Significant predictors of death include New York Heart Association class III‐IV, sustained ventricular arrhythmias, and increased left atrial size.9

Management is focused on the treatment of arrhythmias, heart failure, and thromboembolic events. The use of standard medical therapy for heart failure (including ACE‐inhibitors and beta‐blockers) is not based on large‐scale studies, yet remains the cornerstone of therapy. An implantable cardioverter‐defibrillator is indicated after hemodynamically compromising sustained ventricular tachycardia or aborted sudden cardiac death, but there are no guidelines for primary prophylaxis outside of patients with heart failure and a depressed ejection fraction.10 Cardiac resynchronization therapy has been successful in some patients with isolated left ventricular noncompaction. Long‐term oral anticoagulation is recommended, especially when impaired left ventricular function, thrombi, or atrial fibrillation have been documented. Patients with left ventricular dysfunction in concert with left ventricular noncompaction are at 10% higher risk for embolic complications when compared to those without noncompaction.11 Familial screening with echocardiography is indicated once the diagnosis has been made.2

In this Clinical Care Conundrum, we describe a rare but increasingly recognized condition, and highlight the importance of delineating the underlying cause of cardiomyopathy when possible. Treatment of heart failure in the hospital setting is sometimes more focused on initiation of diuresis and further stabilization of the patient, and less focused on elucidation of the etiology. While recognition of left ventricular failure led to early treatment with standard therapy in this case, identification of the underlying cause allowed for targeted interventions directed at cardiac arrhythmias, embolic events, and familial screening. Of note, the discussant was careful not to let the prior history of syncopal events distract him from the central issues in this case.

This case also serves as a reminder that congenital anomalies should remain on the differential diagnosis when evaluating new complaints in adult patients. The discussant approached the presentation of new‐onset left ventricular dysfunction in a thorough manner, weighing the likelihood of ischemic and nonischemic causes in the context of the history and physical examination. Careful consideration of the patient's new clinical manifestationscoupled with characteristic echocardiographic findings and normal coronary anatomysolidified the diagnosis. By developing a broad differential, the discussant and clinical team arrived at a diagnosis that for this 66‐year‐old gentleman was a lifetime in the making.

Prior studies have identified risk factors for increased severity of illness, readmission, or prolonged length of stay (LOS) in infants admitted with bronchiolitis.123 These risk factors include birth‐related factors (prematurity, birth within six months of respiratory syncytial virus [RSV] season, discharge from the neonatal intensive care unit during winter, multiple birth infant), environmental factors (day care attendance, school‐age siblings, smoke exposure), and underlying diseases (chronic lung disease and other pulmonary conditions, failure to thrive (FTT), congenital heart disease, immunologic disorders, and neuromuscular disease).123 Additional risk factors occurring during the bronchiolitis course that have been associated with prolonged hospital course are mechanical ventilation, intensive care unit (ICU) admission, hypoxia on admission, apnea, feeding problems, and duration of supplemental oxygen.2123

Having a reliable model to identify infants at high risk for prolonged LOS early in the course of an admission would be helpful, both for clinical care and for studies of interventions designed to reduce LOS. Prior attempts at developing a model have yielded mixed results. The Michigan Logistic Regression Model displayed excellent predictive ability with an area under the receiver‐operator curve (ROC) of 0.88 using variables including prematurity, FTT, pulmonary disease, other comorbid diseases, and early mechanical ventilation.21 However, when applied to another patient population it did not perform as well. The Rotterdam Model using the variables of weight and supplemental oxygen had an ROC of 0.65.24

Prior prediction models have focused more on birth‐ and disease‐related risk factors than on hospital course factors, particularly common clinical assessments including respiratory status and caloric intake. An additional limitation of prior models is some loss of ability to study the interaction between various predictor variables when using multivariate regression methods.

Our aims were: 1) to study the associations of various clinical markers identifiable during the first two days of the hospital admission with LOS; and 2) to develop a LOS prediction model, using both previously identified risk factors and more detailed clinical data from the first two days of the hospital admission.

MATERIALS AND METHODS

RESULTS

Three hundred forty‐seven infants were admitted during the 20042005 bronchiolitis season, with 273 placed in the bronchiolitis treatment protocol. The charts of these 273 patients were reviewed. One was excluded because of gastrostomy tube placement during the admission. Of the remaining 272 patients, 47 (17.3%) had a LOS 108 hours. The median LOS was 59 hours (range 10334 hours). Two patients had missing data for caloric intake on days 1 and 2; and 23 patients did not have height obtained, therefore their Waterlow classification could not be determined. Historical details concerning fever, congestion, cough, and diminished caloric intake preceding admission were variably reported, resulting in a smaller sample size for these baseline characteristics as described in Table 1.

Recursive Partitioning Model

Figure 1 depicts the recursive partitioning model that best predicted LOS. Five variables were selected by the recursive partitioning model. Selected variables, in order of appearance (variable importance is related to order of appearance, ie, most important variable is first), were: hours of supplemental oxygen, maximum respiratory rate, minimum supplemental oxygen use, gestation, and kilocalories (kcal)/kilogram (kg)/day consumed. The characteristics of this model were: ROC 0.89 and 0.72 for the learning and test trees, respectively; sensitivity, 0.85; and specificity, 0.82

Figure 1

Infants predicted as having a short LOS had three distinct profiles labeled S1, S2, and S3 in Figure 1. The S1 group required 6.5 hours of oxygen. The S2 group required >6.5 hours of oxygen, but had a maximum respiratory rate 49. The S3 group required >6.5 hours of oxygen, had a maximum respiratory rate >49, but were >36.5 week gestation, and consumed >23.5 kcal/kg/day.

