Week 43 – FREEDOM

“Strategies for Multivessel Revascularization in Patients with Diabetes”

by the FREEDOM (Future Revascularization Evaluation in Patients with Diabetes Mellitus: Optimal Management of Multivessel Disease) Trial investigators

N Engl J Med. 2012 Dec 20;367(25):2375-84. [free full text]

Previous studies, such as the 1996 BARI trial), have demonstrated that patients who have multivessel coronary artery disease (CAD) and diabetes mellitus (DM) and who received coronary artery bypass grafting (CABG) surgery lived longer than patients undergoing balloon angioplasty. However, since that publication, percutaneous coronary intervention (PCI) technology advanced significantly. Prior to the publication of FREEDOM in 2012, there had only been small, underpowered studies comparing PCI with drug-eluting stent (DES) to CABG. FREEDOM was powered appropriately to discover superiority of revascularization strategy (PCI with DES vs. CABG) in patients with DM and multivessel CAD.


Inclusion criteria:

      • 18 years or older
      • Diabetes mellitus – defined by American Diabetes Association
      • Multivessel Coronary Artery Disease
        • > 70% stenosis (angiographically confirmed)
        • 2 or more epicardial vessels
        • 2 or more coronary-artery territories

Selected exclusion criteria:

      • NYHA Class III-IV heart failure
      • Prior CABG, valve surgery, or PCI (< 6 months)
      • Prior significant bleed (< 6 months)
      • Left main stenosis ≥ 50%


Patients meeting criteria were assigned 1:1 into PCI with first-generation paclitaxel-eluting stent (51%) or sirolimus-eluting stent (43%) versus CABG. The PCI group was placed on aspirin and clopidogrel for dual antiplatelet therapy (DAPT) for at least 12 months. For the CABG group, arterial revascularization was encouraged. The mean SYNTAX score (tool used to score complexity of CAD) was 26.2 and did not significantly differ between groups. Guideline-driven targets for lowering medical risk factors were used: LDL <70, BP <130/80, HgbA1c <7. Minimum follow-up was 2 years.


Primary: Composite of death from any cause, non-fatal myocardial infarction (MI), and non-fatal stroke


      1. Rate of major adverse cardiovascular and cerebrovascular events at 30 days and 12 months
      2. Repeat revascularization
      3. Annual all-cause mortality
      4. Annual cardiovascular mortality

953 patients and 947 patients were randomized into the PCI and CABG groups, respectively. At 5 years, the primary outcome (combined death, MI, or stroke) occurred in 200 of the PCI group and 146 of the CABG group (26.6% vs 18.7%, p = 0.005). The curves started diverging at 2 years. All-cause mortality was higher in the PCI group versus the CABG group (16.3% vs 10.9%, p = 0.049). Regarding secondary outcomes, 13.9% of patients in the PCI group had a repeat MI versus 6.0% in the CABG group (p < 0.001). There were fewer strokes in the PCI group than in the CABG group (2.4% vs 5.2%, p = 0.03). There was no statistically significant difference between study groups regarding cardiovascular death (10.9% vs 6.8%, p = 0.12).

At 5 years, the analysis of outcomes according to category of SYNTAX score (≤ 22, 23 to 32, ≥ 33) showed no significant subgroup interaction (p = 0.58).

Regarding safety, major bleeding between the two groups at 30 days was 0.02% for PCI vs 0.04% for CABG (p = 0.13). The incidence of acute renal failure requiring hemodialysis was observed in one patient in the PCI group and eight patients in the CABG group (p = 0.02)

The BARI Trial (1996) was the first trial to show that patients with DM and multivessel CAD derive mortality benefit from bypass grafting over PCI with balloon angioplasty. Furthermore, the BARI 2D (2009) trial demonstrated this benefit of bypass grafting over PCI with bare metal stents (BMS). At the time of the FREEDOM Trial, there had not been a randomized comparison of CABG versus PCI with newer technology and first-generation paclitaxel/sirolimus DES. In this study, CABG showed a 5.3% absolute reduction in all-cause mortality over PCI as well decreased rates of MI and repeat revascularization. CABG was associated with a mild absolute increase in stroke (2.8%). However, this mild increased stroke risk is consistent with most other comparative trials of the two treatment strategies. There was no statistical difference in major bleeding between the two groups.

CABG is likely better than PCI for various reasons. For one, diabetic arteries are affected diffusely and tend to have more extensive atherosclerotic disease compared to those without diabetes, so the likelihood of successful PCI alone is low. Many suspected that with advancement in PCI (i.e. DES) that the BARI data would become irrelevant. However, CABG continued to show benefit despite the technological advancements of drug-eluting stents and PCI. Improvement in surgical technique as well as the use of arterial revascularization (i.e. internal mammary artery) helped maintain superior outcomes with CABG compared to PCI.

The study was limited by the fact that due to low numbers, the subgroup analysis (i.e. SYNTAX scores) was not appropriately powered for statistical significance. Further, the study was not blinded, and patients may have been treated differently on the basis of their surgical procedure. Also, there was variability of STYNAX scores between the study groups, but this circumstance was thought to reflect real world heterogeneity.

Bottom Line:
CABG was superior to PCI with DES in patients with DM and multivessel CAD in that it significantly reduced rates of death and MI despite a small increased risk of stroke.

Further Reading/References:
1. BARI Trial @ NEJM
2. BARI 2D Trial @ NEJM
3. ACCF/AHA 2011 Guideline for Coronary Artery Bypass Graft Surgery
4. FREEDOM @ Wiki Journal Club
5. FREEDOM @ 2 Minute Medicine
5. FREEDOM @ Visualmed

Summary by Patrick Miller, MD.

Image Credit: Jerry Hecht, US Public Domain, via Wikimedia Commons

Week 41 – Transfusion Strategies for Upper GI Bleeding

“Transfusion Strategies for Acute Upper Gastrointestinal Bleeding”

N Engl J Med. 2013 Jan 3;368(1):11-21. [free full text]

A restrictive transfusion strategy of 7 gm/dL was established following the previously discussed 1999 TRICC trial. Notably, both TRICC and its derivative study TRISS excluded patients who had an active bleed. In 2013, Villanueva et al. performed a study to establish whether there was benefit to a restrictive transfusion strategy in patients with acute upper GI bleeding.

