SSRIs: Do They Increase Rates of Suicide?

January 25, 2008

prozac.jpgCommentary by Arthur Sinkman MD, NYU Department of Psychiatry

Three years ago the FDA began requiring that all selective serotonin reuptake inhibitors (SSRIs) carry a black-box warning stating that their use in children and adolescents is associated with an increase in risk for suicidal thinking, feelings and behavior. Recently the FDA ordered that this warning be extended to include treatment for young adults aged 18 to 24.

The 2004 order had a dramatic impact on the treatment of depression in children. The use of SSRIs dropped precipitously, and the overall diagnosis and treatment of pediatric depression also decreased. Clearly physicians and families had been scared away from treating depression. A subsequent increase in the rate of pediatric suicide has been observed.

The FDA’s action was controversial from the start, with several well-conducted studies coming to the opposite conclusion: that SSRIs prevent suicide rather than cause it. The latest research in this burgeoning controversy is a recent group of studies published in the American Journal of Psychiatry.

The first report examined a very large series of cases from a pre-paid health plan and evaluated the rate of suicide attempts both before and after the initiation of treatment.(1) It was found that suicide attempts were highest in the month prior to the start of treatment, second highest in the month after treatment was begun and steadily decreased thereafter. The same pattern held for other treatment modalities, both pharmacologic and nonpharmacologic. The authors concluded that suicidality after starting medication was part of the pattern of the illness and not a direct result of the treatment.

Another study examined an enormous database of the treatment of depression of a quarter of a million veterans(2). It showed that the rate of suicide was much lower in the group that took SSRIs in comparison to those who did not take medication. The suicide rate was also lower in the SSRI group than in the group that took other types of antidepressants. In addition, this study also repeated the findings of the aforementioned study in that the incidence of suicide was higher before treatment than after.

The editorial accompanying the above reports concluded that “it is more likely that suicidal behavior leads to treatment than that treatment leads to suicidal behavior.”(3)It should be noted that these studies mainly focused on adults; thus, the relevance for the treatment of pediatric depression is less clear. However, the VA study did include young adults and its findings are relevant for the newest group of patients to be included in the black-box warning.

A follow-up report in the same journal added further ammunition to those opposed to the black-box warning.(4) It was done by the same research group responsible for the VA report. This study analyzed the prescription rates of SSRI antidepressants and the rates of suicide both before and after the black-box warning. Data from both the U.S. and the Netherlands was examined. A clear correlation between the decreased prescriptions and increased death by suicide was observed. It was also noted that various limitations in the meta-analysis that the FDA used as the basis for its warning exist. Most telling is that there were no completed suicides in this meta-analysis. Rather, there was an increase in the rate of suicide ideation and gestures.

What conclusions can be drawn? One is that withholding an effective treatment is a decision that may have devastating consequences. Although there is a question about the danger of antidepressants, the risks associated with not treating depression are well known and not open to question. The other conclusion is that the use of SSRIs may be far less dangerous than the FDA’s black-box warning implies. Nevertheless, some caution is indicated, particularly when treating children. Prudent behavior includes close follow-up after initiating treatment, considerations for referral for other treatments aside from medication, and evaluation for the presence of side effects, including akathisia, the unpleasant subjective sensation of inner restlessness. Akathisia is usually associated with the use of neuroleptics such as Haldoperidol, but it not infrequently occurs with SSRIs and is often missed. Akathisia can lead to great discomfort and possible suicidal thinking and behavior.


1. Simon GE and Savarino J. Suicide attempts among patients starting depression treatment with medications or psychotherapy. Am J Psychiatry. 2007;164:1029-1034.

2. Gibbons RD, Brown CH, et al. Relationship between antidepressants and suicide attempts: an analysis of the Veterans Health Administration data sets. Am J Psychiatry. 2007;164:1044-1049.

3. Brent D. Antidepressants and suicidal behavior: cause or cure? Am J Psychiatry. 2007;164:989-991.

4. Gibbons RD, Brown CH, et al: Early evidence on the effects of the regulators’ suicidality warnings on the SSRI prescriptions and suicide in children and adolescents. Am J Psychiatry. 2007;164:1356-1363.