Infants predicted as having a long LOS had three distinct profiles labeled L1, L2, and L3 in Figure 1. The L1 group required >6.5 hours of supplemental oxygen, had a maximum respiratory rate >49, and required some level of oxygen support the entire day. The L2 group required >6.5 hours of supplemental oxygen, had a maximum respiratory rate >49, were on room air some portion of the day, but had a gestation 36.5 weeks. The L3 group required >6.5 hours of supplemental oxygen, had a maximum respiratory rate >49, were on room air for some portion of the day, had a gestation >36.5 weeks, but consumed 23.5 kcal/kg/day.

Table 3 compares the performance of our model (the Milwaukee Model), the Michigan Model, and the Rotterdam Model in predicting LOS group. Overall, the Milwaukee Model had the highest ROC 0.89/0.72 for the learning and test trees. All three models had good sensitivity (Milwaukee, 85%; Michigan, 85%), with the Rotterdam Model having the highest (98%). The Milwaukee Model also had good specificity (82%), while the Michigan and Rotterdam Models were less specific (46% and 44%).

Table 3.Comparison of Three Models for Predicting LOS in Infants Admitted with Bronchiolitis

Model	Priors*		Sensitivity	Specificity	Learning Tree ROC	Test Tree ROC
	Long LOS	Short LOS
Abbreviations: LOS, length of stay; ROC, receiver‐operator curve. * The priors (or weighting system) selected for model by Salford Systems' CART software.
Michigan	0.5	0.5	0.85	0.46	0.69	0.56
Rotterdam	0.5	0.5	0.98	0.44	0.73	0.61
Milwaukee	0.5	0.5	0.85	0.82	0.89	0.72

DISCUSSION

We confirmed several previously recognized risk factors for prolonged LOS, including: apnea, at least part of the hospital stay in ICU, use of CPAP, mechanical ventilation, and prematurity. However, most patients admitted with bronchiolitis do not have these risk factors. The major contribution of this study is the evaluation of factors applicable to all patients admitted with bronchiolitis, and a more in‐depth analysis of clinical assessments performed on hospital days 1 and 2 than had been previously reported.

We did find strong associations between a number of clinical assessments and LOS. While some were apparent on day 1 of the admission, the number and degree of clinical differences between infants destined for a short vs prolonged stay were more apparent on hospital day 2. On this day, there were significant differences between the groups in the length and amount of oxygen received, oxygen saturation, maximum respiratory rate, respiratory scores, nasopharyngeal suctioning need, and caloric intake. Interestingly, it was noted the prolonged stay group had overall worsening or a lack of improvement in several clinical markers from day 1 to day 2, in areas where the short stay group showed improvements.

To our knowledge, the Milwaukee Model is the first bronchiolitis LOS prediction model to incorporate several clinical markers occurring early in the hospital stay. These clinical markers were found to be more effective predictors of LOS group in our study population than some of the traditional birth‐ and disease‐related risk factors previously reported. The model highlighted some important interactions among variables, and identified specific profiles of patients likely to have a short or prolonged LOS based on their day 2 clinical status.

The short LOS groups all shared one of the following three features: 1) low duration of oxygen use; 2) absence of tachypnea (tachypnea defined as a respiratory rate >60 in infants <2 months old and >50 for infants between 2 and 12 months old.3234); or 3) absence of severely diminished caloric intake. The prolonged LOS groups shared the common characteristics of higher duration of oxygen use and higher maximum respiratory rates. In addition to these two elements, each long stay group had either a constant oxygen requirement, prematurity (36.5 weeks), or very low caloric intake (<23.5 kcal/kg/day).

While all three models shared good sensitivity, the increased specificity of the Milwaukee Model limits the number of false positives (infants screening as destined for a prolonged LOS who actually will have a short LOS). For clinicians or researchers planning interventions for high‐risk infants, this greater specificity would reduce the number of infants who might unnecessarily receive those interventions. While there are limited proven therapies to hasten the recovery of patients with bronchiolits,35 many treatments and combinations of treatments are currently being used and studied. Nebulized hypertonic saline,36 airway secretion clearance modalities,37 and nutritional supplementation22, 38 are some examples of interventions that could be used and evaluated in infants screened as high risk for prolonged LOS. While it may have been better to identify short vs long stay immediately upon admission or after hospital day 1, it was the more clear separation between the short and long stay groups that occurred on day 2 that allowed us to develop an accurate predictive model. We believe that for infants destined to be in the hospital for at least three more days, a model based on hospital day 2 variables is worthwhile.

When evaluating the characteristics of the three models, it is important to note that they were initially studied in populations with some important differences. Only 11% of Milwaukee patients had chronic respiratory diseases, whereas the previously developed models were generated from a sample with a higher prevalence of chronic respiratory diseases (Michigan, 20%; Rotterdam, 23%). Only 3% of subjects needed placement in the ICU compared to higher rates in prior studies (Michigan, 15%; Rotterdam, 43.5%). In a population of patients with a lower prevalence of chronic respiratory diseases and need for ICU, early clinical markers may become more important in predicting LOS. While our model may generalize well in such a cohort of patients, it might not generalize as well to a cohort with a high prevalence of chronic lung disease and higher need for ICU. It is also important to note that the area under the ROC was lower in the test tree than the learning tree. This variation demonstrates the need for evaluating the performance of this model in additional populations.

