Internal Medicine Residency News: December 2, 2013

Sunday, December 1, 2013

From the Director

DUKE.RESEARCH.NIGHT.03 (1)Happy thanksgiving everyone! I am certainly thankful for being part of the best program in the country. Turkey bowl was spectacular, with the Marines posting a victory, followed by everyone returning triumphantly to Duke to share in a thanksgiving lunch. Again, thanks to our coaching staff of Ryan Schulties and Tony G. for letting Jonah and Jake into the game! It was especially great to hear of all the "friendsgiving" dinners you all had. I've been telling our applicants that they have no idea how important a new group of people will become to them until they start residency - proof of that is in how many of you spent the holiday together.  We also raised nearly $1000 to support the families at Hope Valley Elementary who were in need this thanksgiving.

Kudos this week to Aaron Loochtan for stepping up to fill in a schedule hole on CAD, to all the attendings who let their residents get to Turkey Bowl, and to Marc Samsky, Brian Miller, Rachel Titerance and Hany Elmariah for resident share with the applicants. Also thanks to our ACRs Meredith Clement, Jim Gentry and Chris Hostler for planning the pre turkey bowl MKSAP jeopardy noon conference.

It's a busy week this week - no doubt people will be gathering to watch Duke vs Michigan on Tuesday. On Wdnesday we have the first annual Kussin vs Klotman vs Zaas latke cookoff in the med res library at 4:30, followed by the fellowship match party.  we have SAR talks on Thursday as well, and then Duke football watching on Saturday.  Don't forget to also sign up for Voices in Medicine (see Vaishali for details), and be watching your email for 2014-15(!) schedule request forms.  Yep, time is flying by.

This week' pubmed from the program goes to Mandar Aras for his AHA presentation:  Peripheral metabolite profiles predict cardiomyopathy in a cohort of cardiac catheterization patients. Poster presentation at the American Heart Association annual meeting in Dallas, Texas in November 2013;  mentor: Svati Shah, MD

Have a great week!


QI Corner

First off, Happy Thanksgiving!turkey

QI Noon Conference
Thanks also to Dr. Dani Zipkin for her recent presentation on High-Value Biostatistical Concepts by Dr. Dani Zipkin as part of the HVCC lecture series.

QI Abstracts and Meetings
The submission date is 12/9 for the upcoming Patient Safety and Quality Conference here at Duke in March.  Also, the submission date is 12/14 for the NC ACP conference at the end of February.  If you need us to help or review anything you're planning to submit, please send us an email!

Duke Patient Safety and Quality Conference site:

NC ACP site:

QI Champ
Congrats to Wassim Shatila, our QI Champ for November.  We now need only one more Champion for the We Follow-Up project, primarily based at the DOC. Contact Jon or Joel if interested . .



What Did I Read This Week?

By Vaishali Patel 

“Proton Pump Inhibitors and Risk for Recurrent Clostridium difficile Infection Among Inpatients?”