Breaking News: A Disappointment for Zetia

January 15, 2008

zetia.jpgCommentary by Alana Choy-Shan MD, NYU Chief Resident

The highly anticipated results of the ENHANCE trial will likely be presented at the American College of Cardiology meeting in March, but the preliminary results already have everyone talking. ENHANCE was a multinational, double-blind randomized controlled trial sponsored by Merck and Schering-Plough (the manufacturers of the combination ezetimibe/simvastatin pill). A total of 720 patients who were heterozygotes for Familial Hypercholesterolemia were randomized to treatment with either ezetimibe/simvastatin 10/80mg or simvastatin 80mg. The primary endpoint was change in mean carotid intima-media thickness (IMT) after two years of treatment. Combination therapy resulted in a change in mean carotid IMT of 0.0111mm versus 0.0058mm with simvastatin alone (p=0.29). LDL levels, which at baseline were approximately 320mg/dL in the two groups, were lowered by 58% with ezetimibe/simvastatin and 41% with simvastatin (p<0.01). Based on these results, the addition of ezetimibe to high dose simvastatin appears to improve LDL levels without a concordant impact on carotid IMT in this population. Though the carotid IMT results are not statistically significant and the study population was a particularly high risk group by virtue of their genetic make-up, the lay press is already reporting that ezetimibe fails to slow atherosclerosis and may actually promote it. Not surprisingly, Merck and Schering-Plough share prices have already fallen – probably not quite the ENHANCEment they were hoping for…

NY Times Link

Schering-Plough Release

Evaluation of Asymptomatic Wolff-Parkinson-White EKG Morphology

January 3, 2008

800px-wolff-parkinson-white_syndrome_12_lead_ekg.pngCommentary by David Steckman MD, PGY-2 and William Slater MD, Associate Professor of Medicine, Division of Cardiology

Case: A 42 year-old man presents to clinic for routine follow-up. He is found to be hypertensive for the second consecutive clinic appointment. On routine EKG, you find a shortened PR interval and what looks to be a delta wave in V1-V3. The patient does not report any history of syncope, chest pain, palpitations or shortness of breath. In addition to treating his hypertension, what further work-up is required for his asymptomatic pre-excitation pathway?

The Wolff-Parkinson-White (WPW) EKG morphology was first described in 1930 in 11 patients that had recurrent episodes of tachycardia and a characteristic EKG pattern of a shortened PR interval and a delta wave. The current prevalence of WPW on surface EKG is thought to be 0.15 to 0.25% in the general population. There is a 4-fold increase among family members of an affected patient (1). WPW syndrome is reserved for those patients with both EKG evidence of pre-excitation and tachyarrhythmia, most commonly A-V reentrant tachycardia (AVRT) accounting for 95% of associated tachycardia. Atrial fibrillation (AF) is a potentially life-threatening arrhythmia in patients with WPW syndrome. If an accessory pathway has a short anterograde refractory period, then rapid repetitive conduction to the ventricles during AF can result in a rapid ventricular response with subsequent degeneration to ventricular fibrillation (VF). The incidence of sudden cardiac death (SCD) is estimated to range from 0.15 to 0.38% over 3 and 10 years of follow-up (1). Although cardiac arrest is rare in patients with known WPW syndrome, in approximately half of patients with an unknown accessory tract, the first presenting symptom is cardiac arrest. Factors associated with increased risk of SCD include younger age, male gender, high adrenergic state, multiple accessory pathways, ease of inducing AVRT with electrophysiologic (EP) testing, and short R-R periods during atrial fibrillation (1-3).

Patients with a WPW pattern on ECG can be divided into 3 groups, with different considerations for management: asymptomatic patients, those with spontaneous SVT, and those presenting with AF. We will consider these groups separately.

It is controversial whether patients presenting with asymptomatic EKG findings suggesting pre-excitation (electric stimulation of the ventricles via an accessory pathway) require further testing and subsequent therapy. Current guidelines suggest risk stratifying patients into groups requiring further testing and those requiring simple close observation. The best indicator of low risk is the abrupt disappearance of pre-excitation during exercise, indicating a long refractory period of the accessory pathway (3). This demonstrates that the accessory pathway would be less capable of transmitting a rapid atrial rate to the ventricle in AF. In addition, an intermittent delta wave on a routine EKG indicates longer refractory times and thus implies very low risk. One must be careful to distinguish intermittent pre-excitation from ventricular bigeminy with loss of delta waves on the PVC beat.