This study has several limitations. First, it is a retrospective study of a single bronchiolitis season at a single institution. Second, the authors served as data abstractors and could have been biased, as they were not blinded. Third, four out of the five markers in our model are clinical markers that could vary based on clinical assessment skills and institutional practice. For example, oxygen use is dependent on the practice of nurses and respiratory therapists charged with regulating the oxygen delivery. However, the practice of initiating and weaning oxygen is fairly standardized at our institution. Fourth, environmental and social risk factors, such as day care attendance, school‐age siblings, and smoke exposure, can impact LOS but were not included in our model. Fifth, we do not have data on those infants not included in the bronchiolitis protocol. It is possible that they differed from those in protocol. Finally, six infants were either placed or transferred to the ICU on day 1, which may make them inherently different than the other infants in the model. However, four out of these six infants did go on to have a short stay, highlighting the fact that many other factors affect LOS.

We believe this model may be useful because the clinical markers it uses represent some of the key problems seen in bronchiolitis (poor oxygenation, tachypnea, and poor feeding). Careful assessment of these clinical markers can allow effective prediction of those infants likely to have a prolonged LOS. This early risk assessment could allow more effective targeting of interventions to help high‐risk infants.

CONCLUSIONS

There are important differences between infants with bronchiolitis having short and prolonged hospital stays, including several clinical markers identifiable on hospital day 2, such as the length and amount of oxygen received, minimum oxygen saturation, maximum respiratory rate, clinical respiratory scores, deep suctioning need, and caloric intake. The Milwaukee Model uses the number of hours of supplemental oxygen, respiratory rate, minimum supplemental oxygen use, gestation, and caloric intake to predict short or prolonged LOS. It performed well with a good ROC, sensitivity, and specificity in one population of infants with a low prevalence of chronic respiratory disease.

Prior studies have identified risk factors for increased severity of illness, readmission, or prolonged length of stay (LOS) in infants admitted with bronchiolitis.123 These risk factors include birth‐related factors (prematurity, birth within six months of respiratory syncytial virus [RSV] season, discharge from the neonatal intensive care unit during winter, multiple birth infant), environmental factors (day care attendance, school‐age siblings, smoke exposure), and underlying diseases (chronic lung disease and other pulmonary conditions, failure to thrive (FTT), congenital heart disease, immunologic disorders, and neuromuscular disease).123 Additional risk factors occurring during the bronchiolitis course that have been associated with prolonged hospital course are mechanical ventilation, intensive care unit (ICU) admission, hypoxia on admission, apnea, feeding problems, and duration of supplemental oxygen.2123

Having a reliable model to identify infants at high risk for prolonged LOS early in the course of an admission would be helpful, both for clinical care and for studies of interventions designed to reduce LOS. Prior attempts at developing a model have yielded mixed results. The Michigan Logistic Regression Model displayed excellent predictive ability with an area under the receiver‐operator curve (ROC) of 0.88 using variables including prematurity, FTT, pulmonary disease, other comorbid diseases, and early mechanical ventilation.21 However, when applied to another patient population it did not perform as well. The Rotterdam Model using the variables of weight and supplemental oxygen had an ROC of 0.65.24

Prior prediction models have focused more on birth‐ and disease‐related risk factors than on hospital course factors, particularly common clinical assessments including respiratory status and caloric intake. An additional limitation of prior models is some loss of ability to study the interaction between various predictor variables when using multivariate regression methods.

Our aims were: 1) to study the associations of various clinical markers identifiable during the first two days of the hospital admission with LOS; and 2) to develop a LOS prediction model, using both previously identified risk factors and more detailed clinical data from the first two days of the hospital admission.

MATERIALS AND METHODS

RESULTS

Three hundred forty‐seven infants were admitted during the 20042005 bronchiolitis season, with 273 placed in the bronchiolitis treatment protocol. The charts of these 273 patients were reviewed. One was excluded because of gastrostomy tube placement during the admission. Of the remaining 272 patients, 47 (17.3%) had a LOS 108 hours. The median LOS was 59 hours (range 10334 hours). Two patients had missing data for caloric intake on days 1 and 2; and 23 patients did not have height obtained, therefore their Waterlow classification could not be determined. Historical details concerning fever, congestion, cough, and diminished caloric intake preceding admission were variably reported, resulting in a smaller sample size for these baseline characteristics as described in Table 1.

Recursive Partitioning Model

Figure 1 depicts the recursive partitioning model that best predicted LOS. Five variables were selected by the recursive partitioning model. Selected variables, in order of appearance (variable importance is related to order of appearance, ie, most important variable is first), were: hours of supplemental oxygen, maximum respiratory rate, minimum supplemental oxygen use, gestation, and kilocalories (kcal)/kilogram (kg)/day consumed. The characteristics of this model were: ROC 0.89 and 0.72 for the learning and test trees, respectively; sensitivity, 0.85; and specificity, 0.82

Figure 1

Infants predicted as having a short LOS had three distinct profiles labeled S1, S2, and S3 in Figure 1. The S1 group required 6.5 hours of oxygen. The S2 group required >6.5 hours of oxygen, but had a maximum respiratory rate 49. The S3 group required >6.5 hours of oxygen, had a maximum respiratory rate >49, but were >36.5 week gestation, and consumed >23.5 kcal/kg/day.

