Friday, August 30, 2013

Less common opioids

As part of our palliative care week curriculum, we discussed analgesia management at the end of life. I find there are numerous opioids that residents tend to become familiar with and feel relatively comfortable prescribing, and then there are those that we use less frequently. This blog instalment will focus on those used less frequently: buprenorphine, methadone, fentanyl and tramadol.

Approved in 2007 by Health Canada, buprenorphine is a partial mu receptor agonist, kappa receptor antagonist. This is felt to have a beneficial profile for treating opioid dependence, considering, as a partial agonist, it will provide some activity at the receptor, but carry a ceiling effect that would limit the development of serious toxicity. For this reason buprenorphine alone or incombination with naloxone has become a commonly used drug in addictions medicine. Its been shown to have similar efficacy in reducing illicit substance use, associated with less mortality than methadone and can be taken as an outpatient on a non-witnessed protocol (unlike methadone). Its prescribed in dozens of countries around the world, where over 95 thousand patients receive this for opioid dependence in France. Despite the safety profile being improved from other opioides, there are multiple case series of overdose deaths related to intravenous and oral ingestions. In Finland, buprenorphine is a leading drug of abuse, likely due to its higher rate of availability compared to North America.

Fentanyl binds to mu receptors, having a potency of nearly 100 times that of morphine. It was produced in the 1960's, and in the 90's transitioned to a patch form, mixing drug with a gel alcohol. Its felt to offer a unique profile for pain control, in that it does not have active metabolites, provides a steady stream of drug into the blood, and may be safer in patients with renal insuffuciency. A systematic review in the journal of palliative care found that GI side-effects were reduced with fentanyl compared to oral morphine. There are also health concerns with fentanyl, as there have been many reports of accidental overdose. Many of these cases involve children who may mistake the patch as a band-aid and place it on their skin.

Methadone has been around for a while. Created in Germany during world war II it was used on soldiers to treat pain and boost morale. It likely did one of the two. The German scientists that created methadone lost their patent following the war, which allowed American scientists to manufacture the medication widely, selling the patents to various pharmaceutical companies for the steep price of one dollar. Methadone is still inexpensive when compared to its alternatives (buprenorphine). The half life can vary in individuals, but can spans from 12-160 hours. It provides analgesia through acting on opioid and NMDA receptors. Its long half life is what makes it useful in treating opioid dependence, and its been shown in multiple RCTs to be superior to placebo in preventing illicit substance use and decreasing withdrawal symptoms. This comes with the cost of roughly 5000 overdose cases per year in the US. Other adverse effects include QTc prolongation and drug-drug interactions, as it is metabolized via CYP3A4.

Tramadol is being tested to treat many different conditions, from chronic nociceptive pain, to post-herpetic neuralgia and post-traumatic stress disorder. It acts on the mu opioid receptors and inhibits monoamine reuptake (nor-epinephrine and serotonin). Its mechanism of analgesia is not completely understood, but may offer some effect against neuropathic pain. The effects of tramadol have been said to be similar to codeine, for the treatment of moderate to severe pain. Side-effects are common and similar to other opioids, where one study suggested up to 70% of patients developed some kind of side effects, commonly GI upset. Seizures are also possible when taken in high doses. Because of tramadol's activity on monoamine receptors, its use with SSRI's and other medications resulting in increased levels should warrant careful consideration as not to trigger the serotonin syndrome.

These medications are all potentially dangerous, and some require a special license, or suggested clinical education prior to prescribing. Using these medication in the palliative care setting is also different that using these in patients with chronic pain. A recent Cochrane Review found there to be a lack of evidence for the long term use of opioids to treat chronic low back pain. Something to think about:

Cochrane review summary for opioid use in low back pain

Thursday, August 29, 2013

SLE and lung disease

Lupus can present in many ways, involving nearly any organ in the body. Pulmonary complaints are common in lupus and take on different forms. 

Pleural disease can present with chest pain with inspiration, a symptom described in nearly 50% of SLE patients at some point in their disease. Pleuritis can be identified with a rub on clinical exam and may or may not be associated with a pleural effusion. When present, pleural effusions in SLE are typically exudative, with low glucose, low complement levels and positivity for ANA. This disease is usually mild and can be treated with anti-inflammatory medication. However, it has been associated with additional lung disease in SLE including the shrinking lung sydrome (SLS - see below).

