Understanding the Broken Heart

Posted by Rena Xu • September 2nd, 2015

9-2-15 Understanding the broken heartIn 1897, recently released from prison, the author Oscar Wilde wrote to a close friend, “My desire to live is as intense as ever, and though my heart is broken, hearts are made to be broken: that is why God sends sorrow into the world. The hard heart is the evil thing of life and of art.” Heartbreak, Wilde seemed to imply, wasn’t just an inevitability of life; it was an affirmation of it.

Nearly a hundred years later, doctors in Japan reported that heartbreak could endanger life, too. Takotsubo cardiomyopathy (named after the Japanese term for an octopus trap, which the ballooning heart resembles), also commonly referred to as broken heart syndrome, is a form of acute heart failure that, while often temporary and benign, can also result in significant morbidity and even death. Nine out of ten afflicted individuals are women, and emotional events – a breakup, the death of a loved one, the loss of a job — are classically blamed as triggers.

But broken heart syndrome is more complicated, and less well understood, than its evocative name might suggest. Sometimes the trigger is physical, rather than emotional; traumatic injuries, surgeries, strokes, and asthma attacks are just some of the inciting events that have been reported. Other times, there is no satisfactory explanation for why the condition develops. The near- and long-term health consequences of Takotsubo also haven’t been clearly defined.

To help shed light on these issues, the International Takotsubo Registry was created, and data on seventeen hundred fifty patients afflicted by the condition were collected and studied. The results of the analysis, published this week in NEJM are surprising. Physical triggers were actually more common than emotional ones (36% versus 28%), and in many patients, there was no identifiable trigger (29%). Severe complications during hospitalization were just as likely as with acute coronary syndrome (19% for both) and included death in almost 4% of cases. Long-term complications were also significant, with major adverse cardiac or cardiovascular events occurring in 10% per patient-year, and death from any cause occurring in close to 6% per patient-year.

Takotsubo cardiomyopathy can often be difficult to distinguish from acute coronary syndrome. Notable similarities in an age- and sex-matched comparison included an elevated troponin and evidence of myocardial ischemia on initial EKG (seen in 80% of Takotsubo patients), suggesting that these characteristics aren’t useful for distinguishing between the two conditions. There were, however, a few salient differences. Those with Takotsubo were more likely to demonstrate a reduced left ventricular ejection fraction (87% versus 54%). They were also more likely to have an acute neurological or psychiatric illness (56% versus 26%), suggesting that a connection between the heart and brain could be contributing to the disease process.

The use of an angiotensin converting enzyme inhibitor or angiotensin receptor blocker was linked to improved one-year survival, while beta-blockers, which were previously thought to be protective, didn’t demonstrate a survival benefit. In fact, a third of patients developed Takotsubo cardiomyopathy while on a beta-blocker, and half of patients who suffered a recurrence (a 2% per patient-year risk) were on a beta-blocker at the time of the second event.

“Our data demonstrate that the spectrum of Takotsubo cardiomyopathy is wide and ranges from low to very high risk in the acute phase,” the authors write. A young man who develops the condition after being in a car accident, for instance, has a very different risk profile from an elderly woman whose husband has just passed away.

Perhaps it’s fitting that broken heart syndrome, like heartbreak itself, should be less than straightforward. It blurs the line between the physical and emotional; it implicates the mind in matters of the heart; it sometimes defies reason; and even with our best efforts to comprehend it, it remains, in many ways, a mystery.

Men Who Have Sex with Men

Posted by Carla Rothaus • August 28th, 2015

Primary Care for Men Who Have Sex with MenA new Clinical Practice article covers primary care for men who have sex with men. Care should include a detailed sexual history to assess the risk of infection with HIV, the use of preexposure prophylaxis for those at high risk, and appropriate vaccinations and regular testing for sexually transmitted infections.

Men who have sex with men, many of whom identify themselves as gay, are a distinct population that is at increased risk for certain health conditions. Some, but not all, of these conditions are directly related to sexual behavior.

Clinical Pearls

• What is the current trend in the HIV infection rate among men who have sex with men?

HIV infection continues to be the major health problem affecting men who have sex with men. In this population, the HIV infection rate increased by 12% between 2009 and 2013, with nearly 31,000 new infections annually.

• What measures are recommended for HIV prevention?

