2011 NCS INTERNATIONAL CONSENSUS GUIDELINES ON THE CRITICAL CARE OF SAH: SOME SURPRISES ARE IN STORE

The complex and varied nature of SAH involves a lot of decision making when there is less than definitive evidence. How do you write orders for fluids and pressors? What does it take to trigger hypertensive, hypervolemic therapy in your patients? When do you transfuse blood? Do you uniformly use statins or magnesium? How do you treat fevers, which occur in 70% of patients?

The list goes on and on. To address these many questions, which has lead to a tremendous amount of practice variation throughout the world, NCS sponsored the first International Consensus Conference on the Critical Care of SAH, held in paris in October of 2010. Chaired by former NCS President Michael N. Diringer of Washington University in St Louis, the conference was designed to complement existing guidelines from the American Heart Association, focusing on day-t0-day treatment and decision making in the critical care unit.

This activity was supported by a generous unrestricted educational grant from Actelion Pharaceuticals, sponsors of the Conscious trials of clazosentan, an endothelin-receptor antagonist vasodilator and putative treatment for vasospasm. The motivation of this company, with NCS as its willing partner, was to try to establish greater consistency and adherence to best medical practices for SAH in the ICU by creating an evidence-based document that identifies standards, guidelines, and options for the ICU management of patients with SAH. The company had absolutely no say in the design or content of the proceedings, or in its final product.

A multidisciplinary panel of 27 experts in SAH management, drawn from Europe and North America, representing neurosurgery, neurocritical care, interventional neuroradiology and neuroanesthesia participated. They used the GRADE system to evaluate the quality of the data, and made final recommendations that were classified as strong or weak cased on the quality of the evidence, taking into account (1) on the quality of the evidence, (2) the perceived risk versus benefit trade-offs of a given intervention, and (3) the feasibility of translating clinical trial and cohort studies into clinical practice.

Seventeen individual topics were reviewed. Here, to my mind, are some of the most interesting recommendations:

1. Consider an early, short course of antifibrinolytic therapy prior to early aneurysm repair (begun at diagnosis, for a maximum of 72 hours post onset. Although this is a weak recommendation, it is evidence based, and we have been doing this at Columbia for several years. It makes sense. I tell my fellows that when you get the call to transfer in an SAH patient, telling them to give a 4g load of Amicar is usually the only meaningful recommendation you are going to make.

2. Routine use of anticonvulsant prophylaxis with phenytoin is not recommended after SAH. This is a strong recommendation. It’s based on several studies linking phenytoin to poor functional and cognitive outcome after SAH. Could this statement is going to trigger a sea-change conversion to IV levetiracetam as the standard prophylactic NICU anticonvulsant?

3. Central venous lines should not be placed solely to obtain CVP values, and fluid administration protocols based solely on CVP targets are not recommended. A strong recommendation, but I strongly disagree. True, CVP does not correlate with measurements of intravascular blood volume across the entire range of values, but very low CVP’s still have good specificity for hypovolemia. We have shown that fluid supplemental boluses for CVP values of <5 mm Hg maintains euvolemia, with a very low risk of fluid overload.

4. Give statins if you want: acute statin therapy may be considered for reducing DCI following aneurysmal SAH, pending the outcome of ongoing trials. This is a weak recommendation. My advice would be if you want to use statins in your ICU, either give them to everyone, or no one, to maintain a sense of consistency.

5. No to magnesium drips: inducing hypermagnesemia is not recommended for reducing the risk of delayed cerebral ischemia pending the conclusion of current randomized trials. This is a strong recommendation based on a large negative RCT.

6. Patients should receive packed RBC transfusions to maintain hemoglobin concentration above 8-10 g/dL. Strong recommendation. Wow. At Columbia we maintain hemoglobin >7 g/dL for asymptomatic patients, and >10 g/dL when patients have symptomatic vasospasm. We probably really do need a trial to evaluate this.

And finally:

7. Patients with SAH should be treated at high volume centers, and these centers should have appropriate specialty neurointensive care units staffed by neurointensivists. Yes! This is a strong recommendation, it is strongly supported by medical evidence, and it has important public health implications.