The study enrolled consecutive adults presenting to a single center in Spain with hematemesis (or bloody nasogastric aspirate), melena, or both. Notable exclusion criteria included: a clinical Rockall score* of 0 with a hemoglobin level higher than 12g/dL, massive exsanguinating bleeding, lower GIB, patient refusal of blood transfusion, ACS, stroke/TIA, transfusion within 90 days, recent trauma or surgery

*The Rockall score is a system to assess risk for further bleeding or death on a scale from 0-11. Higher scores (3-11) indicate higher risk. Of the 648 patients excluded, the most common reason for exclusion (n = 329) was low risk of bleeding.

Intervention: restrictive transfusion strategy (transfusion threshold Hgb = 7.0 gm/dL) [n = 444]

Comparison: liberal transfusion strategy (transfusion threshold Hgb = 9.0 gm/dL) [n = 445]

During randomization, patients were stratified by presence or absence of cirrhosis.

As part of the study design, all patients underwent emergent EGD within 6 hours and received relevant hemostatic intervention depending on the cause of bleeding.


Primary outcome: 45-day mortality

Secondary outcomes, selected:

      • Incidence of further bleeding associated with hemodynamic instability or hemoglobin drop > 2 gm/dL in 6 hours
      • Incidence and number of RBC transfusions
      • Other products and fluids transfused
      • Hgb level at nadir, discharge, and 45 days

Subgroup analyses: Patients were stratified by presence of cirrhosis and corresponding Child-Pugh class, variceal bleeding, and peptic ulcer bleeding. An additional subgroup analysis was performed to evaluate changes in hepatic venous pressure gradient between the two strategies.

The primary outcome of 45-day mortality was lower in the restrictive strategy (5% vs. 9%; HR 0.55, 95% CI 0.33-0.92; p = 0.02; NNT = 24.8). In subgroup analysis, this finding remained consistent for patients who had Child-Pugh class A or B but was not statistically significant among patients who had Class C. Further stratification for variceal bleeding and peptic ulcer disease did not make a difference in mortality.

Secondary outcomes:
Rates of further bleeding events and RBC transfusion, as well as number of products transfused, were lower in the restrictive strategy. Subgroup analysis demonstrated that rates of re-bleeding were lower in Child-Pugh class A and B but not in C. As expected, the restrictive strategy also resulted in the lowest hemoglobin levels at 24 hours. Hemoglobin levels among patients in the restrictive strategy were lower at discharge but were not significantly different from the liberal strategy at 45 days. There was no group difference in amount of non-RBC blood products or colloid/crystalloid transfused. Patients in the restrictive strategy experienced fewer adverse events, particularly transfusion reactions such as transfusion-associated circulatory overload and cardiac complications. Patients in the liberal-transfusion group had significant post-transfusion increases in mean hepatic venous pressure gradient following transfusion. Such increases were not seen in the restrictive-strategy patients.

In patients with acute upper GI bleeds, a restrictive strategy with a transfusion threshold 7 gm/dL reduces 45-day mortality, the rate and frequency of transfusions, and the rate of adverse reactions, relative to a liberal strategy with a transfusion threshold of 9 gm/dL.

In their discussion, the authors hypothesize that the “harmful effects of transfusion may be related to an impairment of hemostasis. Transfusion may counteract the splanchnic vasoconstrictive response caused by hypovolemia, inducing an increase in splanchnic blood flow and pressure that may impair the formation of clots. Transfusion may also induce abnormalities in coagulation properties.”

Subgroup analysis suggests that the benefit of the restrictive strategy is less pronounced in patients with more severe hepatic dysfunction. These findings align with prior studies in transfusion thresholds for critically ill patients. However, the authors note that the results conflict with studies in other clinical circumstances, specifically in the pediatric ICU and in hip surgery for high-risk patients.

There are several limitations to this study. First, its exclusion criteria limit its generalizability. Excluding patients with massive exsanguination is understandable given lack of clinical equipoise; however, this choice allows too much discretion with respect to the definition of a massive bleed. (Note that those excluded due to exsanguination comprised only 39 of 648.) Lack of blinding was a second limitation. Potential bias was mitigated by well-defined transfusion protocols. Additionally, there a higher incidence of transfusion-protocol violations in the restrictive group, which probably biased results toward the null. Overall, deviations from the protocol occurred in fewer than 10% of cases.

Further Reading/References:
1. Transfusion Strategies for Acute Upper GI Bleeding @ Wiki Journal Club
2. Transfusion Strategies for Acute Upper GI Bleeding @ 2 Minute Medicine
3. TRISS @ Wiki Journal Club

Summary by Gordon Pelegrin, MD

Image Credit: Jeremias, CC BY-SA 3.0, via Wikimedia Commons

Week 39 – POISE

“Effects of extended-release metoprolol succinate in patients undergoing non-cardiac surgery: a randomised controlled trial”

Lancet. 2008 May 31;371(9627):1839-47. [free full text]

Non-cardiac surgery is commonly associated with major cardiovascular complications. It has been hypothesized that perioperative beta blockade would reduce such events by attenuating the effects of the intraoperative increases in catecholamine levels. Prior to the 2008 POISE trial, small- and moderate-sized trials had revealed inconsistent results, alternately demonstrating benefit and non-benefit with perioperative beta blockade. The POISE trial was a large RCT designed to assess the benefit of extended-release metoprolol succinate (vs. placebo) in reducing major cardiovascular events in patients of elevated cardiovascular risk.

The trial enrolled patients age 45+ undergoing non-cardiac surgery with estimated LOS 24+ hrs and elevated risk of cardiac disease, meaning: either 1) hx of CAD, 2) peripheral vascular disease, 3) hospitalization for CHF within past 3 years, 4) undergoing major vascular surgery, 5) or any three of the following seven risk criteria: undergoing intrathoracic or intraperitoneal surgery, hx CHF, hx TIA, hx DM, Cr > 2.0, age 70+, or undergoing urgent/emergent surgery.