Patients with asymptomatic WPW may undergo non-invasive testing including stress testing or Holter monitoring (i.e., to look for intermittency of pre-excitation). If these studies do show EKG findings suggestive of longer refractory periods, such patients may be discharged safely with an information packet, a copy of the diagnostic EKG and strict counseling to return to the ER if symptomatic arrhythmias occur. Exceptions are made for those patients with high risk occupations (bus driver, scuba diver, pilot). Many clinicians may start low dose beta-blockers if the accessory pathway is proven during EP testing to be incapable of rapid conduction. There have been no randomized-controlled trials of drug prophylaxis for AVRT.

Read more »

Clinical Question: Pharmacology

December 28, 2007

lantus.jpgIs there evidence to support the use of Lantus® (human insulin analog glargine) administered Q12h in Type 1 Diabetes?

Commentary by Kathy Lee, Pharmacy Resident 

The goal of diabetes management is to reduce the risk of long-term complications by maintaining near-normal glycemic control, in addition to reducing other risk factors. Patients with type 1 diabetes have an absolute deficiency in insulin and require exogenous insulin replacement. Lantus®, human insulin analog glargine, is the only long-acting insulin that exhibits a “peakless” action profile with duration of action of up to 24 hours(1). Because the clinical trials demonstrating its safety and efficacy involved a once a day bedtime administration, Lantus® received a FDA indication for once-daily dosing for diabetic patients who require basal insulin.

Theoretically, since insulin glargine has a 24-hour action profile with no pronounced peak, the time of administration should not be significant. A 24-week randomized clinical trial(2) investigated this idea by comparing administration of glargine at breakfast, dinner, or bedtime in patients with type 1 diabetes. The study concluded that the efficacy and safety of insulin glargine was not affected by the timing of administration. Despite limited studies there have been some evidence implying the appropriateness of twice-daily insulin glargine in select type 1 diabetic patients.

In August 2002, a patient case regarding the use of twice-daily insulin glargine was published(3). In this case, a 53 year old male with a history of type 1 diabetes did not receive sufficient insulin coverage with a once daily injection. Before hospitalization, the patient had a history of widely variable blood glucose levels. He was on a 4-injection regimen of premeal insulin lispro and ultralente at dinner. The patient also had a history of heavy alcohol abuse and on admission, the patient developed left arm weakness and progressive loss of consciousness. He was diagnosed with compensated diabetic ketoacidosis, and after treatment, was maintained on IV insulin infusion between 1-2 units/h. On the fifth day of his hospital course, the patient was started on enteral feeding with total nutrition intake of 1680 kcal/day. He was then later given 30-unit dose of insulin glargine at 9:00 p.m. and was weaned off of the insulin infusion over the next 4 hours. From day 6 to day 12 (period 1), the patient received insulin glargine as a single dose. In addition, he received subcutaneous insulin lispro when glucose exceeded 200 mg/dl. Marked hyperglycemia was noted at 10:00 p.m. after the single-dose regimen, requiring that the patient receive a supplemental lispro injection 4 out of the 6 days. From days 13-18 (period 2), glargine was divided into two doses, administered at 9:00 a.m. and 9:00 p.m. The mean glucose level was significantly higher at 10:00 p.m. in treatment period 1 versus period 2. This data suggested that the slow onset of action for the subsequent nighttime dose resulted in higher blood glucose levels at night. Giving insulin glargine as a split dose every 12 hours eliminated the window of insulinopenia and provided effective 24 hour insulin coverage.

Read more »

Inpatient Diabetes Management: Case 7

December 20, 2007

90px-glukometr_ot.jpgCommentary by Mary Vouyiouklis MD, Fellow, and Ann Danoff MD, Director, Division of Endocrinology, Diabetes and Metabolism, NYU Medical Center

Welcome to our final case of our special diabetes series intended to highlight the essentials of diabetes care in the inpatient setting. Over the last few months, we have been presented individual cases followed by some management questions and answers.

Case 7: All our patients go home…
Mr. Smith, Mr. Jones, Mr. Mejia, Mr. Gary, Ms. Samson, and Ms. Longshore are all awaiting discharge. What are the crucial elements for a diabetic’s safe discharge?

A. Specific outline of medication. (i.e. “Take 10 units rapid acting insulin pen prior to your meal. You must make sure your meal is fully prepared and you are ready to eat before you take this type of insulin.”)
B. Inpatient Nutrition consult
C. Inpatient Nurse education regarding fingerstick monitoring, treatment and prevention of hypo- or hyperglycemia.
D. Social work and Case managers’ coordination of home care if necessary.
E. Early phone contact and outpatient re-visit with PMD.
F. ALL of above.