In order to counsel patients about the risk of acquiring HIV and other sexually transmitted infections, providers should first inquire about types of sexual activity (e.g., oral sex, anal sex, and oral-anal sex) and the use or nonuse of condoms during various sexual activities. The consistent use of condoms during anal sex has been associated with a risk of HIV acquisition that is 70% lower than the risk without the consistent use of condoms. Men who have sex with men should be offered routine HIV testing once or twice a year. As part of a comprehensive risk-reduction strategy, preexposure prophylaxis and postexposure prophylaxis should be considered for men who have sex with men if they are at high risk for HIV infection. The CDC recommends preexposure prophylaxis for men who have an HIV-infected sexual partner, who have recently had syphilis or infection with Neisseria gonorrhea or Chlamydia trachomatis, or who have had a high number of sexual partners, a history of anal sex without the use of condoms, or a history of commercial sex work. An alternative approach to the prevention of infection with HIV is postexposure prophylaxis. This strategy involves the use of antiretroviral medications (e.g., tenofovir disoproxil fumarate-emtricitabine and raltegravir) soon after an isolated incident of high-risk exposure (within 72 hours, and preferably within 24 hours) in conjunction with HIV testing; treatment is continued for 28 days.

Table 2. Recommendations for Preexposure Prophylaxis in Men Who Have Sex with Men.

Morning Report Questions

Q: What other screening tests are routinely recommended for men who have sex with men?

A: Even in the absence of symptoms, screening for sexually transmitted infections — including serologic testing for HIV and syphilis and oral, rectal, and urinary testing for N. gonorrhea and C. trachomatis — is recommended for sexually active men who have sex with men and should be performed once a year or, for patients at higher risk, twice a year. Unlike urethral infections, rectal gonorrhea and chlamydial infections are often asymptomatic. A meta-analysis that included 25 studies of lesbian, gay, and bisexual people reported that these groups were at a significantly higher risk for depression, anxiety, and alcohol dependence than heterosexual people. Men who have sex with men should be screened periodically for alcohol and drug use. They are also more likely than heterosexual men to use tobacco and should be asked about tobacco use periodically. Anal infection with high-risk types of human papillomavirus (HPV) can lead to anal cancer. It is unclear whether screening for anal cancer is warranted in men who have sex with men. Screening should be considered for men who have sex with men if trained providers are available to perform these procedures, but patients should understand that diagnostic procedures and precancer treatments for anal cancer have associated risks, including the risk of anal pain and bleeding, and that high-quality data showing that screening reduces the risk of anal cancer are not available. The CDC recommends regular screening for hepatitis C virus (HCV) infection among HIV-infected men who have sex with men and recommends one-time HCV testing before the initiation of preexposure HIV prophylaxis and for all persons born between 1945 and 1965.

Q: What vaccinations are recommended for men who have sex with men?

A: Men who have sex with men are at increased risk for several vaccine-preventable infections. Unvaccinated men who have sex with men should be tested for hepatitis B virus (HBV) infection, and vaccination should be offered if tests for HBV surface antigen and surface antibody are negative. Men who have sex with men are at increased risk for the acquisition of hepatitis A virus (HAV) infection if they engage in sexual practices such as oral-anal sex; outbreaks of HAV infections in this population have been reported. Vaccination should be offered to men who have sex with men if they do not have a documented history of HAV vaccination and are not immune according to the results of serologic testing. The Advisory Committee on Immunization Practices (ACIP) recommends routine human papillomavirus vaccination of men who have sex with men through 26 years of age. A 9-valent version of this vaccine was recently approved by the Food and Drug Administration and is preferred over the quadrivalent vaccine, since it offers broader protection than the quadrivalent vaccine. Data are lacking on the efficacy of vaccination in men older than 26 years of age who have sex with men, but the safety and immunogenicity of vaccination in this age group have been established. This population is reported to have an increased prevalence of oropharyngeal colonization with Neisseria meningitidis as compared with the general population, but population-level data are lacking to confirm whether they are at increased risk for meningococcal disease. Current guidelines from the ACIP do not include men who have sex with men as a group that is at high risk for meningococcal disease and do not recommend routine vaccination. However, vaccination of men who have sex with men is recommended by many local health departments and is prudent for men in this population who are living in or traveling to areas with reported outbreaks.

Table 1. Recommendations for Primary Care in Men Who Have Sex with Men.

Perioperative Bridging Anticoagulation

Posted by Carla Rothaus • August 28th, 2015

Perioperative Bridging AnticoagulationIn this Original Article, a trial assigned patients with atrial fibrillation who required warfarin interruption for an elective procedure to either bridging anticoagulation or placebo. Forgoing bridging was noninferior to bridging for arterial thromboembolism and superior for major bleeding. A video summary is also available with the article on NEJM.org.

For patients with atrial fibrillation who are receiving warfarin and require an elective operation or other elective invasive procedure, the need for bridging anticoagulation during perioperative interruption of warfarin treatment has long been uncertain. A randomized, double-blind, placebo-controlled trial by Douketis, et al. (the Bridging Anticoagulation in Patients who Require Temporary Interruption of Warfarin Therapy for an Elective Invasive Procedure or Surgery [BRIDGE] trial) hypothesized that forgoing bridging anticoagulation would be noninferior to bridging with low-molecular-weight heparin for the prevention of perioperative arterial thromboembolism and would be superior to bridging with respect to major bleeding.