Infants predicted as having a long LOS had three distinct profiles labeled L1, L2, and L3 in Figure 1. The L1 group required >6.5 hours of supplemental oxygen, had a maximum respiratory rate >49, and required some level of oxygen support the entire day. The L2 group required >6.5 hours of supplemental oxygen, had a maximum respiratory rate >49, were on room air some portion of the day, but had a gestation 36.5 weeks. The L3 group required >6.5 hours of supplemental oxygen, had a maximum respiratory rate >49, were on room air for some portion of the day, had a gestation >36.5 weeks, but consumed 23.5 kcal/kg/day.

Table 3 compares the performance of our model (the Milwaukee Model), the Michigan Model, and the Rotterdam Model in predicting LOS group. Overall, the Milwaukee Model had the highest ROC 0.89/0.72 for the learning and test trees. All three models had good sensitivity (Milwaukee, 85%; Michigan, 85%), with the Rotterdam Model having the highest (98%). The Milwaukee Model also had good specificity (82%), while the Michigan and Rotterdam Models were less specific (46% and 44%).

Table 3.Comparison of Three Models for Predicting LOS in Infants Admitted with Bronchiolitis

Model	Priors*		Sensitivity	Specificity	Learning Tree ROC	Test Tree ROC
	Long LOS	Short LOS
Abbreviations: LOS, length of stay; ROC, receiver‐operator curve. * The priors (or weighting system) selected for model by Salford Systems' CART software.
Michigan	0.5	0.5	0.85	0.46	0.69	0.56
Rotterdam	0.5	0.5	0.98	0.44	0.73	0.61
Milwaukee	0.5	0.5	0.85	0.82	0.89	0.72

DISCUSSION

We confirmed several previously recognized risk factors for prolonged LOS, including: apnea, at least part of the hospital stay in ICU, use of CPAP, mechanical ventilation, and prematurity. However, most patients admitted with bronchiolitis do not have these risk factors. The major contribution of this study is the evaluation of factors applicable to all patients admitted with bronchiolitis, and a more in‐depth analysis of clinical assessments performed on hospital days 1 and 2 than had been previously reported.

We did find strong associations between a number of clinical assessments and LOS. While some were apparent on day 1 of the admission, the number and degree of clinical differences between infants destined for a short vs prolonged stay were more apparent on hospital day 2. On this day, there were significant differences between the groups in the length and amount of oxygen received, oxygen saturation, maximum respiratory rate, respiratory scores, nasopharyngeal suctioning need, and caloric intake. Interestingly, it was noted the prolonged stay group had overall worsening or a lack of improvement in several clinical markers from day 1 to day 2, in areas where the short stay group showed improvements.

To our knowledge, the Milwaukee Model is the first bronchiolitis LOS prediction model to incorporate several clinical markers occurring early in the hospital stay. These clinical markers were found to be more effective predictors of LOS group in our study population than some of the traditional birth‐ and disease‐related risk factors previously reported. The model highlighted some important interactions among variables, and identified specific profiles of patients likely to have a short or prolonged LOS based on their day 2 clinical status.

The short LOS groups all shared one of the following three features: 1) low duration of oxygen use; 2) absence of tachypnea (tachypnea defined as a respiratory rate >60 in infants <2 months old and >50 for infants between 2 and 12 months old.3234); or 3) absence of severely diminished caloric intake. The prolonged LOS groups shared the common characteristics of higher duration of oxygen use and higher maximum respiratory rates. In addition to these two elements, each long stay group had either a constant oxygen requirement, prematurity (36.5 weeks), or very low caloric intake (<23.5 kcal/kg/day).

While all three models shared good sensitivity, the increased specificity of the Milwaukee Model limits the number of false positives (infants screening as destined for a prolonged LOS who actually will have a short LOS). For clinicians or researchers planning interventions for high‐risk infants, this greater specificity would reduce the number of infants who might unnecessarily receive those interventions. While there are limited proven therapies to hasten the recovery of patients with bronchiolits,35 many treatments and combinations of treatments are currently being used and studied. Nebulized hypertonic saline,36 airway secretion clearance modalities,37 and nutritional supplementation22, 38 are some examples of interventions that could be used and evaluated in infants screened as high risk for prolonged LOS. While it may have been better to identify short vs long stay immediately upon admission or after hospital day 1, it was the more clear separation between the short and long stay groups that occurred on day 2 that allowed us to develop an accurate predictive model. We believe that for infants destined to be in the hospital for at least three more days, a model based on hospital day 2 variables is worthwhile.

When evaluating the characteristics of the three models, it is important to note that they were initially studied in populations with some important differences. Only 11% of Milwaukee patients had chronic respiratory diseases, whereas the previously developed models were generated from a sample with a higher prevalence of chronic respiratory diseases (Michigan, 20%; Rotterdam, 23%). Only 3% of subjects needed placement in the ICU compared to higher rates in prior studies (Michigan, 15%; Rotterdam, 43.5%). In a population of patients with a lower prevalence of chronic respiratory diseases and need for ICU, early clinical markers may become more important in predicting LOS. While our model may generalize well in such a cohort of patients, it might not generalize as well to a cohort with a high prevalence of chronic lung disease and higher need for ICU. It is also important to note that the area under the ROC was lower in the test tree than the learning tree. This variation demonstrates the need for evaluating the performance of this model in additional populations.