Answer: F

It is essential that the patient has had proper education about his/her diabetes prior to discharge. This must be considered on the day of admission to prevent a last minute scramble on the day (or the hour!) of discharge.  Most patients with newly diagnosed diabetes are quite overwhelmed by the diagnosis.  It is unrealistic to expect most people to learn all of the information that will ultimately be required for mastery of necessary self-management skills.  However, it is critical that they have acquired a minimum of “survival-management” skills education, including how to recognize and treat hypo- and hyperglycemia, how to perform self-monitoring of blood glucose, how to inject insulin and/or take medications (as indicated), a rudimentary grasp of proper nutrition, and how to contact their health-care provider in an emergency or with questions.  In addition, it is important to give an early follow-up (within two weeks) to review the patient’s fingerstick log so that any in-hospital insulin adjustments can be altered if necessary.

FDA Warns of Possible Link Between Chantix Therapy and Mood Disturbances

December 14, 2007

chantix.jpgCommentary by Robert Leonard PharmD, Pharmacy Resident New York Harbor Healthcare System

On November 20, 2007 the FDA announced mounting evidence linking varenicline (Chantix®) therapy for smoking cessation with suicidal ideation and erratic and aggressive behavior. The announcement comes in response to post-marketing case reports submitted to the FDA by the makers of Chantix®, Pfizer Inc. Early review of the cases reveal new onset of depressed mood, suicidal thoughts, and changes in emotions or behaviors within days to weeks of starting Chantix® therapy. A direct causal relationship remains limited at this time due to the nature of smoking cessation and the emergence of withdrawal symptoms with or without treatment. Nicotine withdrawal alone has been associated with acute exacerbation of underlying psychiatric disorders. Further confounding the interpretation of these reports is the presence of co-morbid conditions often associated with tobacco use disorder that predispose patients to mood disorders such as alcohol dependence. However, despite these limitations, reported cases include mood disturbances in patients who have no prior history of psychiatric illness and in patients who continue to smoke.

Pfizer Inc. has also submitted case reports to the FDA describing drowsiness associated with Chantix ®that influences the ability to safely drive or operate machinery.

In light of these preliminary findings, the FDA has made three recommendations concerning the use of Chantix® for smoking cessation. The official recommendations reported in the press release by the FDA are as follows:

“Healthcare professionals should monitor patients taking Chantix for behavior and mood changes. Patients taking Chantix should contact their doctors if they experience behavior or mood changes. Patients should use caution when driving or operating machinery until they know how quitting smoking with Chantix may affect them.” (1)

The FDA, in conjunction with Pfizer Inc., will continue to review data concerning the safety of Chantix®, the potential association with mood disturbances and the impact on driving or operating machinery. Patients and healthcare providers are encouraged to report adverse affects with Chantix® to the FDA via the MedWatch Adverse Event Reporting program available online at

1. Food and Drug Administration. Early communication about an ongoing safety review varenicline (marketed as Chantix). Available at Accessed November 21, 2007.

Meeting Perspectives: Annual Meeting of the Infectious Diseases Society of America (IDSA) October, 2007

December 13, 2007

idsa.jpgCommentary by Neal Steigbigel MD, Professor of Medicine (Infectious Diseases/Immunology)

Methicillin resistant Staphylococcus aureus (MRSA) was a hot topic at this year’s IDSA meeting. The CA (community-acquired)-MRSA strains are spreading rapidly in the United States causing largely soft tissue infections, but also other infections such as serious necrotizing pneumonia. Clinicians can no longer rely on demographics in predicting which patient is at high risk for MRSA when an individual appears with a suspected staph infection. If the infection appears serious and is likely to be caused by S. aureus, treat them all as if they are MR and make antibiotic changes after culture and sensitivities are back. That means typically starting with intravenous vancomycin at the usual doses.