Figure 1. BRIDGE Study Design.

Table 1. Baseline Characteristics of the Patients. 

Clinical Pearls

Do patients with atrial fibrillation who require interruption of their warfarin therapy for an elective operation have a worse outcome with respect to perioperative arterial thromboembolism if they forgo bridging anticoagulation?

In the Douketis study, 1884 patients were enrolled, with 950 assigned to receive no bridging therapy and 934 assigned to receive bridging therapy. The primary efficacy outcome was arterial thromboembolism at 30 days. In this study, a strategy of discontinuing warfarin treatment without the use of bridging anticoagulation was noninferior to the use of bridging anticoagulation for the prevention of arterial thromboembolism. At 30 days after the procedure, the incidence of arterial thromboembolism was 0.4% (four events among 918 patients) in the no-bridging group and 0.3% (three events among 895 patients) in the bridging group (mean between-group dif

ference, 0.1 percentage points; 95% confidence interval [CI], −0.6 to 0.8; P=0.01 for noninferiority; P=0.73 for superiority).

Table 2. Perioperative Anticoagulant Management.

How much greater is the risk of major bleeding with the use of bridging anticoagulation as compared to a strategy that discontinues perioperative warfarin treatment?

In the study by Douketis et al., bridging conferred a risk of major bleeding that was nearly triple the risk associated with no bridging. Major bleeding occurred in 1.3% of the patients (12 of 918) in the no-bridging group and in 3.2% (29 of 895) in the bridging group, which indicated that no bridging was superior to bridging with regard to major bleeding (relative risk, 0.41; 95% CI, 0.20 to 0.78; P=0.005).

Morning Report Questions

Q: Does a strategy forgoing bridging anticoagulation in this clinical setting increase the risk of myocardial infarction or venous thromboembolism?

A: In the Douketis study, there was no significant difference between the groups in the rates of acute myocardial infarction, deep-vein thrombosis, pulmonary embolism, or death.

Table 3. Study Outcomes.

Q: How generalizable are the findings of the Douketis study?

A: There are potential limitations of the BRIDGE trial. First, although it aimed to recruit a representative sample of patients with atrial fibrillation for whom bridging anticoagulation is normally considered, certain groups were underrepresented. Few patients had a CHADS2 score of 5 or 6, although the mean score of 2.3 is similar to that among patients with atrial fibrillation who were assessed in recent trials and patient registries, in which the mean scores were between 2.1 and 2.8. Patients undergoing major surgical procedures associated with high rates of arterial thromboembolism and bleeding (e.g., carotid endarterectomy, major cancer surgery, cardiac surgery, or neurosurgery) were not represented in the trial, although the procedures performed were representative of the most common interventions patients undergo during an interruption of therapeutic anticoagulation, the majority of which are low-risk procedures, such as colonoscopy or ambulatory surgery. In addition, the findings should not be applied to patients with mechanical heart valves, who were specifically not included in the trial.

Does sleep deprivation have any effect on elective daytime operations performed by attending surgeons?

Posted by Andrea Merrill • August 26th, 2015

Outcomes of daytime Procedures with border“How do you stay awake?” I am often asked from friends outside the medical field upon explaining my schedule and frequent 24 hour call shifts. When asked this question, my replies run the gamut from, “You get used to it” to, “Coffee!” or, “I’m a surgeon, we don’t need sleep!” and finally, “You’re so busy most of the time that it keeps you awake.” My friends then try to console me by stating that at least the schedule will get better once I am an attending surgeon and finished with residency.

Unfortunately, this is not the case. As a resident, I am protected by rules regulated by the Accreditation Council for Graduate Medical Education (ACGME). After working a 24 hour shift, it is required that I go home and sleep with a mandatory 14 hour respite from the hospital. An attending surgeon, on the other hand, is not protected by these rules. He or she may be up all night operating for an emergent case, and then have a full day of clinic or surgery the following day.

Concerns have been raised regarding the safety of allowing surgeons to perform elective operations after being awake all night on call. There have been efforts to either prevent this or require surgeons to disclose this information to patients as part of the informed consent process. In fact, this was the thesis of a Perspective article in the NEJM in 2010 by Drs. Michael Nurok, Charles A. Czeisler, and Lisa Soleymani Lehmann (specialists in anesthesia, sleep medicine, and bioethics, respectively). Predictably, this did not go over well in the surgical community. While in theory, restricting attending surgeons’ ability to operate post-call may reduce complications, there are many financial and logistical ramifications of such a policy. Fortunately for surgeons, a large, multi-institutional study by Govindarajan et al. published in this week’s NEJM found no effect of sleep deprivation on elective daytime operations performed by attending surgeons.