Pulmonary infections should be considered, given the relative immunosuppression in SLE patients as well as their typical immunosuppressing drug regiments, these patients are increased risk for infection. Infections typically present as they would in patients without SLE, with cough, dyspnea, and fever. A study looking at risk factors for infection in SLE identified corticosteroids and renal disease as risk factors, and found no association with leukopenia or azathioprine use in isolation. 

Pneumonitis is an uncommon but serious complication of SLE (~1-4%). Short term mortality can reach upwards of 50%. Patients present with significant respiratory symtpoms and ground glass opacities on CT scan, usually bilaterally. This is treated with immunosuppressant, sometimes requiring pulse steroids. 

Interstitial lung disease can occur in up to 10% of patients with SLE. It tends to be a subacute presentation with dyspnea, cough and exertional symtpoms. There is variability in this clinical presentation, where patients may have a fibrotic or inflammatory pattern on imaging. Most commonly patients will have ground glass on CT scans suggesting a non-specific interstitial pneumonia (NSIP) pattern. Those with the fibrotic form can have imaging similar to that of idiopathic pulmonary fibrosis, which is termed a usual interstitial pneumonia (UIP) pattern. Sometimes a lung biopsy is required to confirm the diagnosis, and bronchoscopy can be useufl in excluding infection.

Pulmonary hypertension is a known complication of SLE. It has a similar pathophysiology to that of primary arterial hypertension, which is different than that of left sided heart failure and chronic venous thromboembolic disease. However, as noted in morning report today, patients with SLE may have anti-phospholipid antibodies, putting them at risk for thrombosis and PE. Treatment may require vasodilating agents like bosentan or sildenafil.

Shrinking lung syndrome, characterized by decreasing lung volumes, dyspnea and chest pain. It is felt to be a relatively rare condition, though patients may not be monitored for this disease appropriately with repeated lung volumes. The mechanism may involve worsening diaphragmatic dysfunction though this is somewhat controversial. Some data suggests that nerve conduction to the diaphragm or even chest wall disease may be responsible as opposed to true myopathy of the diaphragm. There have also been multiple studies associating pleural inflammation with development of SLS, pointing to an additional mechanism.

Pulmonary hemorrhage tends to be a more commonly recognized presentation of other rheumatologic conditions (granulomatous polyangiitis, goodpastures syndrome etc), but can rarely occur in SLE. This was documented in 4% of SLE patients in a study spanning ten years. There was a strong association with pulmonary hemorrhage and nephritis, where only one patient who had pulmonary hemorrhage did not have renal involvment. Imaging and bronchoscopy are often indicated to support the diagnosis. 

Acute reversible hypoxemia is a relatively new entity described in the early 1990's. Patient experience transient hypoxemia as a result of leukocyte agreggation and vascular occlusion, which is felt to result from hypercomplementemia and leukocyte activation. See one of the initial articles below from the annals of internal medicine.

Wednesday, August 28, 2013

Sickle Cell Disease

Sickle cell disease (SCD) is a common admission to hospital in the city of Toronto. We have a multicultural population in the city and subspecialty clinics at the University Health Network. These patient suffer from multisystem disease as a result of chronic vaso-occlussion, infections and anemia, often requiring hospitalization for analgesia. However, there is a spectrum of disease activity, and not all forms are equivalent. Different forms of sickle cell disease are based on the specific mutations in the beta globin gene, which can result in varying combinations of alleles Some include:
HbSS - typical form of SCD
HbSC - C genotype is as a different point mutation at the same amino acid residue
HbSthal - patients with one S genotype and beta-thallasemia trait on the other chromosome

These three are the most common forms, but others do exist including:

Hb Lepore disease

These different mutations result in varying disease penetrance. Some patients will have disease very similar to that of typical SCD. Its important to recognize differences in genotype as the frequency of certain sydrome may vary. For example, HbSC is a milder form of disease and hence may present with sequestration crisis later in life (adult years), as opposed to HbSS disease where this occurs solely as a child. Patients with HbSC disease tend to have milder symptoms, developing the same complications but later in life.