However, vancomycin therapy for MRSA infections has been shown in recent years to be too often associated with sub-optimal outcomes, especially when the in vitro MIC of the isolate is at 2ug/ml or above; the current MIC cut off for labeling the isolate as “sensitive” to vancomycin is 2ug/ml or below. Due to this observation, there is a new need for our microbiology laboratories to start including the vancomycin MIC’s in the sensitivity reports for S. aureus. However, there is no alternative antibiotic that is definitively better than vancomycin when the organism is in the sensitive MIC range (2 ug/ml)—only guesses based on pharmacokinetic (PK) data and expert opinion. If the MIC of the isolate is 2 ug/ml, population analysis done by in vitro research studies show that many of these strains are “heteroresistant” to vancomycin, meaning a given isolate has subpopulations that are clearly in the “resistant” MIC ranges (such as 4-8 ug/ml). So why not use higher doses (>2gm/24h) of vancomycin, which has been recently proposed, to achieve trough vancomycin levels of 15-20 ug/ml in patients, for which PK calculations suggest would lead to more effective results against strains of S. aureus that are “heteroresistant?”

There are at least two problems with this approach. First, at the IDSA meeting, George Drusano’s Albany group showed a retrospective study of patients divided into those who received high and usual (2g/24h) doses of vancomycin and found that the high dose group had a significantly higher likelihood of developing nephrotoxicity. Second, PK calculations indicate that to achieve trough levels of vancomycin in the 15-20ug/ml range, the needed dose for a S. aureus strain with an MIC of 2ug/ml or more would be unacceptably high—well over 4 gm per day.

What about alternative antibiotics such as daptomycin or lineziolid? Linezolid is expensive and can cause bone marrow suppression, especially when used for more than 10 days, and is poorly bacteristatic, a potential problem if the patient has endocarditis. Daptomycin is very bactericidal and has many other good characteristics, including a good clinical study showing at least non-inferiority to vancomycin in S. aureus infections, including right sided endocarditis. It can’t be used in bronchial pneumonia because it is inactivated by surfactant.

My opinion until potentially better agents are available (telavancin, dalbavancin, and ceftobiprole are undergoing study for a non-pneumonia patient with a serious MRSA infection who is not doing well on vancomycin and a MIC of at least 2ug/ml, is to switch to daptomycin at 6-8mg/kg daily, assuming normal renal function. For serious MSSA infections nafcillin or oxacillin are still the antibiotics of choice. For dialysis patients with MSSA bacteremia, cefazolin, 1gm/24h is more effective than vancomycin.

For patients with catheter related MSSA or MRSA bacteremia, the “expert” suggestion based on reasonable evidence is to rapidly remove the catheter, culture it, perform TEE in all patients (substantial incidence of vegetations) (not just TTE –too many false negatives) and if TEE negative, along with no other clinical evidence of other metastatic infection and with bacteremia and fever lasting less than 72h, continue the IV antibiotics for the S. aureus for 14 days (not just 10 days). If the patient does not meet that 14 day criteria, treat IV for 4 weeks. Recent studies indicate that catheters inserted properly in ICU’s (hand washing, sterile gloves, chlorhexidine skin prep, and other proper barriers) can lead to an eradication of catheter-related infections.

Read more »

Ethical Dilemmas: Artificial Feeding

December 7, 2007

ethics.pngCommentary by Shrujal Baxi MD, Ethics Section Editor, and Joseph Lowy MD, Medical Director, Palliative Care Service at NYU Hospital

Mrs. Pleasant is an 85 year old woman brought into the ER by her son for poor appetite. Mrs. P is demented secondary to progressive Alzheimer’s disease and has had 3 hospitalizations in the last year for pneumonia, possibly related to aspiration.  Upon physical exam, she is oriented to herself, but is unable to provide any other information. She denies pain and appears to be in no distress.  Her vital signs are within normal limits.  She appears cachectic with temporal and thenar wasting.  Remainder of physical exam is noncontributory.  Laboratory evaluation reveals normal values with the exception of an elevated BUN/Cr consistent with dehydration.  Mrs. P is admitted for IV hydration and evaluation for weight loss. After a thorough work-up, it is determined by her multidisciplinary team, that worsening dementia is the cause of her anorexia.  A family meeting is called to create a treatment plan.  Mrs. P’s son believes that his mother would not want a feeding tube.  Her son is all the family that Mrs. P has and she unfortunately did not sign a living will or a health care proxy.  The team calls an ethics consult to determine what to do next.

Dr. Joseph Lowy:
Decreased oral intake due to anorexia and/or dysphagia is a predictable terminal stage of advanced dementia.  Pneumonia related to aspiration or not, is a common complication.  Anorexia is not an apparent source of suffering for patients with advanced dementia.  Nasogastric tubes for feeding, on the other hand, may be.  Seventy percent of patients with dementia in whom NG tubes are placed require restraints at some point to prevent removal.  Artificial feeding (by feeding gastrostomy) has never been shown to prevent pneumonia and NG tubes predispose to aspiration.