Govindarajan et al. performed a retrospective 1:1 matched cohort study using information from multiple linked health databases in Ontario, Canada of patients undergoing 1 of 12 commonly performed elective operations (ranging from laparoscopic cholecystectomy to coronary artery bypass graft). Patients were placed into the “control group” or the “post-midnight” group (aka the “sleep deprived” group) based on whether their surgeon had a fee code identifying that he or she had had a physician-patient interaction between midnight and 7am the night prior to surgery. Surgical outcomes were compared between the two groups with focus on the primary outcomes of mortality, complications, and readmission within 30 days and secondary outcomes of length of stay and duration of surgery.

Nearly 40,000 patients were included, split evenly between the 2 groups, which were treated by about 1,500 physicians at close to 150 hospitals. In 71% of cases where a physician worked at some point from midnight to 7am, the treatment provided included a billable procedure. The authors found no significant difference in crude or adjusted rates of the primary outcomes (complications, death, or readmission) or secondary outcomes (length of stay or duration of operation). Stratification by hospital academic status, physician age, or procedure type did not affect outcomes. There was, however, a small, but significant, increase in complications in cases where the surgeon had performed 2 or more procedures the night before. No differences were found in the other outcome measures, though.

While this study does shed new light on this issue, there are some limitations inherent to the study. The authors cleverly used billing codes to identify overnight work performed by the surgeons. While this gives some idea of how much a surgeon may have worked overnight, it cannot tell us exactly how many hours a surgeon slept the night prior to surgery, in both groups. Additionally, there was a high level of experience in the physicians studied—a mean of 21 years in practice—which may bias results, although adjusted analysis and stratification did not find any differences.

So what does this mean for surgeons and patients? Deputy Editor, Dr. Mary Beth Hamel says, “Although the authors report reassuring evidence that patients’ surgical outcomes are not compromised when experienced attending physicians work during the night prior to operating, this study does not exclude an adverse effect of sleep loss on attending surgeons’ performance. It is possible that the surgeons compensated by making adjustments in their schedules for days following night work.” While this study surely adds new data to the ongoing discussion on sleep-deprived surgeons, it by no means puts the issue to rest (pun intended). Especially with ongoing review of ACGME resident duty hours, we are sure to see more studies of its kind in the future and with it more controversy.

A 9-Month-Old Girl with Fevers

Posted by Carla Rothaus • August 21st, 2015

A Girl with Recurrent FeversIn a new Case Record of the Massachusetts General Hospital, a 9-month-old girl presented with a 2-month history of recurrent fevers. Examination revealed fever and tachycardia and was otherwise normal; chest and abdominal imaging studies showed no evidence of infection. Additional diagnostic tests were performed.

Septic arthritis is a pediatric orthopedic emergency.

Clinical Pearls

• What is the definition of fever of unknown origin in children, and how often is infection the cause?

While the terms recurrent fever and fever of unknown origin are often used interchangeably, and the differential diagnosis is similar for both, there are specific definitions for fever of unknown origin. In 1961, fever of unknown origin was defined as a temperature of greater than 38.3 degrees C (101 degrees F) “on several occasions,” a duration of illness of greater than 3 weeks, and no cause of fever despite 1 week of inpatient investigation. Over subsequent decades, there have been amendments to the definition and proposals for different categories. Although there is no single generally accepted definition, a working definition would include a temperature of greater than 38.3 degrees C (101 degrees F), a duration of illness of at least 1 week, and a negative initial outpatient or inpatient evaluation, which includes history taking, physical examination, and routine laboratory testing. The differential diagnosis of fever of unknown origin is broad. A systematic review of 18 studies performed between 1950 and 2010 involving children who were evaluated for fever of unknown origin showed that half had infections, fewer than 10% had collagen vascular disease or malignant tumors, and almost one quarter had no diagnosis.

Table 2. Differential Diagnosis of Pediatric Fever of Unknown Origin.

What clinical features can help to distinguish septic arthritis from toxic synovitis?

Features that can help in distinguishing septic arthritis from toxic synovitis in children are an inability to bear weight, fever, and elevations of the erythrocyte sedimentation rate (>40 mm per hour) and white-cell count (>12,000 per cubic millimeter).

Morning Report Questions

Q: What are some of the microorganisms identified in septic arthritis affecting young children?