Presentations of sickle cell disease are many and can include vaso-occlusive crisis, chest crisis, aplastic crisis and sequestration crisis. Patient present with pain, usually located to bone/muscle in VOC, and chest pain/dyspnea in patients with chest crisis. Triggers for VOC are things that impact the oxygen dissociation curve, which results in deoxygenation of Hgb, polymerization and sickling of red blood cells. This include:

Hypothermia, acidosis, hyoxemia, dehydration, and infection to name the common ones.

Treatment is usually supportive, with investigations targeted towards identifying an infection. Fluids, analgesia, oxygen therapy and antibiotics as needed. Transfusions are reserved for only those who truly require it. Usually a cut-off of 50 is targeted, though discussion with the patients hematologist and blood bank should be had.

An acute chest crisis is a dreaded complication of SCD and a previous blog entry has discussed the entity. See this link for details: Acute Chest Crisis.

The long term management usually involves hydroxyurea, which results in increasing HgF levels which improve oxygen carrying capacity and decrease the percentage of HbSS. This medication is usually inititated when patients experience frequent crises ( 3 or more per year), chest crisis, or severe VOC. It was found to have a survival benefit in these patients (see NEJM link).

Hydroxyurea and SCD

Thursday, August 22, 2013

Pleural tuberculosis

Tuberculosis (TB) can involve nearly any part of the body and can present in a number of ways clinically. Some presentations are classic and allow easy identification of the disease through standard procedure (pulmonary TB - sputum AFB smear/culture), while others require additional testing. Pleural TB is the third most common presentation of TB, following pulmonary and lymphatic involvement. This places pleural involvement at a rate of up to 20% of cases of TB in some studies. Pleural involvement is elevated in patient with HIV, specifically in those with higher CD4, which may point to the pathophysiology of this condition.

Pleural involvement can occur in the setting of primary or reactivation TB. Organisms gain access to the pleural space and stimulate an immune mediated response, particularly lymphocytes (Th1). This results in a delayed hypersensitivity reaction in the pleura. Pleural involvment can manifest as a simple effusion or empyema, which is much less common.

Clinical manifestations can develop quickly in this group of patients, where up to 30% present for medical attention within a week of onset. The most common clinical features include pleuritic chest pain, dyspnea and dry cough. Constitutional symptoms may also be present and are more common in HIV positive patients.

Typically, its said that patients with pleural TB are non-contagious and are unlikely to have positive sputums. that being said, extension into the pleura from parenchymal disease is one possible mechanism for effusion formation. CT scans of patients with pleural TB show up to 80% will have evidence of parenchymal involvement. Also, one study in patients with pleural TB found a 50% positive induced sputum culture rate in this population. Making it a test to be considered in all of these patients.

Pleural fluid analysis reveals several things. pH is rarely >7.4 in TB effusion. The cell count differential can initially show neutrophilic predominance, and will tend to shift to lymphocytic predominance thereafter. Eosinophilia is rare, and suggests an alternative diagnosis. AFB smear of pleural fluid is positive less than 10% of the time, with culture identifying TB in up to 30% of cases. The likelihood of identifying the organism increases in patients with HIV. Adenosine deaminase (ADA), an enzyme in the purine pathway associated with lymphocytes, has been found to be increased in pleural TB. The sensitivity and specificity of this test in patient with a high likelihood of TB is roughly 95% at certain levels. That being said, there are studies showing this can also be elevated in rheumatoid arthritis, lung malignancies and other pulmonary infections. PCR testing is useful, given that there are few bacteria present in the pleural fluid and this may allow magnification of the presence of infection. The Amplified Mycobacterium Tuberculosis Direct Test (AMTD) is a nucleic amplification test specifically for TB. This test is positive in 100% of patients with a positive culture, but only 60% in those who are pleural fluid culture negative. The gold standard historically has been pleural biopsy. Not only does it confirm TB, but it can help identify additional disease that may mimic a TB presentation (sarcoid, fungal infection etc). Thoracoscopy has been shown to be the overall most accurate means of diagnosing pleural TB, though it is also the most expensive.