In one survey of nursing home patients, more than two-thirds said they would not want artificial feeding tubes in the setting of advanced dementia.1 In this study, 33% of residents initially said they would want feeding tubes if in an impaired mental state, but 25% of those patients changed their minds and said no to feeding tubes when they learned all that was involved such as the possible placement of restraints.   In the case of Mrs. Pleasant, it is suggested that this patient does not have decision-making capacity, but it should be noted that this capacity is specific to the decision to be made.  Patients disoriented to time and place may still understand a discussion of artificial nutrition and hydration or other life sustaining treatments, and may be deemed to have capacity to comprehend the consequences of their decisions.  In the absence of a living will, verbal statements, or a designated health care agent (pursuant to a Health Care Proxy in NYS), most states allow for a surrogate decision maker to provide substituted judgment for health care decisions in addition to DNR.  New York State does not.

New York State mandates the provision of artificial nutrition for patients who are not receiving ‘sufficient’ nutrition unless there is ‘clear and convincing’ evidence that they would not want it. While the meaning of ‘sufficient’ nutrition is gray and cloudy, the definition of clear and convincing evidence is much more black and white.  This evidence must be either a written statement, such as Living Will, or a verbal statement to which a witness is willing, if necessary, to sign an affidavit.

The belief of the son (who is not a health care proxy) that his mother would not want artificial nutrition doesn’t meet this standard.  However, a patient taking any fluids and/or food need not receive artificial nutrition and hydration if this is not deemed in the best interest of the patient and careful consideration of the risks and benefits as outlined in the first paragraph.  Therefore, the patient should be provided with any fluids or food she is willing and able to handle without the need for the insertion of a feeding tube.

1.  Linda A. O’Brien RN, MA, et al. Tube feeding preferences among nursing home residents. J Gen Int Med. 1997 June; 12: 364-371.

Image courtesy of Wikimedia Commons

Grand Rounds: “Innate Immunity and Viral Pathogenesis”

December 4, 2007

Bellevue AmphitheaterCommentary by Urania Rappo, PGY-2

This week’s Medicine Grand Rounds guest lecturer was Dr. Robert Finberg, currently Chair of the Department of Medicine at the University of Massachusetts Medical School. He earned his MD from Albert Einstein College of Medicine and trained in Medicine at Bellevue Hospital starting in 1974. He was a Fellow in Infectious Diseases at Harvard Medical School, and there established a rich research career over the ensuing twenty years. Dr. Finberg’s research focuses on host-microbial interactions, defining the cell surface proteins that, like CD14 and Toll-like receptors, control host responses to pathogens. He accepted his current position as Department Chair at the University of Massachusetts in 1999. His Medicine Grand Rounds Lecture was entitled: “Innate Immunity and Viral Pathogenesis.”

Dr. Finberg began with a case vignette of an infant who was admitted to the hospital unresponsive with bleeding from the nares. The infant was found to be in septic shock from HSV-1, but the mother was HSV-1 negative. After further investigation, the father was found to have an HSV lesion on his finger, which he used to feed the infant. Why do infants die of viral sepsis, while adults rarely do? Toll-like receptors (TLRs) are part of the answer.

TLRs are a group of ancient pattern recognition proteins found in many diverse species, from lilacs to humans, and discriminate between self and non-self. In mammals, TLRs interact with products of infectious agents to activate cytokine gene transcription, thereby activating the innate immune system (macrophages, NK cells and neutrophils) which then stimulate the adaptive immune system. Mammalian TLRs were initially noted to stimulate proinflammatory cytokines (IL-1, TNF-α, and IL-6), which cause fever and septic shock during bacterial and fungal infections. TLRs were later also found to interact with viral proteins and initiate fever, and sometimes hypotension and a shock-like picture. They also regulate the production of type 1-IFNs which defend against viruses. Not surprisingly, viruses have found ways to inhibit TLR-specific responses, and the interaction between viruses and TLRs is critical in understanding viral pathogenesis and immunity.