Notable exclusion criteria: HR < 50, 2nd or 3rd degree heart block, asthma, already on beta blocker, prior intolerance of beta blocker, hx CABG within 5 years and no cardiac ischemia since

Intervention: metoprolol succinate (extended-release) 100mg PO starting 2-4 hrs before surgery, additional 100mg at 6-12 hrs postoperatively, followed by 200mg daily for 30 days. (Patients unable to take PO meds postoperatively were given metoprolol infusion.)

Comparison: placebo PO / IV at same frequency as metoprolol arm

Primary – composite of cardiovascular death, non-fatal MI, and non-fatal cardiac arrest at 30 days

Secondary (at 30 days)

        • cardiovascular death
        • non-fatal MI
        • non-fatal cardiac arrest
        • all-cause mortality
        • non-cardiovascular death
        • MI
        • cardiac revascularization
        • stroke
        • non-fatal stroke
        • CHF
        • new, clinically significant atrial fibrillation
        • clinically significant hypotension
        • clinically significant bradycardia

Pre-specified subgroup analyses of primary outcome:

9298 patients were randomized. However, fraudulent activity was detected at participating sites in Iran and Colombia, and thus 947 patients from these sites were excluded from the final analyses. Ultimately, 4174 were randomized to the metoprolol group, and 4177 were randomized to the placebo group. There were no significant differences in baseline characteristics, pre-operative cardiac medications, surgery type, or anesthesia type between the two groups (see Table 1).

Regarding the primary outcome, metoprolol patients were less likely than placebo patients to experience the primary composite endpoint of cardiovascular death, non-fatal MI, and non-fatal cardiac arrest (HR 0.84, 95% CI 0.70-0.99, p = 0.0399). See Figure 2A for the relevant Kaplan-Meier curve. Note that the curves separate distinctly within the first several days.

Regarding selected secondary outcomes (see Table 3 for full list), metoprolol patients were more likely to die from any cause (HR 1.33, 95% CI 1.03-1.74, p = 0.0317). See Figure 2D for the Kaplan-Meier curve for all-cause mortality. Note that the curves start to separate around day 10. Cause of death was analyzed, and the only group difference in attributable cause was an increased number of deaths due to sepsis or infection in the metoprolol group (data not shown). Metoprolol patients were more likely to sustain a stroke (HR 2.17, 95% CI 1.26-3.74, p = 0.0053) or a non-fatal stroke (HR 1.94, 95% CI 1.01-3.69, p = 0.0450). Of all patients who sustained a non-fatal stroke, only 15-20% made a full recovery. Metoprolol patients were less likely to sustain new-onset atrial fibrillation (HR 0.76, 95% CI 0.58-0.99, p = 0.0435) and less likely to sustain a non-fatal MI (HR 0.70, 95% CI 0.57-0.86, p = 0.0008). There were no group differences in risk of cardiovascular death or non-fatal cardiac arrest. Metoprolol patients were more likely to sustain clinically significant hypotension (HR 1.55, 95% CI 1.38-1.74, P < 0.0001) and clinically significant bradycardia (HR 2.74, 95% CI 2.19-3.43, p < 0.0001).

Subgroup analysis did not reveal any significant interaction with the primary outcome by RCRI, sex, type of surgery, or anesthesia type.

In patients with cardiovascular risk factors undergoing non-cardiac surgery, the perioperative initiation of beta blockade decreased the composite risk of cardiovascular death, non-fatal MI, and non-fatal cardiac arrest and increased the overall mortality risk and risk of stroke.

This study affirms its central hypothesis – that blunting the catecholamine surge of surgery is beneficial from a cardiac standpoint. (Most patients in this study had an RCRI of 1 or 2.) However, the attendant increase in all-cause mortality is dramatic. The increased mortality is thought to result from delayed recognition of sepsis due to masking of tachycardia. Beta blockade may also limit the physiologic hemodynamic response necessary to successfully fight a serious infection. In retrospective analyses mentioned in the discussion, the investigators state that they cannot fully explain the increased risk of stroke in the metoprolol group. However, hypotension attributable to beta blockade explains about half of the increased number of strokes.

Overall, the authors conclude that “patients are unlikely to accept the risks associated with perioperative extended-release metoprolol.”

A major limitation of this study is the fact that 10% of enrolled patients were discarded in analysis due to fraudulent activity at selected investigation sites. In terms of generalizability, it is important to remember that POISE excluded patients who were already on beta blockers.

Currently, per expert opinion at UpToDate, it is not recommended to initiate beta blockers preoperatively in order improve perioperative outcomes. POISE is an important piece of evidence underpinning the 2014 ACC/AHA Guideline on Perioperative Cardiovascular Evaluation and Management of Patients Undergoing Noncardiac Surgery, which includes the following recommendations regarding beta blockers:

      • Beta blocker therapy should not be started on the day of surgery (Class III – Harm, Level B)
      • Continue beta blockers in patients who are on beta blockers chronically (Class I, Level B)
      • In patients with intermediate- or high-risk preoperative tests, it may be reasonable to begin beta blockers
      • In patients with ≥ 3 RCRI risk factors, it may be reasonable to begin beta blockers before surgery
      • Initiating beta blockers in the perioperative setting as an approach to reduce perioperative risk is of uncertain benefit in those with a long-term indication but no other RCRI risk factors
      • It may be reasonable to begin perioperative beta blockers long enough in advance to assess safety and tolerability, preferably > 1 day before surgery

Further Reading/References:
1. Wiki Journal Club
2. 2 Minute Medicine
3. UpToDate, “Management of cardiac risk for noncardiac surgery”
4. 2014 ACC/AHA guideline on perioperative cardiovascular evaluation and management of patients undergoing noncardiac surgery: a report of the American College of Cardiology/American Heart Association Task Force on practice guidelines.

Image Credit: Mark Oniffrey, CC BY-SA 4.0, via Wikimedia Commons

Summary by Duncan F. Moore, MD

Week 35 – CORTICUS

“Hydrocortisone Therapy for Patients with Septic Shock”

N Engl J Med. 2008 Jan 10;358(2):111-24. [free full text]

Steroid therapy in septic shock has been a hotly debated topic since the 1980s. The Annane trial in 2002 suggested that there was a mortality benefit to early steroid therapy and so for almost a decade this became standard of care. In 2008, the CORTICUS trial was performed suggesting otherwise.

