Primecuts – This Week In The Journals

June 25, 2014

By Ian Henderson, MD

Peer Reviewed

This week we lost an outspoken, revolutionary leader in medicine [1], Dr. Arnold Relman. A former Editor in chief of the New England Journal and professor emeritus of medicine and social medicine at Harvard, Dr. Relman passed peacefully in his home in Cambridge, Mass. at the age of 91. Known for being a for being a pioneering researcher early in his career, Dr. Relman went on to become an out spoken editor of numerous medical journals and won the Geogre Polk award in 2002. Dr. Relman became known for his criticism of the health care system and it becoming a “medical-industrial” complex. A pioneer for his time, Dr. Relman said, “The private health care industry is primarily interested in selling services that are profitable, but patients are interested only in services that they need.”[1]

In this week’s Issue of NEJM, Dr. Relman’s former journal, Russel et al. published the first outpatient study on the bionic pancreas[2], a novel insulin/glucagon pump that is linked to a continuous glucose monitoring run through an algorithm on an iPhone 4s. The report is of two random-order, crossover studies comparing the bionic pancreas to a standard insulin pump. The first study, which is more pertinent to the adult internist, focuses on 20 type 1 diabetic adults for a total of 10 days while the other study collected data from 32 type I diabetic adolescents for a total of 10 days. In the adult study the primary outcome was mean plasma glucose after day one, and the bionic pancreas proved to provide more strict glycemic control than a standard insulin pump(133±13 vs. 159±30 mg/dL, P<0.001). It was also noted there was a difference in episodes of hypoglycemia, the percentage of time with a glucose reading of less than 70 mg/dL was lower (4.1% vs.7.3%, P = 0.01) with the bionic pancreas than a standard insulin pump. This study demonstrates that in a controlled, well-monitored outpatient environment, a dual pump insulin and glucagon system provides better glycemic control than a standard insulin pump with less risk of hypoglycemia. While very promising, this study must be viewed in light of its size, a total of only 20 patients, and environment. Patients during the bionic pancreas period were restricted to a 3 square mile area of Boston and accompanied by a staff member at all times. Therefore, the bionic pancreas requires further testing on larger scales in less controlled environments before the system can be hailed as the future of outpatient diabetes management.

Warfarin dosing is a labile process different from most other medications, and is a challenge faced commonly by physicians. In order to combat this issue, genetic tests have been developed with the hope of creating dosing regimens based on a patient’s genotype. In this meta-analysis, Stergiopoulos et al.[3] address the question of how genotype guided dosing of warfarin compares to standard clinical dosing. The analysis is comprised of 9 randomized controlled trials, consisting of 2812 patients, found via MEDLINE, EMBASE, and the Cochrane Library Central Register of Controlled Trials searches. Genotypic variants previously found to influence warfarin dosing requirements and included in the meta-analysis were CYP2C9, VKORC1, and CYP4F2. The authors found a genotype guided strategy was not superior to clinical dosing based on their primary outcome of percentage time spent at a therapeutic INR, with the difference in means of the percentage of time that the INR was within the therapeutic range was 0.14 for the genotype-guided dosing cohort compared to clinical dosing (P = .25, 95%CI: ?0.10 to 0.39, I2=88%). Genotype guided dosing was also found to be not superior on other outcomes such as number of patients with an INR greater than 4, and thromboembolic or major bleeding events. While these findings suggest that time-tested clinical dosing can continue, this study does have some limitations. In terms of the primary outcome the studies used have a large degree of heterogenicity (I2 >75%), weakening the strength of this finding. In addition, genotype guided dosing was only used for determining the patients initial dose of warfarin and future adjustments were made using clinical judgment. This meta-analysis shows that while conceptually promising genotype guided dosing is not clinically useful currently, and physicians are better off spending resources on good clinical dosing than ordering warfarin related genotypes.