Freedberg DE et al.  Am J Gastroenterol 2013; 108:1794–1801


 I know what you’re thinking.  Yes, Vaishali picked a GI article.  As biased as I may be, give me a chance to prove to you why delving into the controversy surrounding the recurrent infection risk associated with proton-pump inhibitors (PPIs) in inpatients may worth your time (and money!).  There has been conflicting data exploring the association between PPIs and several infections, including pneumonia, spontaneous bacterial peritonitis and clostridium difficile infections (CDI).  One of the challenges of treating CDIs is the high rate of recurrence after treatment. The risk of recurrence is the highest between 7-30 days after finishing antibiotic treatment, but actually persists for up to 90 days.  This combined with an increasing rate of community-acquired infection has resulted in a continuously rising incidence of CDI.  This makes it imperative to identify risk factors for recurrence, so that they can be targeted by preventive measures.  Previous studies have suggested that PPIs are a risk factor for about new CDIs (systematic reviews and meta-analyses, Janarthanan S et al, Am J Gastroenterol 2012; Kwok CS et al, Am J Gastroenterol 2012; Tleyjeh IM et al, PLoS One 2012).  Similar to the data regarding PPI use and the risk of pneumonia, the evidence suggesting the risk of CDI associated with PPI use has sparked controversy and is conflicted.  There is evidence to suggest that PPI use is associated with recurrent CDI, though some of this data comes from smaller observational studies (one was a larger study done across eight VA medical centers in outpatients and inpatients, Linsky et al, Arch Intern Med 2010).  Furthermore, there is very little data focusing on the population most at risk for recurrent CDI – hospitalized patients!This retrospective cohort analysis by Freedberg et al, was conducted in hospitalized adult patients. Patients with new CDI (defined as having a first positive C. difficile stool toxin B PCR test and having received treatment) who had undergone CDI treatment were identified by electronic records.  After excluding patients who were lost to follow-up by their hospital system within 90 days, 894 patient records were reviewed for in-hospital administration of PPIs (minimum of 2 days exposure) as well as other acid suppressives, non-CDI treatment antibiotics and other comorbidities.  They looked for a second positive stool test within 15-90 days of the first to define recurrence (seems reasonable since most patients should respond to CDI therapy within ~3 days and over 90% should be cured after 14 days of treatment, after which is it okay to test again for recurrent symptoms).  Recurrence rates were compared between patients unexposed and exposed to in-hospital PPI therapy either during or shortly after CDI treatment, and analyses were controlled for potential confounders including age, sex, race, the modified Charlson score which signifies comorbidities, type of Cdiff treatment, ICU stay, hospital length of stay, receipt of antibiotics and inpatient immunosuppressant exposure.

The cumulative incidence of recurrent CDI in this cohort was 23% - this is close to most estimates of recurrent CDI of 25%-30% in patients who have been treated with either metronidazole or vancomycin.  The hazard ratio for recurrent CDI associated with concurrent PPI treatment was HR =0.82 (95% CI = 0.58-1.16).  In the subset of patients who survived 90 days of follow-up, there was again no association between PPIs and CDI recurrence (HR=0.87, 95% CI = 0.60-1.28).  Increased duration or dose of PPIs also did not make a difference.  Factors that actually were associated with increased risk for recurrent CDI included increased age, black race, and increased comorbidities.  Older age has been previously identified as a risk factor in both inpatients and outpatients.  This is the first study to find an association between black race and increased risk of recurrent CDI.

Previous observational studies suggesting that PPI therapy is a risk factor for new CDI is relatively convincing, but it remains unclear whether there is a causal relationship or if there are underlying confounding factors.  Perhaps the greatest strength of this study is that it was focused on hospitalized patients, which somewhat helps to minimize heterogeneity between patients.  Furthermore, 41% of their patients were treated with both metronidazole and vancomycin compared to 10% in other studies, and they had a higher proportion of these patients in the PPI group compared to the non-PPI group.  This may suggest that this study had a higher portion of patients with more severe incident CDI, and would have a higher rate of refractory or recurrent CDI.  If anything, this should have moved the results away from the null hypothesis of finding no association between PPI therapy and recurrent CDI.  This was a much larger study than those that have been done previously.  In spite of this, however, the paper did not include information about their power analysis.  Clinically, you may actually care about detecting a small difference in CDI recurrence rate (especially in older, sicker patients) and this study may not have been adequately powered to detect it.  It also would have been important to control for patients being discharged on PPI therapy, non-CDI antibiotic therapy, or immunosuppression.  Lastly, when you are interpreting the results of retrospective chart review studies, it is important to remember that the results are only as good as the data that was available for analysis: though it was good that patients who did not have good follow-up were excluded, it would have been helpful to know if the investigators obtained outside medical records to find out if patients were diagnosed with recurrence elsewhere.  The recurrence rate found in their study is similar to previous estimates, so that is reassuring.