A: The microbiologic features of septic arthritis can vary depending on age, immunization history, possible exposures, and the presence of chronic conditions. Methicillin-susceptible Staphylococcus aureus (MSSA) and methicillin-resistant Staphylococcus aureus (MRSA) are the most common causes in all age groups. Of the types of streptococci, group B streptococcus is most common in infants younger than 3 months of age, group A streptococcus is most common in infants older than 3 months of age, and Streptococcus pneumoniae is most common in infants older than 3 months of age who are not immunized. The gram-negative bacillus Kingella kingae is increasingly identified in patients with septic arthritis who are between 3 months and 3 years of age.

Q: What long term sequelae may result from septic arthritis in children?

A: Septic arthritis is a pediatric orthopedic emergency. Early diagnosis and expedited treatment are critical for a satisfactory outcome. Although bone has the ability to repair itself, articular and epiphyseal cartilage does not. In addition to the effect of the infection on the child’s general health, there is a risk of considerable damage to the joint. Potent proteolytic enzymes are released in the joint directly from the bacteria and host tissue (i.e., synovial cells and chondrocytes) in response to the infection; these enzymes destroy the hyaline cartilage. Some studies have shown that cartilage destruction begins within 6 to 8 hours after bacterial colonization. Capsular distention, which occurs as a result of the effusion, coupled with muscle spasm, which occurs as a result of pain, can lead to a pathological dislocation of the joint and necessitate the application of a brace. Increased intraarticular pressure, which occurs as a result of the pus accumulation, can lead to avascular necrosis of the femoral head. All these features can lead to limb-length discrepancy, proximal femoral deformity, acetabular dysplasia, and joint stiffness and may lead to degenerative arthritis in the long term.

Figure 3. Imaging Studies of the Pelvis and Hips.

Unprovoked Venous Thromboembolism

Posted by Carla Rothaus • August 21st, 2015

Insight post 8-19-15In this Original Article, a trial showed that the addition of abdominopelvic CT to routine measures in patients with unprovoked venous thrombosis did not detect additional occult cancers. The incidence of cancer in first unprovoked venous thrombosis was 4%, not 10% as had been previously reported.

Carrier et al. conducted a multicenter, open-label, randomized, controlled trial in which patients who had a first unprovoked venous thromboembolism were randomly assigned to undergo limited occult-cancer screening (basic blood testing, chest radiography, and screening for breast, cervical, and prostate cancer) or limited occult-cancer screening in combination with comprehensive computed tomography (CT) of the abdomen and pelvis. CT included a virtual colonoscopy and gastroscopy, biphasic enhanced CT of the liver, parenchymal pancreatography, and uniphasic enhanced CT of the distended bladder. The primary outcome measure was confirmed cancer that was missed by the screening strategy and detected by the end of the 1-year follow-up period.

Clinical Pearls

• What is the association between unprovoked venous thromboembolism and occult malignancy?

Venous thromboembolism is classified as provoked when it is associated with a transient risk factor (e.g., trauma, surgery, prolonged immobility, or pregnancy or the puerperium) and as unprovoked when it is associated with neither a strong transient risk factor nor overt cancer. Unprovoked venous thromboembolism may be the earliest sign of cancer; up to 10% of patients with unprovoked venous thromboembolism receive a diagnosis of cancer in the year after their diagnosis of venous thromboembolism. More than 60% of occult cancers are diagnosed shortly after the diagnosis of unprovoked venous thromboembolism. Thereafter, the incidence rate of cancer diagnosis gradually declines and returns to the rate in the general population after 1 year.

• Is there a consensus about how to evaluate a patient with a first unprovoked venous thromboembolism?

Clinicians, patients, and policymakers struggle with how aggressive to be in screening for occult cancers in patients who present with unprovoked venous thromboembolism. The rationale for screening is to allow early detection and intervention and ultimately reduce cancer-related mortality. However, owing to the paucity of data in this context, there is great variation in practice. Whereas some studies have suggested that a limited screening strategy for occult cancer — including history taking, physical examination, routine blood testing, and chest radiography — is adequate to detect most occult cancers, other studies have suggested that a more extensive screening strategy (e.g., incorporating ultrasonography or CT of the abdomen and pelvis, measurement of tumor markers, or a combination of these) can substantially increase the rate of detection of occult cancer.

Morning Report Questions

Q: Does adding a CT of the abdomen and pelvis to a limited screening strategy for occult cancer lead to fewer missed cancers?

A: In the study by Carrier et al., a screening strategy for occult cancer that included comprehensive CT of the abdomen and pelvis did not lead to fewer missed cancers than the number missed with a limited screening strategy. In the primary outcome analysis, 4 of 14 occult cancers (29%; 95% CI, 8 to 58) were missed by the limited screening strategy (i.e., cancer was diagnosed after the screening strategy had deemed the patient as being free from cancer and before the end of the 1-year follow-up period), whereas 5 of 19 occult cancers (26%; 95% CI, 9 to 51) were missed by the strategy of limited screening plus CT (P=1.0).