The trial enrolled ICU patients with septic shock onset with past 72 hrs (defined as SBP < 90 despite fluids or need for vasopressors and hypoperfusion or organ dysfunction from sepsis). Excluded patients included those with an “underlying disease with a poor prognosis,” life expectancy < 24hrs, immunosuppression, and recent corticosteroid use. Patients were randomized to hydrocortisone 50mg IV q6h x5 days plus taper or to placebo injections q6h x5 days plus taper. The primary outcome was 28-day mortality among patients who did not have a response to ACTH stim test (cortisol rise < 9mcg/dL). Secondary outcomes included 28-day mortality in patients who had a positive response to ACTH stim test, 28-day mortality in all patients, reversal of shock (defined as SBP ≥ 90 for at least 24hrs without vasopressors) in all patients and time to reversal of shock in all patients.

In ACTH non-responders (n = 233), intervention vs. control 28-day mortality was 39.2% vs. 36.1%, respectively (p = 0.69). In ACTH responders (n = 254), intervention vs. control 28-day mortality was 28.8% vs. 28.7% respectively (p = 1.00). Reversal of was shock 84.7%% vs. 76.5% (p = 0.13). Among all patients, intervention vs. control 28-day mortality was 34.3% vs. 31.5% (p = 0.51) and reversal of shock 79.7% vs. 74.2% (p = 0.18). The duration of time to reversal of shock was significantly shorter among patients receiving hydrocortisone (per Kaplan-Meier analysis, p<0.001; see Figure 2) with median time to of reversal 3.3 days vs. 5.8 days (95% CI 5.2 – 6.9).

In conclusion, the CORTICUS trial demonstrated no mortality benefit of steroid therapy in septic shock regardless of a patient’s response to ACTH. Despite the lack of mortality benefit, it demonstrated an earlier resolution of shock with steroids. This lack of mortality benefit sharply contrasted with the previous Annane 2002 study. Several reasons have been posited for this difference including poor powering of the CORTICUS study (which did not reach the desired n = 800), inclusion starting within 72 hrs of septic shock vs. Annane starting within 8 hrs, and the overall sicker nature of Annane patients (who were all mechanically ventilated). Subsequent meta-analyses disagree about the mortality benefit of steroids, but meta-regression analyses suggest benefit among the sickest patients. All studies agree about the improvement in shock reversal. The 2016 Surviving Sepsis Campaign guidelines recommend IV hydrocortisone in septic shock in patients who continue to be hemodynamically unstable despite adequate fluid resuscitation and vasopressor therapy.

Per Drs. Sonti and Vinayak of the GUH MICU (excepted from their excellent Georgetown Critical Care Top 40): “Practically, we use steroids when reaching for a second pressor or if there is multiorgan system dysfunction. Our liver patients may have deficient cortisol production due to inadequate precursor lipid production; use of corticosteroids in these patients represents physiologic replacement rather than adjunct supplement.”

The ANZICS collaborative group published the ADRENAL trial in NEJM in 2018 – which demonstrated that “among patients with septic shock undergoing mechanical ventilation, a continuous infusion of hydrocortisone did not result in lower 90-day mortality than placebo.” The authors did note “a more rapid resolution of shock and a lower incidence of blood transfusion” among patients receiving hydrocortisone. The folks at EmCrit argued [https://emcrit.org/emnerd/cc-nerd-case-relative-insufficiency/] that this was essentially a negative study, and thus in the existing context of CORTICUS, the results of the ADRENAL trial do not change our management of refractory septic shock.

Finally, the 2018 APPROCCHSS trial (also by Annane) evaluated the survival benefit hydrocortisone plus fludocortisone vs. placebo in patients with septic shock and found that this intervention reduced 90-day all-cause mortality. At this time, it is difficult truly discern the added information of this trial given its timeframe, sample size, and severity of underlying illness. See the excellent discussion in the following links: WikiJournal Club, PulmCrit, PulmCCM, and UpToDate.

References / Additional Reading:
1. CORTICUS @ Wiki Journal Club
2. CORTICUS @ Minute Medicine
3. Surviving Sepsis Campaign: International Guidelines for Management of Sepsis and Septic Shock (2016), section “Corticosteroids”
4. Annane trial (2002) full text
5. PulmCCM, “Corticosteroids do help in sepsis: ADRENAL trial”
6. UpToDate, “Glucocorticoid therapy in septic shock”

Post by Gordon Pelegrin, MD

Image Credit: LHcheM, CC BY-SA 3.0, via Wikimedia Commons

Week 31 – Early TIPS in Cirrhosis with Variceal Bleeding

“Early Use of TIPS in Patients with Cirrhosis and Variceal Bleeding”

N Engl J Med. 2010 Jun 24;362(25):2370-9. [free full text]

Variceal bleeding is a major cause of morbidity and mortality in decompensated cirrhosis. The standard of care for an acute variceal bleed includes a combination of vasoactive drugs, prophylactic antibiotics, and endoscopic techniques (e.g. banding). Transjugular intrahepatic portosystemic shunt (TIPS) can be used to treat refractory bleeding. This 2010 trial sought to determine the utility of early TIPS during the initial bleed in high-risk patients when compared to standard therapy.

The trial enrolled cirrhotic patients (Child-Pugh class B or C with score ≤ 13) with acute esophageal variceal bleeding. All patients received endoscopic band ligation (EBL) or endoscopic injection sclerotherapy (EIS) at the time of diagnostic endoscopy. All patients also received vasoactive drugs (terlipressin, somatostatin, or octreotide). Patients were randomized to either TIPS performed within 72 hours after diagnostic endoscopy or to “standard therapy” by 1) treatment with vasoactive drugs with transition to nonselective beta blocker when patients were free of bleeding followed by 2) addition of isosorbide mononitrate to maximum tolerated dose, and 3) a second session of EBL at 7-14 days after the initial session (repeated q10-14 days until variceal eradication was achieved). The primary outcome was a composite of failure to control acute bleeding or failure to prevent “clinically significant” variceal bleeding (requiring hospital admission or transfusion) at 1 year after enrollment. Selected secondary outcomes included 1-year mortality, development of hepatic encephalopathy (HE), ICU days, and hospital LOS.