See this review from CHEST for details:

Pleural TB diagnosis and management

Wednesday, August 21, 2013

Intracerebral hemorrhage management

ent with an intracerebral hemorrhage will sometimes be admitted to the internal medicine service when there is a lack of surgical indication, and/or additional medical problems. As a result, internists need to be familiar with how to manage this condition with the support of neurologists and neurosurgeons. Focusing on the medical management of patients, several categories of therapy have been studied: blood pressure control, coagulation, hyperglycemia and others.

Balancing blood pressure in a patient with ICH attempts to control intracerebral pressure (ICP) and cerebral perfusion pressure (CPP). Hypertension is associated with larger volume ICH and hence functional impairment, and as a result its suspected that lowering blood pressure may improve outcomes. The trade-off being that lowering the blood pressure too much would impair cerebral perfusion and would cause hypoxia in susceptible brain tissue. A study looking at CPP in patients treated with anti-hypertensives in ICH found that dropping systemic BP didnt alter CPP. Though the patients in the study were not significantly hypertensive to begin with. The current guidelines recommend targeting a sysBP of less than 160/90 in patients who do not have evidence of high ICP. If high ICP is presents an ICP monitor may be considered to guide CPP.

A recent study from June 2013 (INTERACT2) examined whether targeting a systolic of less than 140 would be superior to less than 180 in patients with ICH. Overall, there was no statistically significant difference between the two group (p=0.06) in terms of mortality. Functional scores may have been slightly improved in the intensive therapy group. Subgroup analysis showed a possible improvement in outcomes in patients without preexisting hypertension.

Hyperglycemia is associated with worse outcomes in patients with ischemic/hemorrhagic stroke. A general target of less than 10mmol is thought to be appropriate according to current guidelines. Patients with ICH associated with anti-coagulant medications are more likely to have worse outcomes. Treatment with activate FVII has been studied and was associated with less hematoma expansion, and decreased mortality in soem studies. Factor aVII does place patients at increased risk for clotting and should not be given lightly. At certain hospitals, only neurosurgeons and transfusionists can approve the use of this medication. Other less helpful treatments that have been pursued include corticosteroids, which were actually associated with an increased rate of infection.

See the below guidelines for details.
ICH guidelines
See the below guidelines for details.
ICH guidelines

Friday, August 16, 2013

Silent Myocardial Infarction

We are all taught  that myocardial ischemia can present in many ways. The classic retrosternal chest pain, radiating to the neck and left arm is a pattern that is universally recognized amongst the public, but is not the only way a heart attack may manifest.

In 1912, Herrick published an article describing the "clinical features of obstruction of the coronary arteries" in JAMA. There, he describes the entity of unrecognized myocardial infarction. His title still suggests that the pathophysiology in ischemia is the same in patients with silent MI, where poor perfusion to the cardiomyocytes leads to tissue hypoxia and infarction. The difference must be in how this abnormality is perceived and acted upon by patients. Autonomic neuropathy of afferent nerves has been a proposed mechanism, especially in diabetic patients. In 1977, a study looking at 5 patients with silent myocardial infarction and diabetes, and found pathologic changes in the autonomic nerves supplying the myocardium. These changes were consistent with diabetic neuropathy. These changes were not seen in diabetics and non-diabetics in patients with painful ACL presentations. "Gating" phenomenon has been described as another potential mechanism, where muting of afferent pain signals in the dorsal horns by additional sensory input (ex. dypnea) may overwhelm pain input and dampen down the pain perception. Additional factors have focused on supratentorial interpretation of pain, which can be influenced by other medical conditions such as depression. There was a theory that endorphins may play a role in suppressing pain in these patients, however studies involving the administration of naloxone in the setting of silent ischemia in exercise testing had no influence.

Silent MI is likely an under-recognized condition. Data from the Framinham Study suggested approximately 30% of all MI's are silent. This statistic was slightly more common in women. This was also seen in an additional study, where ~20% of patients had silent myocardial infarctions. Q waves on ECG are the most common way to identify this. Considering these ECG changes can resolve in some patients over time, its possible we are missing a portion of silent MI's as a result. Not to mention those that die of sudden cardiac death.