Q: Does screening with the addition of CT detect significantly more cancers or reduce cancer-related mortality?

A: In the Carrier et al. study, the screening strategy that included CT did not appear to detect significantly more occult cancers (including early cancers), shorten the time to cancer diagnosis, or reduce cancer-related mortality. A total of 14 patients (3.2%; 95% CI, 1.9 to 5.4) in the limited-screening group and 19 patients (4.5%; 95% CI, 2.9 to 6.9) in the limited-screening-plus-CT group received a diagnosis of occult cancer (P=0.28). In the secondary outcome analyses, there was no significant between-group difference in the mean time to cancer diagnosis (4.2 months in the limited-screening group and 4.0 months in the limited-screening-plus-CT group, P=0.88), the rate of recurrent venous thromboembolism (3.3% and 3.4%, P=1.0), overall mortality (1.4% and 1.2%, P=1.0), or cancer-related mortality (1.4% and 0.9%, P=0.75). The rate of detection of early cancers was 0.23% among those in the limited-screening group and 0.71% in the limited-screening-plus-CT group (P=0.37).

Blood Clots and Buried Cancers

Posted by Rena Xu • August 19th, 2015

Insight post 8-19-15An unprovoked blood clot presents a dilemma. Deep vein thrombosis or pulmonary embolism may be the presenting sign of occult cancer, which makes it tempting to search high and low for a source. Early detection could enable prompt treatment and perhaps a better prognosis, in addition to changing the type of anticoagulation a patient receives. On the other hand, exhaustive screening is expensive and, when imaging is involved, exposes patients to radiation that can actually induce cancer. What’s more, there hasn’t been strong evidence historically to suggest that more comprehensive screening translates to fewer missed malignancies.

The NEJM recently reported the results of a large multicenter study that makes a case against more aggressive screening. The Screening for Occult Malignancy in Patients with Idiopathic Venous Thromboembolism (SOME) trial randomized over 850 patients with an unprovoked clotting event to undergo either limited cancer screening – basic labs, a chest x-ray, mammography and Pap smears for women, PSA testing for men – or those tests plus a comprehensive CT scan of the abdomen and pelvis. This scan included virtual colonoscopy and gastroscopy, as well as multi-phased imaging of the liver and pancreas. The primary study endpoint was the number of missed cancers after one year (most occult malignancies get diagnosed within that time frame after a clotting event).

In the two study arms, similar percentages of patients were diagnosed with cancer in the first year (3.2% in the limited screening arm, versus 4.5% in the limited-screening-plus-CT arm). In the limited screening arm, 4 out of 14 cancers, or 29%, were missed by the initial screening. In the limited-screening-plus-CT arm, a similar proportion of diagnoses were missed – 5 out of 19 cancers, or 26%. There was also no difference between the two groups in the average time to diagnosis, which for both groups was around four months.

“In our trial, a screening strategy for occult cancer that included comprehensive CT of the abdomen and pelvis did not lead to fewer missed cancers than the number missed with a limited screening strategy,” the authors write. For patients with a negative result at the time of limited screening, the incidence of a cancer diagnosis later in the first year was less than 1% — similar to the reported incidence among patients without clot.

Previous studies have suggested that in the year after an unprovoked clot, up to 10% of patients are diagnosed with cancer. In this study, across the two groups, the proportion of patients diagnosed was just under 4%. In an accompanying editorial, Dr. Alok Khorana of the Cleveland Clinic suggests that a relatively younger study patient population could have contributed to this discrepancy. The average age of participants in the SOME trial was 54 years; in contrast, in a prior study of over 500,000 patients, the average patient was more than a decade older. The authors, in turn, suggest that the actual prevalence of occult cancer could be decreasing due to better cancer screening.

“Routine screening with CT of the abdomen and pelvis did not provide a clinically significant benefit,” the authors conclude. It’s possible that adding other forms of imaging would have changed these findings, although a third of study participants also got a CT chest as part of their workup for pulmonary embolism. And among the occult cancers most frequently missed were gynecologic tumors and colorectal tumors; if a comprehensive CT scan of the abdomen and pelvis was unable to make the diagnosis in these patients, it seems unlikely that routine imaging of other regions of the body would have helped.

How do you screen patients for malignancy after an unprovoked clot? When would you perform a CT scan as part of your work-up? What other studies, if any, would you consider?

A Girl with a Chest-Wall Mass

Posted by Carla Rothaus • August 14th, 2015

8-Year-Old Girl with a Chest-Wall Mass 8-14-15In a new Case Record of the Massachusetts General Hospital, an 8-year-old girl with placement of a long-term tracheostomy tube presented with a chest-wall mass. Imaging studies revealed a soft-tissue mass, a pleural effusion, and pulmonary consolidation. A diagnostic procedure was performed.