What did I take home from all of this?  Well, I think carefully about the indication for PPI therapy in my elderly patient with several comorbidities who may be at risk for CDI infection.  If my patient has already had a recent episode of CDI, they have about a ~30% risk of recurrence, especially if they are elderly and have several comorbidities – and giving them in-hospital PPI therapy will not increase their risk for recurrent CDI…but I should think carefully about the indication for giving them PPI therapy anyway!  As any other medication, PPI use obviously has unanticipated consequences.  Pathophysiologically, this sort of makes sense for new CDI – altering the pH of the stomach and its contents may affect gastrointestinal flora that play a role in the immune barrier mechanisms of GI mucosa.  If they have had a recent CDI, the GI microbiome is already altered, so perhaps a PPI does not have much effect.  PPIs are highly effective for the treatment of pathology associated with increased gastric-acid but also are overused and often prescribed when not indicated.  As with any other medication, it is important to use PPIs for the correct indication and with the minimum dose and duration needed for effective therapy.


From the Chief Residents

Turkey Bowl 2013 RevisitedGroup Shot TB

Great game, no major injuries, and another turkey bowl goes down into the record books.  TB Action 1The trophy is now residing in the office of the Duke Chief Resident after making its way across the street.  There were a few surprise plays - and players - as Jona and Jake BOTH came in off the bench.  The collection of pictures will tell the story for many years to come.

Hats off to the Duke Marines !score

Sar Talks

     Thursday, December 5, 2013

     Meredith Clement and Aaron Mitchell


Grand Rounds

Dr. David Holland HIV Prevention in 2013

Noon Conference

Date Topic Lecturer Time Vendor Room
12/2 FUO Evaluation Susanna Naggie 12:00 Rudinos 2002
12/4 MSK Exam Part 1 Irene Whitt & Lisa Criscione 12:00 China King 2002
12/5 SAR talks Meredith Clement /    Aaron Mitchell 12:00 Chick-Fil-A 2001


From the Residency Office

Wanted: Future leaders in ambulatory care

Have you thought about how your training provides the kinds of knowledge and skills you'll need in your career?  For many medicine residents, a background in leadership, clinical teaching, advanced EBM, communication, and strong ambulatory clinical skills will best support their career.  This is particularly true for our residents wanting to pursue careers in academic medicine or perhaps as physician leaders -- which led me to create the Ambulatory Care Leadership Track (ACLT) a few years ago, with the support of Aimee and Randy and the MedRes Office.

Four JAR spots and two SAR spots are now open for the 2014-15 academic year.  I encourage you to consider applying, and talk to residents in the program or ambulatory faculty to see if the ACLT is the right choice for you.  The track was designed not only for residents interested in primary care, but also for those of you who are interested in ambulatory subspecialty careers.  We can promise you social events and camaraderie with like-minded residents and faculty, too, organized by Sharon Rubin and others.

If interested or if you have questions please contact current ACLT leaders Alex Cho, Stephen Bergin, Daniella Zipkin, or me.  A brief, one-page application will be due Monday, December 30. You can also go to for more information.  Thanks!

Alex Cho  MD, MBA

ACLT application form

Noon Conference Lunch Options

Lynsey is picking up the ball from Lauren and has continued to explore options for the lunch menu.  This week we are giving China King a chance on Wednesday, and the following week she has arranged for a baked potato bar.  It is quite a challenge to find alternatives that work, so please share your feedback.

One request please:  Both of these options do bring an increased risk of creating a "mess", so if you would - help clean up if something drops/dribbles/spills.

Faces of Flu Prevention 2013

Throughout Duke, our team members are stepping up to protect our patients, their loved ones and each other by getting the flu vaccine. Follow the link below to check out this year’s Faces of Flu Prevention poster, read some of the stories behind the vaccinations, and find out how to
share your story.

Stephanie L. Giattino, MD
PGY-1, Internal Medicine
Duke University Hospital

Duke Hospital Resident Stephanie Giattino, MD, came in on her day off to get a flu shot. She shared, “It just makes sense to get the flu shot. How many other health care interventions can you think of that take only a few minutes and then last for a full year? It is fast and easy, and by getting a flu shot, you can help protect your own health as well as that of your family, friends and patients!”




Upcoming Dates and Events

  • December 4:  Fellowship Match Party
  • December 14:  DoM Holiday Party
  • January 15th:  "Voices in Medicine"


Useful links