Many studies have found age and hypertension to be associated with silent MI. However, just given these factors are associated in general with coronary artery disease may confound these findings. In a study from Iceland, the risk of silent MI increased by 10% per year of age in those with myocardial infarction. Dementia and cognitive impairment are some of the possible explanations for this. As previously stated, diabetes is a risk factor for silent MI, interestingly diabetics tend to report less pain in confirmed MI as well. A study in JAMA reviewed over 400,000 patients presenting myocardial infarction and found that silent MI was common, and was associated with differences in treatment, where they were less likely to receive aspirin or thrombolysis/PCI. Diabetes was more common in this group as well. The Cardiovascular Health Study found women to be 45% more likely to have unrecognized infarction compared to men, although it was not identified as an independent risk factor. Public perceptions regarding the baseline risk for women to have cardiovascular disease may play a role in the under-identification of MI.

You would think that silent disease would represent milder disease and improved survival, however this may lead to a lack of appropriate therapy. Long term follow up of patients with silent MI show little differences in mortality. The Cardiovascular Health Study showed a 21% vs 25% mortality at seven years in unrecognized and recognized MI respectively. The Honolulu Heart Study actually found an increase in mortality associated with unrecognized MI.

Patients presenting without chest pain but additional concerning features for ischemia need to be investigated. These individuals carry a prognosis similar to recognized myocardial infarction and shouldn't be treated differently. See below for a review article.

Thursday, August 15, 2013


Derived from the term "scleros", meaning thickened, scleroderma is a heterogeneous disease that can effect the skin and multiple internal organs. The terminology is usually categorized as systemic versus localized sclerosis, and within systemic sclerosis there is limited and diffuse disease, which is based on cutaneous distribution.

Localized scleroderma can be subdivided in linear and morphea:

Linear scleroderma - "coup de sabre", which gets its name because it appears like a scar from a blade. This is usually limited to dermatomal distribution and is common in younger patients.

Morphea - can be circumscribed or generalized or focal. There is significant variability within this group, with multiple subtypes. Morphea often spares the face, which is helpful in separating this from systemic sclerosis.

Systemic scleroderma

Limited cutaneous systemic sclerosis (lcSSc) - have cutaneous changes limited to the hands and forearms, on formthat is well described is CREST:
1. Calconisis cutis
2. Reynauds 
3. Esophageal dysmotility
4. Sclerodactyly
5. Telengactasia

Diffuse cutaenous systemic sclerosis (dcSSc) can have additional clinical manifestations including:

1. Vascular abnormalities: Reynauds
2. MSK: myalgias, arthralgias, fatigue. Inflammatory arthritis is uncommon. Osteolysis and bone destruction can occur.
3. Pulmonary: interstitial lung disease or pulmonary hypertension can develop. This can be quite debiliatating for patients. 
4. Cardiac: arrhythmias, pericardial and myocardial disease. Myocardial disease is felt to be secondary to small vessel spasm, similar to Reyanaud's phenomenon. 
5. Renal: "Renal crisis", with azotemia, hypertension and bland urine sediment. 
6. Neurologic: multiple different types of neuropathies have been described, with autonomic, peripheral, cranial and entrapment documented.

Another entity is systemic sclerosis sine scleroderma, which has internal organ manifestations without skin findings.

As mentioned today, there is an increased risk of malignancy, both lung and GI, though the mechanism for this is unclear. Smoking was found to be a risk factor in previous studies.

The case discussed today involved a patient with systemic sclerosis with acute on chronic kidney injury, which always raises the possibility of renal crisis. This is defined by an acute, progressive worsening of renal failure, increased blood pressure and bland urine sediment (mild proteniuria with few casts/cells). Up to 80% of patients with SSc will have renal involvement, with only 20% experiencing renal crisis. Average BP in these patients was found to be ~ 175/105, with signs suggestive of hypertensive emergency (encephalopathy). Additional tesing can reveal microangiopathic hemolytic anemia, with schistocytes and low platelets. Risk factors include severe skin disease, corticosteroid use, cyclosporine use, and anti-RNA polymerase antibodies. 