Bacterial tracheobronchitis and pneumonia are common in children who have undergone tracheostomy, with an annual incidence of up to 88%.

Clinical Pearls

Describe some of the features of pulmonary actinomycosis.

Pulmonary actinomycosis is often polymicrobial, with gram-positive bacteria or anaerobes as common coisolates, and its parenchymal component may cavitate. Small-to-medium-sized pleural effusions may occur, and chest-wall invasion may result in early rib erosion.

What is empyema necessitatis?

Empyema necessitatis is a process characterized by extension of pleural empyema into the chest wall. This condition was first described by Gullan de Baillon in 1640. Empyema necessitatis typically develops over a period of 4 to 8 weeks and is associated with pain, swelling, and lymphadenopathy of the anterolateral chest, often without fever. It can also result in a pleurocutaneous or bronchopleurocutaneous fistula. The publication of only a few relevant case reports since the 19th century indicates that this condition remains extremely rare in children.

Morning Report Questions

Q: What pathogens are associated with empyema necessitatis?

A: Empyema necessitatis was more common in the era before antibiotics. At that time, it was associated with an overall mortality of approximately 66%; the most common pathogens were Mycobacterium tuberculosis (mortality, 87%) and Streptococcus pneumoniae (mortality, 28%). Since antibiotics have been in use, cases of empyema necessitatis are rarely fatal, and actinomyces species are a more common cause than is Streptococcus pneumoniae. Less common pathogens include Staphylococcus aureus, Streptococcus milleri, Fusobacterium nucleatum, Mycobacterium avium, Mycobacterium intracellulare, Burkholderia cepacia, blastomyces species, and Nocardia asteroides.

Q: What are some of the tumors included in the differential diagnosis of a chest-wall mass in a child?

A: Benign tumors that occur in children, such as lymphangioma (cystic hygroma) or hemangioma, arise in the soft tissue of the chest wall and are generally confined to it. Patients with an inflammatory myofibroblastic tumor can present with a soft-tissue mass of the chest wall. It arises from lung parenchyma and can extend into the chest wall, with localized mass effect. Chest-wall invasion is inconsistent and generally occurs no sooner than 12 to 18 months after the onset of symptoms, pleural effusions are uncommon, and lymphadenopathy is rare. In children, rhabdomyosarcoma is the most common malignant tumor arising in the soft tissue of the chest wall. Other soft-tissue sarcomas are relatively rare in children. Pleuropulmonary blastoma arises from the pleura or lung and can invade the chest wall, and usually occurs in children 5 years of age or younger. Malignant fibrous histiocytoma is usually confined to the chest wall and is more common in elderly persons than in children. Lymphomas in the thorax typically arise in the mediastinum and can extend into the anterior chest wall.

Table 2. Differential Diagnosis of Acquired Chest-Wall Deformity or Mass in a Child.

Ischemic Limb Gangrene

Posted by Carla Rothaus • August 14th, 2015

Ischemic Limb Gangrene with Pulses for fb-blogA new review article covers ischemic limb gangrene. A variety of systemic illnesses can be associated with limb gangrene with preservation of arterial pulses. Many such disorders involve venous thrombosis caused by a procoagulant–anticoagulant imbalance often related to low levels of protein C.

There are two distinct syndromes of microthrombosis-associated ischemic limb injury. Venous limb gangrene can complicate thrombocytopenic disorders that are strongly associated with deep-vein thrombosis (e.g., cancer-associated disseminated intravascular coagulation and heparin-induced thrombocytopenia). Usually, only one limb is affected. The potentially reversible, prodromal state of limb-threatening ischemia is phlegmasia cerulean dolens. In contrast, two and sometimes all four limbs are affected in symmetric peripheral gangrene, also featuring acral limb ischemic necrosis but usually without deep-vein thrombosis. The two syndromes have common pathophysiological features of microthrombosis associated with a disturbed procoagulant-anticoagulant balance.

Table 1. Two Syndromes of Ischemic Limb Gangrene with Pulses.  

Clinical Pearls

In patients with cancer, disseminated intravascular coagulation, and evidence of a deep venous thrombosis, how does the administration of warfarin lead to venous limb gangrene?

Laboratory studies support a model of profoundly disturbed procoagulant-anticoagulant balance in patients with cancer in whom venous gangrene develops during warfarin anticoagulation. Uncontrolled thrombin generation is shown by greatly elevated thrombin-antithrombin complexes (a marker of in vivo thrombin generation) together with greatly reduced levels of protein C activity — in other words, the ratio of thrombin-antithrombin complex to protein C is elevated as compared with that in controls. In essence, warfarin does not inhibit cancer-associated hypercoagulability while at the same time it predisposes the patient to microthrombosis by depleting protein C activity (often to <10% of normal levels). The characteristic laboratory picture includes thrombocytopenia and an international normalized ratio (INR) that typically exceeds 4.0; a supratherapeutic INR is a proxy for a severely reduced protein C level.