359 patients were screened for inclusion, but ultimately only 63 were randomized. Baseline characteristics were similar among the two groups except that the early TIPS group had a higher rate of patients with previous hepatic encephalopathy. The primary composite endpoint of failure to control acute bleeding or rebleeding within 1 year occurred in 14 of 31 (45%) patients in the pharmacotherapy-EBL group and in only 1 of 32 (3%) of the early TIPS group (p = 0.001). The 1-year actuarial probability of remaining free of the primary outcome was 97% in the early TIPS group vs. 50% in the pharmacotherapy-EBL group (ARR 47 percentage points, 95% CI 25-69 percentage points, NNT 2.1). Regarding mortality, at one year, 12 of 31 (39%) patients in the pharmacotherapy-EBL group had died, while only 4 of 32 (13%) in the early TIPS group had died (p = 0.001, NNT = 4.0). There were no group differences in prevalence of HE at one year (28% in the early TIPS group vs. 40% in the pharmacotherapy-EBL group, p = 0.13). Additionally, there were no group differences in 1-year actuarial probability of new or worsening ascites. There were also no differences in length of ICU stay or hospitalization duration.

Early TIPS in acute esophageal variceal bleeding, when compared to standard pharmacotherapy and endoscopic band ligation, improved control of index bleeding, reduced recurrent variceal bleeding at 1 year, and reduced all-cause mortality. Prior studies had demonstrated that TIPS reduced the rebleeding rate but increased the rate of hepatic encephalopathy without improving survival. As such, TIPS had only been recommended as a rescue therapy. Obviously, this study presents compelling data that challenge these paradigms. The authors note that in “patients with Child-Pugh class C or in class B with active variceal bleeding, failure to initially control the bleeding or early rebleeding contributes to further deterioration in liver function, which in turn worsens the prognosis and may preclude the use of rescue TIPS.” Authors at UpToDate note that, given the totality of evidence to date, the benefit of early TIPS in preventing rebleeding “is offset by its failure to consistently improve survival and increasing morbidity due to the development of liver failure and encephalopathy.” Today, TIPS remains primarily a salvage therapy for use in cases of recurrent bleeding despite standard pharmacotherapy and EBL. There may be a subset of patients in whom early TIPS is the ideal strategy, but further trials will be required to identify this subset.

Further Reading/References:
1. Wiki Journal Club []
2. 2 Minute Medicine []
3. UpToDate, “Prevention of recurrent variceal hemorrhage in patients with cirrhosis

Summary by Duncan F. Moore, MD

Week 29 – PneumA

“Comparison of 8 vs 15 Days of Antibiotic Therapy for Ventilator-Associated Pneumonia in Adults”

JAMA. 2003 November 19;290(19):2588-2598. [free full text]

Ventilator-associated pneumonia (VAP) is a frequent complication of mechanical ventilation and, prior to this study, few trials had addressed the optimal duration of antibiotic therapy in VAP. Thus, patients frequently received 14- to 21-day antibiotic courses. As antibiotic stewardship efforts increased and awareness grew of the association between prolonged antibiotic courses and the development of multidrug resistant (MDR) infections, more data were needed to clarify the optimal VAP treatment duration.

This 2003 trial by the PneumA Trial Group was the first large randomized trial to compare shorter (8-day) versus longer (15-day) treatment courses for VAP.

The noninferiority study, carried out in 51 French ICUs, enrolled intubated patients with clinical suspicion for VAP and randomized them to either 8 or 15 days of antimicrobials. Antimicrobial regimens were chosen by the treating clinician. 401 patients met eligibility criteria. 197 were randomized to the 8-day regimen. 204 patients were randomized to the 15-day regimen. Study participants were blinded to randomization assignment until day 8. Analysis was performed using an intention-to-treat model. The primary outcomes measured were death from any cause at 28 days, antibiotic-free days, and microbiologically documented pulmonary infection recurrence.

Study findings demonstrated a similar 28-day mortality in both groups (18.8% mortality in 8-day group vs. 17.2% in 15-day group, group difference 90% CI -3.7% to 6.9%). The 8-day group did not develop more recurrent infections (28.9% in 8-day group vs. 26.0% in 15-day group, group difference 90% CI -3.2% to 9.1%). The 8-day group did have more antibiotic-free days when measured at the 28-day point (13.1 in 8-day group vs. 8.7 in 15-day group, p<0.001). A subgroup analysis did show that more 8-day-group patients who had an initial infection with lactose-nonfermenting GNRs developed a recurrent pulmonary infection, so noninferiority was not established in this specific subgroup (40.6% recurrent GNR infection in 8-day group vs. 25.4% in 15-day group, group difference 90% CI 3.9% to 26.6%).

There is no benefit to prolonging VAP treatment to 15 days (except perhaps when Pseudomonas aeruginosa is suspected based on gram stain/culture data). Shorter courses of antibiotics for VAP treatment allow for less antibiotic exposure without increasing rates of recurrent infection or mortality.

The 2016 IDSA guidelines on VAP treatment recommend a 7-day course of antimicrobials for treatment of VAP (as opposed to a longer treatment course such as 8-15 days). These guidelines are based on the IDSA’s own large meta-analysis (of 10 randomized trials, including PneumA, as well as an observational study) which demonstrated that shorter courses of antibiotics (7 days) reduce antibiotic exposure and recurrent pneumonia due to MDR organisms without affecting clinical outcomes, such as mortality. Of note, this 7-day course recommendation also applies to treatment of lactose-nonfermenting GNRs, such as Pseudomonas.

When considering the PneumA trial within the context of the newest IDSA guidelines, we see that we now have over 15 years of evidence supporting the use of shorter VAP treatment courses.