Wednesday, August 14, 2013

IgG4 disease

An common dilemma in medicine is distinguishing between worsening autoimmune disease and infection in patients with chronic autoimmune disease when they present with worsening symptoms. Increasing dyspnea may represent a worsening inflammation in lungs, versus a pulmonary infection secondary to immunosuppressant drugs. In a recently discussed case,  we encountered this scenario in a patient with IgG4 disease (pancreatitis and sclerosing cholangitis) and a positive cytomegalovirus PCR in bronchoalveolar fluid lavage.

IgG4 related disease is relatively new entity, first being recognized in 2003. Initially felt to be solely with autoimmune pancreatitis, it has now been identified to cause disease in nearly all organs. Its involvement in disease development is currently unclear, where its role as a primary or secondary factor in pathogenesis is yet to be confirmed. Interestingly IgG4 levels in serum and tissue are helpful but not necessary (negative ~30%) to confirm a diagnosis, where additional pathologic features can be helpful. Clinically, patients tend to present with subacute disease involving one ore more organs. Fever and elevated inflammatory markers are uncommon. Organ sclerofibrosis with tumour like lesions and atopic symptoms (asthma, allergy) are a common manifestations. A list of organs that can be involved are listed from a recent NEJM review:

1. Mikulicz’s syndrome (affecting the salivary and lacrimal glands)
2. K├╝ttner’s tumor (affecting the submandibular glands)
3. Riedel’s thyroiditis
4. Eosinophilic angiocentric fibrosis (affecting the orbits and upper respiratory 7.tract)
5. Multifocal fibrosclerosis (orbits, thyroid gland, retroperitoneum, mediastinum etc)
6.Inflammatory pseudotumor (affecting the orbits, lungs, kidneys, and other organs)
7. Mediastinal fibrosis
8. Retroperitoneal fibrosis (Ormond’s disease)
9. Periaortitis and periarteritis
10. Inflammatory aortic aneurysm
11. Tubulointerstitial nephritis with extensive tubulointerstitial deposits

Treatment can be monitoring (in slowly progressing asymtpomatic) or involve immunosuppressants. Prednisone is the mainstay with slow tapers over months +/- sparing agents like azathiaprine, methotrexate or mycophenylate. Rituximab use has also been reported. 

Given the patients medications, infection was considered as the underlying cause, specifically CMV pneumonitis. CMV infection most commonly occurs in the GI tract and lungs, when clinically significant. There are multiple ways to test for infection, including serology, cultures, antigen detection and molecular amplification. There are no consensus guidelines for the diagnosis CMV pneumonitis. PCR testing is quite sensitive, but there are no cut-off levels for diagnosing pneumonitis, where low levels are not interpretable. High levels in the appropriate clincal context chould prompt consideration of gancylovir treatment. CMV is often detected in patients with autoimmune disease without clinical symptoms on BAL (~20% -CHEST 2001). In patients with pneumonitis (dyspnea, hypoxia, infiltrates) immunostaining for CMV in BAL samples has a high sensitivity and specificity and is useful test in confirming disease. This is a challenging scenario and may require further discussion with both ID and respiratory medicine specialists. See below for additional information. 

Friday, August 9, 2013

COPD exacerbations and Antibiotics

In the current era of antimicrobial stewardship, select use of antibiotics is increasingly important. Physicians are aware of the numerous side effects of antimicrobial therapy, increasing resistance patterns and additional costs associated with prescribing, and hence try to target only those patients who will benefit from these medications because of an underlying infection. Infections are felt to cause up to 75% of all exacerbations, and can be due to bacteria, virus' or additional infectious agents. In 2002, a study in NEJM showed that exacerbations were not only associated with bacterial infections, but with new organisms as opposed to those previously seen in colonization.

Antibiotic therapy in COPD exacerbations has been a controversial topic. Ignoring the early, statistically flawed, small sampled trials from the mid-20th century, Anthonisen et al. performed a RCT in 1987 looking at antibiotics in COPD exacerbation. They categorized patients based on the presence of sputum purulence, increased sputum volume, and increased dyspnea. Patients with all of these characteristics had the largest benefit with antibiotics compared to placebo with an >15% reduction in deteriorations. As the number of criteria reduced, there was less of an effect between the antibiotics (doxy, septra or amox) vs placebo. This is why these three qualifiers have become engrained into our COPDE history and used in the consideration of antimicrobials. The GOLD guidelines recommend antibiotics for exacerbations with sputum purulence with increased volume or dyspnea. The severity of illness also plays a role in considering antimicrobial therapy. A Cochrane Review found that patients admitted to the ICU have a significant benefit from antibiotics therapy, but for inpatients without ICU admission the benefits was not as clear. A retrospective study published in JAMA (2010), examined over 80,000 patients who were hospitalized for COPDE. Failure to receive antibiotics in the first 48 hours was associated with increased mortality, repeated admission and mechanical ventilation.