On this last point, we need you to help spread the word to the public. NCS will issue a press release just prior to the annual meeting in Montreal entitled "Despite Proven Benefits, Few Brain Aneurysm Patients Receive Specialized Care." You can download the press release on the NCS website (see the homepage) and distribute it to your hospital for widespread release through regional news outlets, or pitch it yourself to any health and science reporters that you know.

TEN SPECIFIC RECOMMENDATIONS FOR DEVELOPING A SUCCESSFUL NEURO-ICU PROGRAM

The introduction of neurointensivists into a hospital can lead to uncertainty on the part of other specialists, and in particular other neurologists and neurosurgeons, regarding “turf” and authority. To establish a successful program, several concepts should be kept in mind.

First, avoid placing complete responsibility for decision making on one individual (i.e. the intensivist or surgeon). The neurointensivist should work to build consensus and to facilitate team consensus decision making for all important decisions. No one person is in charge. The team is in charge.

Second, it is undesirable to have competing or duplicative “services” within different departments at the same medical center. For instance, two competing services (i.e. neurology and neurosurgery) that that can admit and provide care to an intracerebral hemorrhage patient in the same can lead to conflict and tension on a daily basis. Neurointensivists function best when they can work with everyone, and when everyone works together.

Finally, critical care is not a 9-to-5 job: intensive care must be provided 365 days a year, 7 days a week, 24 hours a day. It has been said that until you are part of a team delivering around the clock care, your are not truly an intensivist. This creates a special challenge for solo neurointensivists who are starting their own program at a hospital, without adequate coverage to create a constant standard of care. Often when starting out this is unavoidable, but the goal should eventually be to create at the very least a team of caregivers who can deliver a consistent level of care.

Apart from these principles, here is a to-do list of key things that can help make your neurocritical care program highly successful.

1. Identify a Neuro-ICU directors from neurology, neurosurgery, anesthesiology, or another discipline, who is interested in a commitment to working together, sharing responsibility, andy leading a multidisciplinary team. Specific administrative responsibilities should include final responsibility for patient transfers, bed flow, nursing and resident education, QA, and protocol development. Key point: the old-school model of a neurosurgeon who rules with an iron fist, and is physically absent most of the time, no longer cuts it.

2. Emphasize the multidisciplinary nature of the ICU, use a “global” descriptive term for the unit such as “neurological,” “neuroscience,” or “neurocritical” rather than "neurological" or “neurosurgical” alone. These terms can be used to divide rather than unify, and reinforce old concepts of what a neurologist or neurosurgeon does or doesn’t do.

3. A minimum of 2 neurointensivists are required at any medical center to start a neuro-ICU program, to avoid physician burn-out. If this is not possible covering medical or surgical intensivists should be identified, but it will hard to really get rolling without the critical mass that results from two like-minded individuals. Approximately 25% to 60% of a neurointensivist’s time should be devoted to patient care.

4. Specialized neurocritical care nurses are possibly the single most important aspect of care in the neuro-ICU. The hospital should identify a nurse manager for the ICU with expertise in neuro-ICU care nursing, and make this individual responsible for nursing education, care standards, patient flow, and protocol development, working in conjunction with the unit directors.

5. A common pathway (i.e. central phone number and coverage system) and mechanism for triaging and arranging interhospital transfers should be developed. A corollary of this that an efficient protocol for identifying and expediting transfers out of the ICU should be developed. This function absolutely cannot be delegated to team members who are not absolutely committed to bringing in every transfer possible, with extreme prejudice.

6. Bullet rounds should be held every morning with neurosurgery, and the ICU team. The doctors who are authorized to make the decisions must be involved. Wasting time later in the day to “get an okay” or “run things by” attending physicians is murder in terms of efficiency. The focus is on establishing an agreed-upon action plan for each patient when it comes to interventions of mutual interest (i.e. angiography, pulling an EVD, CT scans, ICP monitor, establishing threshold criteria for a hemicraniectomy), and on patient flow into and out of the ICU. Decision-making should be made by consensus.

7. Things can change quickly. A clearly delineated protocol should be established for communicating clinical events throughout the day. For instance, in a teaching hospital, the nurse should communicate changes to the primary ICU resident; and the resident calls the fellow (or neurointensivist attending), and is also responsible for directly calling the neurosurgery resident on-call for any “red flag” event. If the nurses constantly run directly to the ICU director with every problem, everyone else might as well go home.