Further Reading/References:
1. 2016 IDSA Guidelines for the Management of HAP/VAP
2. Wiki Journal Club
3. PulmCCM “IDSA Guidelines 2016: HAP, VAP & It’s the End of HCAP as We Know It (And I Feel Fine)”
4. PulmCrit “The siren’s call: Double-coverage for ventilator associated PNA”

Summary by Liz Novick, MD

Image Credit: Joseaperez, CC BY-SA 3.0, via Wikimedia Commons

Week 28 – Symptom-Triggered Benzodiazepines in Alcohol Withdrawal

“Symptom-Triggered vs Fixed-Schedule Doses of Benzodiazepine for Alcohol Withdrawal”

Arch Intern Med. 2002 May 27;162(10):1117-21. [free full text]

Treatment of alcohol withdrawal with benzodiazepines has been the standard of care for decades. However, in the 1990s, benzodiazepine therapy for alcohol withdrawal was generally given via fixed doses. In 1994, a double-blind RCT by Saitz et al. demonstrated that symptom-triggered therapy based on responses to the CIWA-Ar scale reduced treatment duration and the amount of benzodiazepine used relative to a fixed-schedule regimen. This trial had little immediate impact in the treatment of alcohol withdrawal. The authors of the 2002 double-blind RCT sought to confirm the findings from 1994 in a larger population that did not exclude patients with a history of seizures or severe alcohol withdrawal.

The trial enrolled consecutive patients admitted to the inpatient alcohol treatment units at two European universities (excluding those with “major cognitive, psychiatric, or medical comorbidity”) and randomized them to treatment with either scheduled placebo (30mg q6hrs x4, followed by 15mg q6hrs x8) with additional PRN oxazepam 15mg for CIWA score 8-15 and 30mg for CIWA score > 15 or to treatment with scheduled oxazepam (30mg q6hrs x4, followed by 15mg q6hrs x8) with additional PRN oxazepam 15mg for CIWA score 8-15 and 30mg for CIWA score > 15.

The primary outcomes were cumulative oxazepam dose at 72 hours and duration of treatment with oxazepam. Subgroup analysis included the exclusion of symptomatic patients who did not require any oxazepam. Secondary outcomes included incidence of seizures, hallucinations, and delirium tremens at 72 hours.

117 patients completed the trial. 56 had been randomized to the symptom-triggered group, and 61 had been randomized to the fixed-schedule group. The groups were similar in all baseline characteristics except that the fixed-schedule group had on average a 5-hour longer interval since last drink prior to admission. While only 39% of the symptom-triggered group actually received oxazepam, 100% of the fixed-schedule group did (p < 0.001). Patients in the symptom-triggered group received a mean cumulative dose of 37.5mg versus 231.4mg in the fixed-schedule group (p < 0.001). The mean duration of oxazepam treatment was 20.0 hours in the symptom-triggered group versus 62.7 hours in the fixed-schedule group. The group difference in total oxazepam dose persisted even when patients who did not receive any oxazepam were excluded. Among patients who did receive oxazepam, patients in the symptom-triggered group received 95.4 ± 107.7mg versus 231.4 ± 29.4mg in the fixed-dose group (p < 0.001). Only one patient in the symptom-triggered group sustained a seizure. There were no seizures, hallucinations, or episodes of delirium tremens in any of the other 116 patients. The two treatment groups had similar quality-of-life and symptom scores aside from slightly higher physical functioning in the symptom-triggered group (p < 0.01). See Table 2.

Symptom-triggered administration of benzodiazepines in alcohol withdrawal led to a six-fold reduction in cumulative benzodiazepine use and a much shorter duration of pharmacotherapy than fixed-schedule administration. This more restrictive and responsive strategy did not increase the risk of major adverse outcomes such as seizure or DTs and also did not result in increased patient discomfort.

Overall, this study confirmed the findings of the landmark study by Saitz et al. from eight years prior. Additionally, this trial was larger and did not exclude patients with a prior history of withdrawal seizures or severe withdrawal. The fact that both studies took place in inpatient specialty psychiatry units limits their generalizability to our inpatient general medicine populations.

Why the initial 1994 study did not gain clinical traction remains unclear. Both studies have been well-cited over the ensuing decades, and the paradigm has shifted firmly toward symptom-triggered benzodiazepine regimens using the CIWA scale. While a 2010 Cochrane review cites only the 1994 study, Wiki Journal Club and 2 Minute Medicine have entries on this 2002 study but not on the equally impressive 1994 study.

Further Reading/References:
1. “Individualized treatment for alcohol withdrawal. A randomized double-blind controlled trial.” JAMA. 1994.
2. Clinical Institute Withdrawal Assessment of Alcohol Scale, Revised (CIWA-Ar)
3. Wiki Journal Club
4. 2 Minute Medicine
5. “Benzodiazepines for alcohol withdrawal.” Cochrane Database Syst Rev. 2010

Summary by Duncan F. Moore, MD

Image Credit: VisualBeo, CC BY-SA 3.0, via Wikimedia Commons

Week 26 – HACA

“Mild Therapeutic Hypothermia to Improve the Neurologic Outcome After Cardiac Arrest”

by the Hypothermia After Cardiac Arrest Study Group

N Engl J Med. 2002 Feb 21;346(8):549-56. [free full text]

Neurologic injury after cardiac arrest is a significant source of morbidity and mortality. It is hypothesized that brain reperfusion injury (via the generation of free radicals and other inflammatory mediators) following ischemic time is the primary pathophysiologic basis. Animal models and limited human studies have demonstrated that patients treated with mild hypothermia following cardiac arrest have improved neurologic outcome. The 2002 HACA study sought to evaluate prospectively the utility of therapeutic hypothermia in reducing neurologic sequelae and mortality post-arrest.

Population: European patients who achieve return of spontaneous circulation (ROSC) after presenting to the ED in cardiac arrest

inclusion criteria: witnessed arrest, ventricular fibrillation or non-perfusing ventricular tachycardia as initial rhythm, estimated interval 5 to 15 min from collapse to first resuscitation attempt, no more than 60 min from collapse to ROSC, age 18-75

pertinent exclusion: pt already < 30ºC on admission, comatose state prior to arrest d/t CNS drugs, response to commands following ROSC

Intervention: Cooling to target temperature 32-34ºC with maintenance for 24 hrs followed by passive rewarming. Pts received pancuronium for neuromuscular blockade to prevent shivering.