This study has several limitations. First, it is a retrospective study of a single bronchiolitis season at a single institution. Second, the authors served as data abstractors and could have been biased, as they were not blinded. Third, four out of the five markers in our model are clinical markers that could vary based on clinical assessment skills and institutional practice. For example, oxygen use is dependent on the practice of nurses and respiratory therapists charged with regulating the oxygen delivery. However, the practice of initiating and weaning oxygen is fairly standardized at our institution. Fourth, environmental and social risk factors, such as day care attendance, school‐age siblings, and smoke exposure, can impact LOS but were not included in our model. Fifth, we do not have data on those infants not included in the bronchiolitis protocol. It is possible that they differed from those in protocol. Finally, six infants were either placed or transferred to the ICU on day 1, which may make them inherently different than the other infants in the model. However, four out of these six infants did go on to have a short stay, highlighting the fact that many other factors affect LOS.

We believe this model may be useful because the clinical markers it uses represent some of the key problems seen in bronchiolitis (poor oxygenation, tachypnea, and poor feeding). Careful assessment of these clinical markers can allow effective prediction of those infants likely to have a prolonged LOS. This early risk assessment could allow more effective targeting of interventions to help high‐risk infants.

CONCLUSIONS

There are important differences between infants with bronchiolitis having short and prolonged hospital stays, including several clinical markers identifiable on hospital day 2, such as the length and amount of oxygen received, minimum oxygen saturation, maximum respiratory rate, clinical respiratory scores, deep suctioning need, and caloric intake. The Milwaukee Model uses the number of hours of supplemental oxygen, respiratory rate, minimum supplemental oxygen use, gestation, and caloric intake to predict short or prolonged LOS. It performed well with a good ROC, sensitivity, and specificity in one population of infants with a low prevalence of chronic respiratory disease.

Prior studies have identified risk factors for increased severity of illness, readmission, or prolonged length of stay (LOS) in infants admitted with bronchiolitis.123 These risk factors include birth‐related factors (prematurity, birth within six months of respiratory syncytial virus [RSV] season, discharge from the neonatal intensive care unit during winter, multiple birth infant), environmental factors (day care attendance, school‐age siblings, smoke exposure), and underlying diseases (chronic lung disease and other pulmonary conditions, failure to thrive (FTT), congenital heart disease, immunologic disorders, and neuromuscular disease).123 Additional risk factors occurring during the bronchiolitis course that have been associated with prolonged hospital course are mechanical ventilation, intensive care unit (ICU) admission, hypoxia on admission, apnea, feeding problems, and duration of supplemental oxygen.2123

Having a reliable model to identify infants at high risk for prolonged LOS early in the course of an admission would be helpful, both for clinical care and for studies of interventions designed to reduce LOS. Prior attempts at developing a model have yielded mixed results. The Michigan Logistic Regression Model displayed excellent predictive ability with an area under the receiver‐operator curve (ROC) of 0.88 using variables including prematurity, FTT, pulmonary disease, other comorbid diseases, and early mechanical ventilation.21 However, when applied to another patient population it did not perform as well. The Rotterdam Model using the variables of weight and supplemental oxygen had an ROC of 0.65.24

Prior prediction models have focused more on birth‐ and disease‐related risk factors than on hospital course factors, particularly common clinical assessments including respiratory status and caloric intake. An additional limitation of prior models is some loss of ability to study the interaction between various predictor variables when using multivariate regression methods.

Our aims were: 1) to study the associations of various clinical markers identifiable during the first two days of the hospital admission with LOS; and 2) to develop a LOS prediction model, using both previously identified risk factors and more detailed clinical data from the first two days of the hospital admission.

MATERIALS AND METHODS

RESULTS

Three hundred forty‐seven infants were admitted during the 20042005 bronchiolitis season, with 273 placed in the bronchiolitis treatment protocol. The charts of these 273 patients were reviewed. One was excluded because of gastrostomy tube placement during the admission. Of the remaining 272 patients, 47 (17.3%) had a LOS 108 hours. The median LOS was 59 hours (range 10334 hours). Two patients had missing data for caloric intake on days 1 and 2; and 23 patients did not have height obtained, therefore their Waterlow classification could not be determined. Historical details concerning fever, congestion, cough, and diminished caloric intake preceding admission were variably reported, resulting in a smaller sample size for these baseline characteristics as described in Table 1.

Recursive Partitioning Model

Figure 1 depicts the recursive partitioning model that best predicted LOS. Five variables were selected by the recursive partitioning model. Selected variables, in order of appearance (variable importance is related to order of appearance, ie, most important variable is first), were: hours of supplemental oxygen, maximum respiratory rate, minimum supplemental oxygen use, gestation, and kilocalories (kcal)/kilogram (kg)/day consumed. The characteristics of this model were: ROC 0.89 and 0.72 for the learning and test trees, respectively; sensitivity, 0.85; and specificity, 0.82

Figure 1

Infants predicted as having a short LOS had three distinct profiles labeled S1, S2, and S3 in Figure 1. The S1 group required 6.5 hours of oxygen. The S2 group required >6.5 hours of oxygen, but had a maximum respiratory rate 49. The S3 group required >6.5 hours of oxygen, had a maximum respiratory rate >49, but were >36.5 week gestation, and consumed >23.5 kcal/kg/day.