Use of antibiotics for outpatients has less clinical evidence. There have been randomized trials showing showing decreased duration of symptoms (by days) and longer times until repeat exacerbation with the use of Amox-Clav. However, given the individual risks of harm and mild benefits with antibiotics, utility is questionable. Amox-Clav has been compared to other antibiotics (moxi), without showing any significant difference. Duration of five days may be an appropriate length of treatment in the outpatient setting, where a meta-analysis in COPD showed that five days compared to seven days showed no differences in treatment success and lower adverse events.

Overall, there is no specific antibiotic regiment that should be chosen, and patients should have antimicrobials based on their previous history, local resistance patterns and risk factors for pseudomonas. Whether or not antibiotics should be given at all should be based on an individual basis. I also consider additional clinical factors used in the diagnosis of community acquired pneumonia, such as presence of fever and focal consolidation on chest x-ray.

Anthonisen Annals

Thursday, August 8, 2013


A recently discussed case was of a patient with sepsis secondary to fungemia with Candida parapsilosis. Candida in the blood should never be assumed to be contaminant, though this can sometimes be a result of a contaminated line. As with all cases of sepsis, Candidemia has been increasing in the US and Canada.

Candidemia can be invasive or non-invasive, the presence of end organ involvement (endophthalmitis, endocarditis, renal and CNS involvement) defines invasive disease. Diagnosis should be based on positivie blood cultures. Skin biopsies if lesions are present can also be used to identify systemic disease. Recently, assay for beta-D-glucan, a component of the cell wall of Candida has been shown to be useful in diagnosis candidemia. This test is not specific to Candida and will be positive in other invasive fungal infections. This test has been shown to have reasonable sensitivity and specificity for fungal infections. Risk factors identified by several case-control studies in hospitalized and ICU patients included the following:

1. Hickman lines
2. ESRD on dialysis
3. Broad spectrum antibiotics use
4. Gastric acid suppression
5. ICU admission
6. Total parenteral nutrition (present in our patient)
7. Immunsuppression

Candida albicans is the most common species identified in candidemia, however additional strains are being recognized with increasing frequency. C. glabrata, C. parapsilosis, C. krusei and C. tropicalis are amongst the most common species seen. C. glabrata can be resistant to azole therapy, while all C. krusei is resistant to this class of medications. A less common species C. lusitaniae can develop resistance to amphotericin, which should be recocgnized by treating physicians.

Empiric treatment of fungemia should consider multiple factors: history of recent azole antifungals, nuetropenia, severity of illness, local epidemiology, invasive disease and history of resistant organisims in the individual. If the patient was well and had few risk factors, initial treatment with fluconazole IV is reasonable. otherwise, neutropenic/septic patients should receive an echinocandin (caspofungin/micafungin) until speciation occurs. Amphotericin has significant toxicity and tends to be used less. Duration of therapy is based on consensus guidelines, and the IDSA has said 2 weeks following negative blood cultures is a reasonable duration in patient without invasive disease (abscess/endocarditis).