Comparison: Standard intensive care

Primary: a “favorable neurologic outcome” at 6 months defined as Pittsburgh cerebral-performance scale category 1 (good recovery) or 2 (moderate disability). (Of note, the examiner was blinded to treatment group allocation.)


  • all-cause mortality at 6 months
  • specific complications within the first 7 days: bleeding “of any severity,” pneumonia, sepsis, pancreatitis, renal failure, pulmonary edema, seizures, arrhythmias, and pressure sores

3551 consecutive patients were assessed for enrollment and ultimately 275 met inclusion criteria and were randomized. The normothermia group had more baseline DM and CAD and were more likely to have received BLS from a bystander prior to the ED.

Regarding neurologic outcome at 6 months, 75 of 136 (55%) of the hypothermia group had a favorable neurologic outcome, versus 54/137 (39%) in the normothermia group (RR 1.40, 95% CI 1.08-1.81, p = 0.009; NNT = 6). After adjusting for all baseline characteristics, the RR increased slightly to 1.47 (95% CI 1.09-1.82).

Regarding death at 6 months, 41% of the hypothermia group had died, versus 55% of the normothermia group (RR 0.74, 95% CI 0.58-0.95, p = 0.02; NNT = 7). After adjusting for all baseline characteristics, RR = 0.62 (95% CI 0.36-0.95). There was no difference among the two groups in the rate of any complication or in the total number of complications during the first 7 days.

In ED patients with Vfib or pulseless VT arrest who did not have meaningful response to commands after ROSC, immediate therapeutic hypothermia reduced the rate of neurologic sequelae and mortality at 6 months.

Corresponding practice point from Dr. Sonti and Dr. Vinayak and their Georgetown Critical Care Top 40: “If after ROSC your patient remains unresponsive and does not have refractory hypoxemia/hypotension/coagulopathy, you should initiate therapeutic hypothermia even if the arrest was PEA. The benefit seen was substantial and any proposed biologic mechanism would seemingly apply to all causes of cardiac arrest. The investigators used pancuronium to prevent shivering; [at MGUH] there is a ‘shivering’ protocol in place and if refractory, paralytics can be used.”

This trial, as well as a concurrent publication by Benard et al. ushered in a new paradigm of therapeutic hypothermia or “targeted temperature management” (TTM) following cardiac arrest. Numerous trials in related populations and with modified interventions (e.g. target temperature 36º C) were performed over the following decade, and ultimately led to the current standard of practice.

Per UpToDate, the collective trial data suggest that “active control of the post-cardiac arrest patient’s core temperature, with a target between 32 and 36ºC, followed by active avoidance of fever, is the optimal strategy to promote patient survival.” TTM should be undertaken in all patients who do not follow commands or have purposeful movements following ROSC. Expert opinion at UpToDate recommends maintaining temperature control for at least 48 hours.

Further Reading/References:
1. HACA @ 2 Minute Medicine
2. HACA @ Wiki Journal Club
3. HACA @ Visualmed
4. Georgetown Critical Care Top 40, page 23 (Jan. 2016)
5. PulmCCM.org, “Hypothermia did not help after out-of-hospital cardiac arrest, in largest study yet”
6. Cochrane Review, “Hypothermia for neuroprotection in adults after cardiopulmonary resuscitation”
7. The NNT, “Mild Therapeutic Hypothermia for Neuroprotection Following CPR”
8. UpToDate, “Post-cardiac arrest management in adults”

Summary by Duncan F. Moore, MD

Week 23 – TRICC

“A Multicenter, Randomized, Controlled Clinical Trial of Transfusion Requirements in Critical Care”

N Engl J Med. 1999 Feb 11; 340(6): 409-417. [free full text]

Although intuitively a hemoglobin closer to normal physiologic concentration seems like it would be beneficial, the vast majority of the time in inpatient settings we use a hemoglobin concentration of 7g/dL as our threshold for transfusion in anemia. Historically, higher hemoglobin cutoffs were used with aims to keep Hgb > 10g/dL. In 1999, the landmark TRICC trial demonstrated no mortality benefit in the liberal transfusion strategy and harm in certain subgroup analyses.


Inclusion: critically ill patients expected to be in ICU > 24h, Hgb ≤ 9g/dL within 72hr of ICU admission, and clinically euvolemic after fluid resuscitation

Exclusion criteria: age < 16, inability to receive blood products, active bleed, chronic anemia, pregnancy, brain death, consideration of withdrawal of care, and admission after routine cardiac procedure.

Patients were randomized to either a liberal transfusion strategy (transfuse to Hgb goal 10-12g/dL, n = 420) or a restrictive strategy (transfuse to Hgb goal 7-9g/dL, n = 418). The primary outcome was 30-day all-cause mortality. Secondary outcomes included 60-day all-cause mortality, mortality during hospital stay (ICU plus step-down), multiple-organ dysfunction score, and change in organ dysfunction from baseline. Subgroup analyses included APACHE II score ≤ 20 (i.e. less-ill patients), patients younger than 55, cardiac disease, severe infection/septic shock, and trauma.

The primary outcome of 30-day mortality was similar between the two groups (18.7% vs. 23.3%, p = 0.11). The secondary outcome of mortality rate during hospitalization was lower in the restrictive strategy (22.2% vs. 28.1%, p = 0.05). (Of note, the mean length of stay was about 35 days for both groups.) 60-day all-cause mortality trended towards lower in the restrictive strategy although did not reach statistical significance (22.7% vs. 26.5 %, p = 0.23). Between the two groups, there was no significant difference in multiple-organ dysfunction score or change in organ dysfunction from baseline.

Subgroup analyses in patients with APACHE II score ≤ 20 and patients younger than 55 demonstrated lower 30-day mortality and lower multiple-organ dysfunction score among patients treated with the restrictive strategy. In the subgroups of primary disease process (i.e. cardiac disease, severe infection/septic shock, and trauma) there was no significant differences among treatment arms.