Infants predicted as having a long LOS had three distinct profiles labeled L1, L2, and L3 in Figure 1. The L1 group required >6.5 hours of supplemental oxygen, had a maximum respiratory rate >49, and required some level of oxygen support the entire day. The L2 group required >6.5 hours of supplemental oxygen, had a maximum respiratory rate >49, were on room air some portion of the day, but had a gestation 36.5 weeks. The L3 group required >6.5 hours of supplemental oxygen, had a maximum respiratory rate >49, were on room air for some portion of the day, had a gestation >36.5 weeks, but consumed 23.5 kcal/kg/day.

Table 3 compares the performance of our model (the Milwaukee Model), the Michigan Model, and the Rotterdam Model in predicting LOS group. Overall, the Milwaukee Model had the highest ROC 0.89/0.72 for the learning and test trees. All three models had good sensitivity (Milwaukee, 85%; Michigan, 85%), with the Rotterdam Model having the highest (98%). The Milwaukee Model also had good specificity (82%), while the Michigan and Rotterdam Models were less specific (46% and 44%).

Table 3.Comparison of Three Models for Predicting LOS in Infants Admitted with Bronchiolitis

Model	Priors*		Sensitivity	Specificity	Learning Tree ROC	Test Tree ROC
	Long LOS	Short LOS
Abbreviations: LOS, length of stay; ROC, receiver‐operator curve. * The priors (or weighting system) selected for model by Salford Systems' CART software.
Michigan	0.5	0.5	0.85	0.46	0.69	0.56
Rotterdam	0.5	0.5	0.98	0.44	0.73	0.61
Milwaukee	0.5	0.5	0.85	0.82	0.89	0.72

DISCUSSION

We confirmed several previously recognized risk factors for prolonged LOS, including: apnea, at least part of the hospital stay in ICU, use of CPAP, mechanical ventilation, and prematurity. However, most patients admitted with bronchiolitis do not have these risk factors. The major contribution of this study is the evaluation of factors applicable to all patients admitted with bronchiolitis, and a more in‐depth analysis of clinical assessments performed on hospital days 1 and 2 than had been previously reported.

We did find strong associations between a number of clinical assessments and LOS. While some were apparent on day 1 of the admission, the number and degree of clinical differences between infants destined for a short vs prolonged stay were more apparent on hospital day 2. On this day, there were significant differences between the groups in the length and amount of oxygen received, oxygen saturation, maximum respiratory rate, respiratory scores, nasopharyngeal suctioning need, and caloric intake. Interestingly, it was noted the prolonged stay group had overall worsening or a lack of improvement in several clinical markers from day 1 to day 2, in areas where the short stay group showed improvements.

To our knowledge, the Milwaukee Model is the first bronchiolitis LOS prediction model to incorporate several clinical markers occurring early in the hospital stay. These clinical markers were found to be more effective predictors of LOS group in our study population than some of the traditional birth‐ and disease‐related risk factors previously reported. The model highlighted some important interactions among variables, and identified specific profiles of patients likely to have a short or prolonged LOS based on their day 2 clinical status.

The short LOS groups all shared one of the following three features: 1) low duration of oxygen use; 2) absence of tachypnea (tachypnea defined as a respiratory rate >60 in infants <2 months old and >50 for infants between 2 and 12 months old.3234); or 3) absence of severely diminished caloric intake. The prolonged LOS groups shared the common characteristics of higher duration of oxygen use and higher maximum respiratory rates. In addition to these two elements, each long stay group had either a constant oxygen requirement, prematurity (36.5 weeks), or very low caloric intake (<23.5 kcal/kg/day).

While all three models shared good sensitivity, the increased specificity of the Milwaukee Model limits the number of false positives (infants screening as destined for a prolonged LOS who actually will have a short LOS). For clinicians or researchers planning interventions for high‐risk infants, this greater specificity would reduce the number of infants who might unnecessarily receive those interventions. While there are limited proven therapies to hasten the recovery of patients with bronchiolits,35 many treatments and combinations of treatments are currently being used and studied. Nebulized hypertonic saline,36 airway secretion clearance modalities,37 and nutritional supplementation22, 38 are some examples of interventions that could be used and evaluated in infants screened as high risk for prolonged LOS. While it may have been better to identify short vs long stay immediately upon admission or after hospital day 1, it was the more clear separation between the short and long stay groups that occurred on day 2 that allowed us to develop an accurate predictive model. We believe that for infants destined to be in the hospital for at least three more days, a model based on hospital day 2 variables is worthwhile.

When evaluating the characteristics of the three models, it is important to note that they were initially studied in populations with some important differences. Only 11% of Milwaukee patients had chronic respiratory diseases, whereas the previously developed models were generated from a sample with a higher prevalence of chronic respiratory diseases (Michigan, 20%; Rotterdam, 23%). Only 3% of subjects needed placement in the ICU compared to higher rates in prior studies (Michigan, 15%; Rotterdam, 43.5%). In a population of patients with a lower prevalence of chronic respiratory diseases and need for ICU, early clinical markers may become more important in predicting LOS. While our model may generalize well in such a cohort of patients, it might not generalize as well to a cohort with a high prevalence of chronic lung disease and higher need for ICU. It is also important to note that the area under the ROC was lower in the test tree than the learning tree. This variation demonstrates the need for evaluating the performance of this model in additional populations.