See the link below for candidal guidelines:

IDSA guidlines for candidemia

Wednesday, August 7, 2013

Myasthenic Crisis

Myasthenic crisis is a worsening of muscle weakness precipitating respiratory failure and need for ICU admission and intubation in patients with myasthenia gravis (MG). MG is an autoimmune disorder characterized by antibodies directed against acetylcholine receptors. This results in diffuse muscle weakness, primarily involving bulbar, peripheral and respiratory muscles. Although MG is a rare condition (~5-10/100,000), approximately 15% of patients will develop a crisis during the course of their disease. Despite advances in understanding disease pathogenesis and treatment options, myasthenic crisis carries a mortality of up to 8%. Patients clinical condition can worsen precipitously, sometimes as a result of a triggers and other times without clear cause. Precipitants include:

1. Infection -  usually pnuemonia/aspiration (~70% of causes for crisis)
2. Medications - Tapering of steroids, acetycholinesterase. Or worsening muscle dysfunction from additional medications: anaesthetics, antibiotics, anticonvulsants, antiarrhythmics, beta blockers, magnesium and many others
3. Surgery
4. Pregnancy and childbirth

The cholinergic crisis (worsening of muscle weakness from toxicity due to acetylcholinesterase inhibitors (AChEI)), should be considered in pateints as it can look similar and has a very different treatment approach. Patients on less than 120mg of pyridostigmine q3h are reported to be at low risk of developing this problem.

Investigations that can aid in the diagnosis of myasthenic crisis include: ABG, CXR, vital capacity (VC) and mean inspiratory pressure (MIP). As weakness worsens, an ABG will reveal hypoxemia and hypercarbia. Imaging can identify precipitating infections and alternative causes of dyspnea. A VC of less than 1 litre or 20cc/kg is concerning of imminent failure and warrants intubation. MIP's of less than -30cmH20 should raise concern and lead to endotracheal intubation.

There are several treatment options for myasthenic crisis. These include:

1. AChEI's
2. Steroids
3. Plasmapheresis
4. IVIg

The use of AChEI's in a crisis state is controversial. The awareness of cholinesterase crisis has led many physicians to monitor patients for several days before increasing these drugs. Many prefer to hold these medications and monitor the patient clinically to try and identify the underlying cause, while others continue the medications. There have been randomized studies comparing AChEI +/- steroids, or plasmapheresis, which did not show any additional benefit in patient outcomes.

There is not a large amount of evidence for increasing steroid dosage in acute crisis. Guidelines suggest 60-80mg of prednisone a day and monitoring for acute myopathy from this medication. This should be considered primarily in patients with severe disease in the ICU setting.

Plasmapheresis and IVIg have been promoted as both first line therapies in myasthenic crisis. This idea is that removal or disruption of autoantibodies would result in improvment of muscle function. Both of these treatment options have realatively quick onset of action (days), but do not result in persistent improvement (weeks). Head to head studies have suggested that plasmapheresis has better short term response, but increased side-effects compared to IVIg. Long-term therapy should be considered in conjunction with acute management given this can take time to work (i.e. azathioprine), and the acute therapies will have worn off.

See link below for more info:

Myasthenic crisis

Thursday, August 1, 2013

Pel-Ebstein Fever

This eponym has been raised several times in the month of July in regards to patients presenting with either fever of unknown origin or fever in the context of a previous malignancy. Pel-Ebstein fever refers to a cyclical fever in the context of Hodgkins lymphoma (HL). The description is of fevers lasting days to weeks with defervescence of similar duration. The duration of the absence of fever may decrease as disease progresses. Pal and Ebstein described this separately in two publication in the mid 1880's. Editorials have been published commenting on the original articles, which recognize that the prevalence of HL was not confirmed in many of the subjects described in the original article. It has also become more recognized that other causes of fever can produce a similar pattern, where case reports of CMV and tuberculosis report cyclical fevers (Schattner 2010.). The diagram posted is from a NEJM clinical image from 1995 which charts the fevers of a patient with HL, demonstrating the classic pattern. How often patient with HL present with this pattern is unclear, but its thought to be rare. If nothing else, the eponym has labelled this phenomenon allowing it to be remembered, which may be the only reason its discussed. There are other systemic manifestations of lymphoma that are just as interesting, but don't seem to have the same notoriety.

Alcohol induced pain, usually at the site of involved lymph nodes or bony disease was first described in the mid 20th century. Its reported to occur quickly after the ingestion of alcohol in  less than 10% of patients. Pruritis is a common manifestation of lymphoma, where ~10-15% will experience this, sometimes being quite severe. It may be an early presenting symptom and when severe may carry some prognostic value. I would argue that pruritis is less commonly mentioned as a symptom of HL than pruritis, and may be a more valuable piece of knowledge.