Complications in the ICU were monitored, and there was a significant increase in cardiac events (primarily pulmonary edema) in the liberal strategy group when compared to the restrictive strategy group.

The TRICC trial demonstrated that, among ICU patients with anemia, there was no difference in 30-day mortality between a restrictive and liberal transfusion strategy. Secondary outcomes were notable for a decrease in inpatient mortality with the restrictive strategy. Furthermore, subgroup analyses showed benefit in various metrics for a restrictive transfusion strategy when adjusting for younger and less ill patients. This evidence laid the groundwork for our current standard of transfusing to hemoglobin 7g/dL. A restrictive strategy has also been supported by more recent studies. In 2014 the Transfusion Thresholds in Septic Shock (TRISS) study showed no change in 90-day mortality with a restrictive strategy. Additionally, in 2013 the Transfusion Strategy for Acute Upper Gastrointestinal Bleeding study showed reduced 40-day mortality in the restrictive strategy. However, the study’s exclusion of patients who had massive exsanguination or low rebleeding risk reduced generalizability. Currently, the Surviving Sepsis Campaign endorses transfusing RBCs only when Hgb < 7g/dL unless there are extenuating circumstances such as MI, severe hypoxemia, or active hemorrhage.

Further reading:
1. TRICC @ Wiki Journal Club, @ 2 Minute Medicine
2. TRISS @ Wiki Journal Club, full text, Georgetown Critical Care Top 40 pages 14-15
3. “Transfusion strategies for acute upper gastrointestinal bleeding” (NEJM 2013) @ 52 in 52 (2017-2018) Week 46), @ Wiki Journal Club, full text
4. “Surviving Sepsis Campaign: International Guidelines for Management of Sepsis and Septic Shock 2016”

Summary by Gordon Pelegrin, MD

Image Credit: U.S. Air Force Master Sgt. Tracy L. DeMarco, US public domain, via WikiMedia Commons

Week 20 – Omeprazole for Bleeding Peptic Ulcers

“Effect of Intravenous Omeprazole on Recurrent Bleeding After Endoscopic Treatment of Bleeding Peptic Ulcers”

N Engl J Med. 2000 Aug 3;343(5):310-6. [free full text]

Intravenous proton-pump inhibitor (PPI) therapy is a cornerstone of modern therapy for bleeding peptic ulcers. However, prior to this 2000 study by Lau et al., the role of PPIs in the prevention of recurrent bleeding after endoscopic treatment was unclear. At the time, re-bleeding rates after endoscopic treatment were noted to be approximately 15-20%. Although other studies had approached this question, no high-quality, large, blinded RCT had examined adjuvant PPI use immediately following endoscopic treatment.

The study enrolled patients who had a bleeding gastroduodenal ulcer visualized on endoscopy and in whom hemostasis was achieved following epinephrine injection and thermocoagulation. Enrollees were randomized to treatment with either omeprazole 80mg IV bolus followed by 8mg/hr infusion x72 hours then followed by omeprazole 20mg PO x8 weeks or to placebo bolus + drip x72 hours followed by omeprazole 20mg PO x8 weeks. The primary outcome was recurrent bleeding within 30 days. Secondary outcomes included recurrent bleeding within 72 hours, amount of blood transfused by day 30, hospitalization duration, and all-cause 30-day mortality.

120 patients were randomized to each arm. The trial was terminated early due to the finding on interim analysis of a significantly lower recurrent bleeding rate in the omeprazole arm. Bleeding re-occurred within 30 days in 8 (6.7%) omeprazole patients versus 27 (22.5%) placebo patients (HR 3.9, 95% CI 1.7-9.0; NNT 6.3). A Cox proportional-hazards model, when adjusted for size and location of ulcers, presence/absence of coexisting illness, and history of ulcer disease, revealed a similar hazard ratio (HR 3.9, 95% CI 1.7-9.1). Recurrent bleeding was most common during the first 72 hrs (4.2% of the omeprazole group versus 20% of the placebo group, RR 4.80, 95% CI 1.89-12.2, p<0.001). For a nice visualization of the early separation of re-bleeding rates, see the Kaplan-Meier curve in Figure 1. The mean number of units of blood transfused within 30 days was 2.7 ± 2.5 in the omeprazole group versus 3.5 ± 3.8 in the placebo group (p = 0.04). Regarding duration of hospitalization, 46.7% of omeprazole patients were admitted for < 5 days versus 31.7% of placebo patients (p = 0.02). Median stay was 4 days in the omeprazole group versus 5 days in the placebo group (p = 0.006). 4.2% of the omeprazole patients died within 30 days, whereas 10% of the placebo patients died (p = 0.13).

Treatment with intravenous omeprazole immediately following endoscopic intervention for bleeding peptic ulcer significantly reduced the rate of recurrent bleeding. This effect was most prominent within the first 3 days of therapy. This intervention also reduced blood transfusion requirements and shortened hospital stays. The presumed mechanism of action is increased gastric pH facilitating platelet aggregation. In 2018, the benefit of this intervention seems so obvious based on its description alone that one would not imagine that such a trial would be funded or published in such a high-profile journal. However, the annals of medicine are littered with now-discarded interventions that made sense from a theoretical or mechanistic perspective but were demonstrated to be ineffective or even harmful (e.g. pharmacologic suppression of ventricular arrhythmias post-MI or renal denervation for refractory HTN).

Today, bleeding peptic ulcers are treated with an IV PPI twice daily. Per UpToDate, meta-analyses have not shown a benefit of continuous PPI infusion over this IV BID dosing. However, per 2012 guidelines in the American Journal of Gastroenterology, patients with active bleeding or non-bleeding visible vessels should receive both endoscopic intervention and IV PPI bolus followed by infusion.

Further Reading/References:
1. Wiki Journal Club
2. 2 Minute Medicine
3. UpToDate, “Overview of the Treatment of Bleeding Peptic Ulcers”
4. Laine L, Jensen DM. “Management of patients with ulcer bleeding.” Am J Gastroenterol. 2012

Summary by Duncan F. Moore, MD

Image credit: Wesalius, CC BY 4.0, via Wikimedia Commons