This study has several limitations. First, it is a retrospective study of a single bronchiolitis season at a single institution. Second, the authors served as data abstractors and could have been biased, as they were not blinded. Third, four out of the five markers in our model are clinical markers that could vary based on clinical assessment skills and institutional practice. For example, oxygen use is dependent on the practice of nurses and respiratory therapists charged with regulating the oxygen delivery. However, the practice of initiating and weaning oxygen is fairly standardized at our institution. Fourth, environmental and social risk factors, such as day care attendance, school‐age siblings, and smoke exposure, can impact LOS but were not included in our model. Fifth, we do not have data on those infants not included in the bronchiolitis protocol. It is possible that they differed from those in protocol. Finally, six infants were either placed or transferred to the ICU on day 1, which may make them inherently different than the other infants in the model. However, four out of these six infants did go on to have a short stay, highlighting the fact that many other factors affect LOS.

We believe this model may be useful because the clinical markers it uses represent some of the key problems seen in bronchiolitis (poor oxygenation, tachypnea, and poor feeding). Careful assessment of these clinical markers can allow effective prediction of those infants likely to have a prolonged LOS. This early risk assessment could allow more effective targeting of interventions to help high‐risk infants.

CONCLUSIONS

There are important differences between infants with bronchiolitis having short and prolonged hospital stays, including several clinical markers identifiable on hospital day 2, such as the length and amount of oxygen received, minimum oxygen saturation, maximum respiratory rate, clinical respiratory scores, deep suctioning need, and caloric intake. The Milwaukee Model uses the number of hours of supplemental oxygen, respiratory rate, minimum supplemental oxygen use, gestation, and caloric intake to predict short or prolonged LOS. It performed well with a good ROC, sensitivity, and specificity in one population of infants with a low prevalence of chronic respiratory disease.

Many hospitalized patients have microbiology test results pending at the time of discharge.1, 2 Failure to follow up on these results in a timely fashion can lead to delays in diagnosis and adequate treatment of important infections. Prompt communication of the results of these pending tests to the responsible providers is crucial to minimize these delays.36 As hospitalized patients are increasingly cared for by clinicians other than their primary care providers,7 important information may be lost during the discharge process.8 This increasing fragmentation makes reliable communication of pending tests even more crucial.9, 10

Studies to date have primarily investigated tests from general medical services. In that setting, there is clearly room for improvement in test result communication. Discharge summaries often do not reach the outpatient providers at the time of the patients' follow‐up visits after hospitalization.11 When the discharge summaries are present, the majority of pending tests are not mentioned in them,2, 12, 13 and both inpatient and outpatient physicians are unaware of most of these results when they return.1 However, the specific characteristics of postdischarge microbiology results and the extent to which these results represent potential follow‐up errors in specialties other than general medicine have not been adequately studied.

We aimed to describe the issue of microbiology tests pending at the time of discharge from a hospital‐wide perspective. Specifically, we sought to determine: (1) frequency and characteristics of these results across all admitting services; and (2) how often these results potentially require a change in antimicrobial therapy.

Methods

Manual Review

We manually reviewed a random sample of 94 of the clinically important and untreated postdischarge results to determine if the results potentially required a change in therapy and therefore required follow‐up. For each case, the electronic patient chart was reviewed by two internal medicine‐trained physicians on the study staff. Each reviewer was blinded to events that occurred after the cultures returned, and determined whether the results necessitated a potential change in antibiotic. The reviewer then indicated the level of certainty of that determination on a 6‐point Likert scale that had been previously used in reviews to identify adverse medical events15, 16: 1 = little or no evidence, 2 = slight evidence, 3 = not quite likely (<50:50 but close call), 4 = more likely than not (>50:50 but close call), 5 = strong evidence, and 6 = virtually certain evidence. To standardize the assignment of certainty for potential need for antibiotic change, we used a set of review guidelines developed by our study staff (Figure 1). A microbiology result was defined as potentially necessitating antibiotic change if both reviewers indicated as such and recorded a certainty with a score 4. Differences in assessments were resolved through discussion of the case between the reviewers.

Figure 1

Results

Of the 111,331 blood, urine, sputum, and CSF cultures analyzed, 77,349 (69%) were collected from hospitalized patients. The majority (63%) of the inpatient results were for blood cultures and one quarter (24%) were for urine cultures. Table 1 shows the distribution of the microbiology results across primary services responsible for the patients at the time of discharge. Half (49%) of the patients from whom the specimens were collected were female. The mean age of patients was 55 years. Most (68%) were white and most (86%) had either commercial insurance or Medicare (Table 1).

Of the 77,349 microbiology tests obtained during hospital stays, 8668 (11%) of the inpatient microbiology results were reported after the patients were discharged from the hospital. Most (54%) of these postdischarge results were for blood cultures. The distribution of results across primary hospital service, patient sex, race, insurance, and mean patient age were similar to those for all inpatient results (Table 1). Of the 8668 postdischarge results, 385 (4%) met our screening criteria of being both clinically important and not treated by an antibiotic to which the organism was found susceptible at the time of discharge from the hospital. After manual review of a random subset of 94 of these screen‐positive cases, 50 (53%) required follow‐up (Figure 2). The interrater reliability for the reviewers was found to be kappa = 0.58 (P < 0.001). From our results, we estimated that 2.4% of the postdischarge microbiology results required follow‐up and potential change in therapy.

Figure 2

Potential need for antibiotic change was present in 30 of 45 (67%) urine cultures, 12 of 32 (38%) blood cultures, 8 of 16 (50%) sputum cultures, and 0 of 1 (0%) CSF cultures. By primary service, reviewers identified a potential need for antibiotic change in 19 of 25 (76%) of results from surgical services, 17 of 29 (59%) from general medicine, 6 of 16 (38%) from oncology, and 8 of 23 (35%) from medical subspecialties. Examples of cases that potentially required antibiotic change are shown in Table 2.