Neonates, unlike adults who develop only localized disease, can develop a sepsis-like picture likely from activation of TLR2 on macrophages causing a “cytokine storm.” Neonates appear to have an enhanced ability to respond to TLR2 ligands, and are thus more susceptible to an overwhelming and harmful immune response. Thus, it appears that TLR activation is a “double-edged sword.”

The next logical question would be: would a TLR2 inhibitor prevent this lethal response? It is possible that a TLR2 specific inhibitor might have some effect but the response to many infections involves multiple TLRs as well as cytoplasmic DNA and RNA receptors. Along with biotechnology companies, Dr. Finberg is investigating TLR agonists, including TLR activating compounds, to increase viral-specific immunity and as vaccine adjuvants.

Perhaps TLR inhibitors can be designed to dull the sword edge that occurs with viral hemorrhagic fever and causes damage in viral encephalitis, yet preserve the edge that is an ancient defender against microbial invaders.


Finberg RW, Wang JP, Kurt-Jones EA. Toll like receptors and viruses. Rev Med Virol. 2007;17:35-43.

Grand Rounds: Breast Cancer Genomics

November 20, 2007

Bellevue AmphitheaterCommentary by Jonathan Willner MD, PGY-2

This week’s Medicine Grand Rounds speaker was Lisa Carey, MD, Associate Professor in Hematology/Oncology at the University of North Carolina and Medical Director of the UNC Breast Center.  Much of Dr. Carey’s research focuses on how an understanding of breast cancer genomics may tailor clinical therapy.

While the incidence of breast cancer has plateaued over the past few years, there has been a decline in the number of breast cancer deaths. The reason is thought to be two-fold: improved screening and more effective medical therapy.  As treatment has improved, so too has the variability in choice and duration of therapy. However our ability to personalize adjuvant therapy so that women are receiving only the drugs they need is limited; most women are either over or under-treated.

The traditional model of cancer development holds that malignancy develops as a linear progression of genetic “hits:” successive loss of genetic integrity over time allowing for unchecked proliferation and spread of malignant cells. The biologic model, by contrast, argues that inborn traits and biologic variability sum to create a cancer subtype, which is later evidenced by the phenotypic variability we see in disease progression and treatment response.  The biologic model has been largely informed by genomics, the study of large-scale genetic mapping, as its tool for defining breast cancer subtypes. Analysis of approximately 8000 genes via microarray has elucidated a gene sets that reliably identify several breast cancer subtypes. These subtypes are definable early in the course of disease, and are present both after chemotherapy and in metastatic disease.  Dr. Carey’s work has focused on the ‘basal-like’ subtype in particular, so named because of its high density of genes coding for EGFR, basal cytokeratins, and other basal proteins. The basal-like subtype also has typically low HER2 and ER/PR expression, and is very proliferative.

Breast cancer subtyping being increasing used to predict the clinicopathologic characteristics of a patient’s disease. Population-based studies, such as the Carolina Breast Cancer Study, have found a preponderance of the basal-like subtype among pre-menopausal, African-American women. Patients that carry germline BRCA1 mutations also tend to develop the basal-like subtype. Somewhat surprisingly, traditional breast cancer risk factors (degree of parity or OCP use, for example) fail to predict ER-negative subtypes, and some factors may actually be protective in some subtypes.. Such variability has led researchers to try to identify particular gene profiles that can be used to predict outcome across all subtypes.  From these studies, the basal-like subtype has emerged as the subtype most powerfully predictive of clinical progression and outcome.

Breast cancer relapse is known to be heterogenous. While ER-positive breast cancer tends to maintain a low-but-constant relapse rate, the basal-like subtype typically evinces a high early rate followed by a rapid decline in number of relapses. This may suggest why these two subtypes have different responses to chemotherapy: chemotherapy tends to decrease early relapses and is more effective in ER-negative subtypes like the basal-like subtype; ERER-positive breast cancer has a constant risk of  relapse over many years and is less impacted by chemotherapy.

Lastly, though genomic subtyping has shown promise as a model for predicting clinical outcome, it has yet to prove itself as a tool for choice of medical therapy. A number of agents have shown promise in the treatment of basal-like breast cancer, but it has not yet been proven that chemotherapy regimen  can be tailored on the basis of genetic subtyping. We do not yet have known targeted drugs for this subtype, which is the subject of active study.   Future efforts will work to clarify risk factors by breast cancer subtype, which may inform either directed lifestyle or chemopreventive strategies, improving chemotherapy, and identifying targeted drugs to effectively treat with a minimum of toxicity.