Methotrexate, the first line disease modifying treatment for rheumatoid arthritis, comes with a large side effect profile. In the recent years, numerous biologic agents have been developed for use in RA and other autoimmune diseases. In this phase three randomized, double blind, parallel group trial of 986 patients, [4] featured in NEJM this week, Lee, et al. looked at the effectiveness of tofacitinib, an oral Janus kinase (JAK) inhibitor, compared to methotrexate for monotherapy of moderate to severe rheumatoid arthritis. This trial had co-primary outcomes of modified Sharp score, a means of representing structural joint damage (with a greater scores representing more damage) and ACR 70 response. The ACR 70 response is defined as at least a 70% reduction from baseline in the number of tender and swollen joints and improvement in three or more of the following; the patient’s assessment of pain, level of disability, C-reactive protein level or erythrocyte sedimentation rate, global assessment of disease by the patient, and global assessment of disease by the physician. Tofacitinib was demonstrated to be superior to methotrexate both of these co-primary outcomes. The mean proportion of patients who had an ACR 70 response at 6 months was 25.5±2.3% in the 5-mg tofacitinib group and 37.7±2.4% in the 10-mg tofacitinib group, as compared with 12.0±2.4% in the methotrexate group (P<0.001 for either dose vs. methotrexate). The mean changes in the Sharp score from baseline at 6 months were 0.2±0.1 points in the 5-mg tofacitinib group and <0.1±0.1 points in the 10-mg tofacitinib group, as compared with 0.8±0.2 points in the methotrexate group (P<0.001 for both comparisons). This finding was again demonstrated at 12- and 24-month assessments. Despite showing superiority in clinical outcomes, tofacitinib had a more severe side effect profile with more numbers of malignancies detected (6 vs. 0), patients with a raise in serum creatinine, and patients with infections, most commonly herpes zoster. While tofacitinib may offer a better clinical response than methotrexate, it does not come without its risks, and at this point cannot be promoted to first line in place of methotrexate as a result.

Non-invasive ventilation(NIV) has become an excellent tool in respiratory support of patients with acute respiratory failure (ARF). Despite its efficacy, NIV is frequently limited by patient cooperation and comfort. In a randomized, double blind, placebo control trial,[5] Devlin et al. look at the role of the non-respiratory depressant sedative dexemedtomidine as a means of early sedation in patients with ARF receiving NIV. A total of 33 patients were enrolled in the study, 17 were randomized within 8 hours of starting NIV to receive IV dexmedetomidine according to dosing protocol to maintain a Sedation-Agitation Scale score of 3 to 4, while 16 were randomized to receive placebo. Patients were treated for break through discomfort or agitation with IV midazolam or fentanyl as needed and were followed for up to 72 hours, until NIV was stopped for >2 hours, or until intubation. The primary outcome was NIV tolerance graded on the four point Non-invasive ventilation tolerance scale. According to this scale, patients receiving early dexmedetomidine did not better tolerate NIV when compared to those receiving placebo (OR, 1.44; 95% CI, 0.44-4.70; P = .54). Dexemedetomidine did lead to a longer time spent on NIV, but had equal total ventilation time (NIV+invasive) as placebo and lead to a similar number of intubations as placebo. These finding suggest that early sedation with dexmedetomidine is not helpful in improving tolerance of NIV and does not avoid invasive ventilation. Despite this, the study is quite small and therefore underpowered to evaluate the study subgroups. Therefore, there may be a role for IV dexmedetomidine in patients who began NIV with a high level of intolerance, and further study is required.

Other interesting articles of note this week:

A meta-analysis of the current trials on thrombolytics for Pulmonary Embolism[6].

A study on the effects of screening and life style interventions on ischemic heart disease in a Danish population [7].

A meta-analysis of radiotherapy after mastectomy and axillary dissection for breast cancer[8] and its effect on recurrence and mortality.

Dr. Ian Henderson is a 1st year resident at NYU Langone Medical Center

Peer reviewed by Gregory Schrank, contributing editor, Clinical Correlations


1- Martin, D. Dr. Arnold Relman, Outspoken Medical Editor, Dies at 91. html?ref=health&_r=0 Published 6/21/14

2- Russel, et al. Outpatient Glycemic Control with a Bionic Pancreas in Type 1 Diabetes. N Eng J Med 2014. Jun 15.

3- Stergiopoulos, et al. Genotype-Guided vs Clinical Dosing of Warfarin and Its AnaloguesMeta-analysis of Randomized Clinical Trials. JAMA Intern Med. Jun 16.

4- Lee, et al. Tofacitinib versus methotrexate in rheumatoid arthritis. N Engl J Med. 2014 Jun 19;370(25):2377-86

5- Devlin, et al. Efficacy and Safety of Early Dexmedetomidine during Non-Invasive Ventilation for Patients with Acute Respiratory Failure: A Randomized, Double-Blind, Placebo-Controlled, Pilot Study. Chest. 2014 Feb 27. http://

6- Chatterjee, et al. Thrombolysis for Pulmonary Embolism and Risk of All-Cause Mortality, Major Bleeding, and Intracranial HemorrhageA Meta-analysis. JAMA. 2014 Jun 18;311(23):2414-21.

7- Jørgensen, et al. Effect of screening and lifestyle counselling on incidence of ischaemic heart disease in general population: Inter99 randomised trial. BMJ. 2014 Jun 9;348.

8- Godwin, et al. Effect of radiotherapy after mastectomy and axillary surgery on 10-year recurrence and 20-year breast cancer mortality: meta-analysis of individual patient data for 8135 women in 22 randomised trials. The Lancet, Volume 383, Issue 9935, Pages 2127 – 2135, 21 June 2014.