8. The ultimate goal for every patient, with the exception of end-of-life cases, is to stabilize the patient and get them into a recovery environment as quickly as possible. If you direct every action to getting the patient out of the ICU and discharged as quickly as possible, it simply follows that you have to make them better or help them have their peace, and you will always be making the right decision. A multidisciplinary team should meet daily to create athe plan for post-ICU and post-hospital care. This should include doctors, nurses, social work, care coordinators, and most importantly the providers of post-ICU floor or step-down unit care.

9. A commitment should be made by the hospital to provide adequate technological resources and support staff for advanced neuromonitoring (i.e. continuous EEG monitoring, ICP and brain tissue oxygen monitors, hypothermia devices, multimodality data management systems, transcranial Doppler). This is what makes the neuro-ICU special.

10. Finally, it has long been said that “hospitals are essentially turning into big ICUs.” This may never actually be the case, but it is true that well-integrated care delivery teams on the hospital floor can promote patient stability after transfer from the ICU, lead to the more widespread adoption of best medical practices, and facilitate discharge planning for rehabilitation. Neurontensivists can improve outcomes for patients discharged from the ICU and supplement their practice by creating or covering a “neurohospitalist” service which covers floor and step-down patients, runs interdisciplinary rounds on the floor, provides hospital consultations, and covers the emergency room.

Okay, that’s it, says me. Am I missing anything important?

IS A NEURO-ICU REALLY OPTIONAL?

Efforts to re-examine criteria and performance standards for comprehensive stroke centers (CSCs) are back in full swing. Now that the nation has seen a proliferation of primary stroke centers, capable of providing IV t-PA 24 hours a day, the next task will be for JCAHO to put teeth into the CSC concept and provide certification for centers that apply for this designation. Both JCAHO and the AHA are working behind the scenes to define both what a comprehensive stroke center is, and what the key quality measures should be.

In 2005 the first consensus statement from the AHA on the topic, entitled "Recommendations for Comprehensive Stroke Centers," made some waves (1,2). This document mapped out the basic idea of what a CSC should be. It said that vascular neurologists, neurosurgeons, and general intensivists are essential, but that neurointensivists are optional. The document said that an ICU is essential, but a dedicated neuro-ICU is optional.

On the other hand, there is a great deal of outcomes research showing that dedicated neuro-ICU's, neurointensivist involvement, and transition from an open to intensivist-driven semi-closed model are all associated with improved outcomes, more efficient resource utilization, or both (3). The evidence is compelling.

The main argument for making neuro-ICU's and neurointensivists optional in 2005 went something like this: "Yes, you guys are great, awesome, and in a perfect world every CSC would have a dedicated neuro-ICU and a team of neurointensivsts. But it's just not realistic. There are not enough of you guys. And some hospitals are never going to be able to structurally accomodate a dedicated neuro-ICU."

But the world is a different place. That was before we had 389 UNCS-certified Neurocritical Care diplomates, and more coming every year. Before we had over 1000 NCS members. Before we had Leapfrog recognition. The time has come to push forcefully for change.

There may be a bias among those who write stroke guidelines emphasizing ischemic stroke, as opposed to SAH and ICH. It may also be that some of those who write the stroke guidelines are vascular neurologists or neurosurgeons who are perfectly comfortable with the "old-school" model of working general intensivists who manage the ventilator, drips, and antibiotics, but otherwise defer completely neurological decsion-making.

The Advocacy mission of NCS is to "make the case that complex, life-threatening neurological diseases are best cared for by a multidisciplinary team with special expertise in neurocritical care." In line with that, we need to make sure that the same type of team-based integrated care that occurs in the stroke unit does not fall apart when patients are moved into an open ICU. Any new CSC guidelines need to "strongly encourage" a dedicated neuro-ICU whenever possible, involvement of neurointensivists or general intensivists who document ongoing CME training in neurocritical care, and a team-based collaborative practice model that involves protocols and order sets emphasizing best practices, shared decision making, and multidisciplinary rounds.

At the local level, think about this: is your hospital ready to become a comprehensive stroke center? What could be improved in terms of the way that ICU care is organized and delivered? If you have any thoughts on this, please respond!