Treatment of obesity with bariatric surgery: evidence and implications

November 15, 2007

vertical_banded_gastroplasty.jpgCommentary by Jatin Roper MD PGY-3 and Christine Ren MD Associate Professor, Department of Surgery

Bariatric procedures to treat obesity involve the restriction of the gastric reservoir, bypass of part of the gastrointestinal tract, or both. Worldwide, an estimated 300 million people are obese, and in the United States, the percentage of adults who are obese increased from 15% in 1995 to 24% in 2005. (Obesity is defined as a Body Mass Index or BMI of 30 or more, measured as kg / m2; for example, a woman 5 feet, 4 inches tall weighing 175 pounds.) Obesity is associated with, but not limited to, increased risk of coronary artery disease, stroke, diabetes, obstructive sleep apnea, osteoarthritis, several forms of cancer, and depression, as well as decreased quality of life.

Non-surgical therapy for obesity

Therapy for obesity has included dieting, exercise, and medications. However, reduction in caloric intake combined with exercise typically results in 5-10% reduction in weight (roughly 10-15 pounds) over a number of months, which is rarely sustained. There are no truly effective pharmacologic therapies for obesity; the only currently FDA-approved compounds, sibutramine (trade name Meridia), orlistat (trade name Xenical), and phentermine are limited by side effects and rebound weight gain after discontinuation.

Types of bariatric surgery

Surgical modalities of bariatric surgery can be grouped by their intended physiologic effects: Vertical-banded gastroplasty and adjustable gastric banding involve gastric restriction, while jejunoileal bypass and bilio-pancreatic diversion induce malabsorption; Roux-en-Y gastric bypass contains components of both of these basic types. Adjustable gastric banding and Roux-en-Y gastric bypass are now the favored procedures, and are often performed laproscopically. In gastric banding, a band is placed around the proximal portion of the stomach. In gastric bypass, the stomach is divided into a small proximal gastric pouch and the remaining main stomach is bypassed. The gastric pouch is connected to the small bowel by a jejunal segment, called a Roux limb.

Clinical evidence

A meta-analysis of 136 studies involving 22,094 patients by Buchwald et al. (JAMA 2004;292(14):1724-37) found a peri-operative mortality of 0.1% of purely restrictive procedures and 0.5% for gastric bypass. Almost all patients pooled in the study were followed for approximately two years. Bariatric surgery resulted in resolution of co-morbidities such as diabetes (77% of patients), hypertension (62%), hyperlipidemia (70%), and obstructive sleep apnea (86%). They report an overall excess weight loss of 48% for gastric banding and 62% for gastric bypass.

Read more »

FDA Approves Raltegravir- A First in New Class of HIV Medications

November 14, 2007

220px-raltegravir.pngCommentary by Helen Kourlas PharmD, Pharmacology Section Editor

On October 16th the FDA announced the approval of raltegravir (Isentress®) for the treatment of Human Immunodeficiency Virus (HIV)-1 infection in combination with other antiretroviral agents. The use of raltegravir is recommended for patients who have HIV-1 strains resistant to multiple antiretroviral medications. Raltegravir belongs to a new pharmacologic class of antiretrovirals called HIV integrase strand transfer inhibitors. Integrase is one of the three enzymes necessary for the HIV-1 virus to replicate, and integrase inhibitors can stop the HIV-1 virus from inserting its genes into uninfected DNA. The other two enzymes necessary for viral replication are reverse transcriptase and protease which already are targeted by a variety of antiretrovirals.

According to the FDA’s press release, raltegravir was approved based on data from two double-blind, placebo-controlled studies in 699 HIV-1 infected adult patients with histories of extensive antiretroviral use. HIV viral loads dropped below a measurable threshold of 50 copies per milliliter in 65% of the patients taking the drug, nearly five times as many as the placebo group. Common adverse events reported with raltegravir use were diarrhea, nausea, and headache and abnormal elevated levels of muscle enzymes. Caution is advised when using raltegravir in patients at increased risk for certain types of muscle problems. Currently, raltegravir has not been shown to affect disease progression and patients may still be at risk for developing opportunitistic infections.

Reference: FDA Approves New HIV Drug. Raltegravir Tablets Used in Combination with Other Antiretroviral Agents. FDA News. Accessed 10/18/07

Image of Raltegravir, courtesy of Wikipedia