In our logistic regression model, both the type of culture and the primary hospital service were found to be significant predictors of a potential need for antibiotic change in the manually reviewed cases. Urine cultures were more likely than non‐urine cultures to potentially require antibiotic change (P = 0.03; OR 2.8, 95% CI 1.1‐7.2). Results from surgical services were most likely to potentially require antibiotic change, followed by general medicine, oncology, and medical subspecialties (Table 3).

Discussion

We performed a retrospective analysis of all blood, urine, sputum, and CSF cultures finalized at our institution in 2007 and found that many returned after patients were discharged. Overall, we estimated that 2.4% of these postdischarge results potentially required a change in antibiotic. This proportion varied by culture type and by primary hospital service at the time of discharge, with urine cultures and cultures from surgical services being most likely to potentially need change in antibiotic.

We speculate that postdischarge urine cultures may have been more likely to require antibiotic change in part due to different urgency that clinicians assign to different culture types. Urinary tract infections may present with more vague, transient, or minor complaints compared with bacteremia, pneumonia, and cerebrospinal fluid infections. For that reason, clinicians may be more likely to forego empiric antibiotics for pending urine cultures in favor of watchful waiting. Therefore, the postdischarge urine cultures with growth may include a higher proportion of untreated true infections compared with other culture types.

A similar difference in prescription of empiric antibiotics may help explain the differences seen across primary hospital specialties. For example, if patients on surgical services were less likely to receive empiric antibiotics, then the pool of postdischarge results would be more likely to include true infections that require antibiotic change. Furthermore, it is possible that surgical services may tend to order cultures for patients only if they already have convincing evidence of infections. It may be that selecting a group with higher likelihood of infection led to a higher proportion of true infections in surgical patients with cultures with growth.

Prior studies led by Roy and Were illustrated that pending microbiology results from general medicine services were often not communicated and followed up adequately.1, 2 For patients discharged with pending test results, between 47% and 89% of discharge summaries did not mention the pending tests.2, 12, 13, 17 These deficiencies in discharge summaries likely have a substantial impact on the proportion of tests followed up by outpatient clinicians. By extending the analysis hospital‐wide, our study suggests that pending microbiology results occur for a wide range of hospital services. While our study was not designed to determine whether these results were followed up appropriately, opportunities for miscommunication and missed follow‐up likely exist for all specialties.

The potential harms associated with inadequate test follow‐up have gained the attention of the patient safety community. In 2005, the Joint Commission underscored the importance of proper communication of critical lab results.3, 5, 18 Their recommendations included the development of systems to ensure adequate follow‐up of critical results in high‐risk scenarios including the postdischarge period.5 While many of the microbiology results do not fall into the criticalcategory, we feel that these results should be considered for inclusion in hospital efforts to track postdischarge results. These efforts should also address issues specific to microbiology results, such as preliminary status before antibiotic sensitivities are known.

Developing a comprehensive strategy for test result communication is challenging, and more so for results that return after transitions of care. Even defining the proper target of communication interventions can involve complex organizational and cultural issues. As these results span the inpatient and outpatient domains, there may be some ambiguity as to which provider is responsible when the results return. The inpatient clinicians ordering the microbiology cultures are in the best position to put the results into the patient's clinical context. However, these clinicians may no longer be on clinical duty when the results return, or they may not have a system to ensure that they are notified about these results. While the outpatient providers may be available, they have often not seen the patient in follow‐up at the time the results return and would need to repeat a clinical assessment to determine whether a change in antibiotics is required. While many feel that the ordering provider is a logical choice to perform the follow‐up of the result, not all agree and few institutions have developed clear policies on this issue. To avoid this ambiguity, future work will require institutions to clearly outline which party is responsible for test result follow‐up during transitions of care.

Potential solutions to improve communication of these results must be tailored to the local infrastructure of the institution. In hospitals that do not have extensive electronic systems, a solution might involve a registered nurse, nurse practitioner, or lab technician whose responsibilities include identifying postdischarge results and communicating them to the ordering clinician, primary care provider, and patient. In settings with more advanced electronic infrastructure, solutions could be designed to automatically notify the responsible providers electronically, as well as post the results to a patient portal. Regardless of the level of technical sophistication, it is vital to create a system that has is highly reliable to prevent these important results from falling through the cracks.

Our study did have some limitations. First, we evaluated results from only one institution. It is unclear how substantially differences in practice patterns or patient populations would affect the number of postdischarge microbiology results in other settings. Second, we did not assess whether these results were actually followed up or whether treatment regimens were altered. As this study was retrospective in nature, we could not expect clinicians to recall the clinical scenarios surrounding each result and decided that documentation in clinical notes would be an unreliable indicator of whether any follow‐up action had been taken. Even without this information, however, we would submit that our findings represent a substantial near‐miss rate and threat to patient safety (approximately one potentially actionable, postdischarge microbiology result every other day for our hospital), and call for a fail‐safe system to ensure appropriate actions are taken.

In conclusion, microbiology results are often pending at the time patients are discharged from the hospital and roughly 2.4% of these results potentially require a change in therapy. This proportion was highest for urine cultures and cultures drawn from surgical patients. Our results suggest that a hospital‐wide system is warranted to ensure adequate communication of postdischarge microbiology results. Further research is required to evaluate the impact of such a system on the follow‐up rates of pending microbiology tests.