Is warfarin therapy also implicated in venous limb gangrene associated with heparin-induced thrombocytopenia?

Warfarin therapy is implicated in the majority of patients with heparin-induced thrombocytopenia in whom venous limb gangrene develops. Again, a characteristic feature is a supratherapeutic INR. In such patients, a markedly elevated ratio of thrombin-antithrombin complex to protein C supports a model of profoundly disturbed procoagulant-anticoagulant balance. In the minority of patients with heparin-induced thrombocytopenia in whom venous gangrene develops in the absence of warfarin administration, unusually severe thrombocytopenia (platelet count, <20,000 per cubic millimeter) and laboratory evidence of decompensated disseminated intravascular coagulation (e.g., elevated INR, hypofibrinogenemia, and circulating nucleated red cells) is found.

Figure 1. Clinical Profile of Venous Limb Gangrene and Symmetric Peripheral Gangrene.

Figure 2. Changes in Platelet Count and INR in Three Clinical Scenarios Associated with Ischemic Limb Gangrene with Pulses.

Morning Report Questions

Q: What are some of the clinical features of symmetric peripheral gangrene, and what microorganisms are associated with purpura fulminans?

A: Symmetric peripheral gangrene indicates predominantly acral necrosis, which affects the distal limbs (with more frequent and extensive involvement of the feet than the fingers or hands) but sometimes also the nose, lips, ears, scalp, and genitalia. The term purpura fulminans is used when there is extensive, multicentric, nonacral skin necrosis, although patients usually have acral limb necrosis as well. Septicemia and cardiac failure are the most common underlying disorders, and patients usually have metabolic (lactic) acidosis. Concomitant multiple organ failure (e.g., respiratory, renal, and hepatic) is common. Patients with septicemia-associated disseminated intravascular coagulation that is complicated by dermal manifestations usually present with fever, hypotension, and a petechial rash that evolves to more extensive confluent nonacral and acral purpuric areas of evolving ischemic necrosis. Limb ischemic necrosis typically involves lower limbs before upper limbs; approximately one quarter of patients require four-limb amputations. Mortality exceeds 50%. Purpura fulminans in young children and adolescents is usually associated with meningococcemia (caused by Neisseria meningitidis), whereas in adults Streptococcus pneumoniae (pneumococcus) is most often implicated.

Figure 3. Symmetric Peripheral Gangrene.

Q: How are these syndromes treated?

A: Venous limb gangrene and symmetric peripheral gangrene are observed in a small minority (<1%) of patients with disseminated intravascular coagulation, so treatment considerations are based primarily on theoretical considerations and case-based observations, rather than on results of clinical trials. In a patient who is recognized to have phlegmasia or venous limb gangrene, treatment is based on two principles. The first — for a patient with a prolonged INR that is caused by treatment with a vitamin K antagonist — is the administration of vitamin K (at least 10 mg by slow intravenous infusion, with 5 to 10 mg repeated 12 to 24 hours later if the prolongation in the INR persists or recurs). The second is therapeutic-dose anticoagulation. These measures can be limb-saving in a patient with phlegmasia. Treatment of symmetric peripheral gangrene (with or without purpura fulminans) theoretically involves heparin-based anticoagulation and the substitution of natural anticoagulants. In choosing an anticoagulant, many practitioners favor heparin, since its anticoagulant effect can be monitored directly (by measuring anti-factor Xa levels). However, heparin requires its cofactor, antithrombin, and antithrombin levels can be reduced in patients with consumptive coagulopathies, particularly with concomitant liver dysfunction.

New Interactive Medical Case: “A Man with Bizarre Behavior”

Posted by Jennifer Zeis • August 13th, 2015

IMC813A 24-year-old man was brought to the emergency department by ambulance after exhibiting abnormal and agitated behavior. On presentation, he was wearing only underwear and yelling “I am God!” repeatedly. He was alert, awake, and talkative but was not responsive to direct questioning. Although his speech was fluent, its content was illogical and was not directed toward caregivers.

Test your diagnostic and therapeutic skills with this free Interactive Medical Case, and earn CME credit or MOC points.

Interactive Medical Cases are online simulations based on a real patient’s experience of illness. You follow interactive steps through an evolving patient’s history, diagnosis, and management, from presentation to outcome. During the presentation of the case, you access videos, lab results and brief commentary that explain concepts important for diagnosis and treatment.

You can also browse the list of 38 previous Interactive Medical Cases.