Is anesthesia getting safer? What are the biggest risks of modern anesthesia? How can you work to address them? To get a sense of the issues and answers, we interviewed practitioners who've taken the time to examine them. Over the next several pages, our group talks about issues ranging from airway management to runaway sedation and from obesity to office-based surgery. Here's what they had to say.

Current State of Affairs
Robert S. Lagasse, MD, is a professor and vice chairman of the department of anesthesiology, Weiler Division of Montefiore Medical Center, in New York, N.Y. He authored "Anesthesia Safety: Model or Myth," a review article that appeared in the December 2002 issue of Anesthesiology.

Question: Anesthesia poses two basic kinds of risks: human- and system-related. What are they and how prevalent are they?

Answer: About 90 percent of anesthesia-related mortalities are "system-related" in that we're doing what we're supposed to do, but something still goes wrong. The most common system errors stem from the limitations in our current abilities to diagnose and treat patients. For example, we don't yet fully understand how to predict a difficult airway with 100 percent certainty. Our current resuscitation methods aren't guaranteed to work in every case even when we do everything correctly. Our current standards don't allow us to know every single pre-existing condition or allergy.

Other system-related errors ' over which we do have control ' include equipment malfunctions, a lack of resources (like blood products) or supervision (like insufficient oversight of residents in teaching institutions), communication errors and technical accidents. A communication error occurs when everyone follows the prescribed lines of communication but something in the system breaks down; for example, an important medical consultant's report is delayed. A technical accident occurs when a preventable adverse event develops even though the practitioner performed the procedure properly. For example, we noted higher epidural wet tap rates at one of our hospitals because practitioners there were still using glass syringes. The only way to uncover problems like this is to analyze outcomes.

The remaining 10 percent of anesthesia mortalities are human-related. They involve direct one-on-one patient care errors like improper technique, misuse of equipment (neglecting to perform equipment checks, for example), failure to avoid a known drug allergen, failure to seek appropriate data about the patient (such as when the practitioner doesn't adhere to extubation criteria) or inadequate knowledge or training (which could lead to incorrect interpretation of hemodynamic variables, for example).

Despite reports to the contrary, human errors don't cause more serious outcomes than system-related errors. The 90:10 ratio of system-to-human errors is the same whether you're looking at mortality or all adverse events. Also, this 10 percent rate is similar to that of other industries, which suggests that it may simply represent inherent human limitations.

Question: Why is there such a strong emphasis in the literature on airway management problems?

Answer: Premature extubation is the most common human error, and practitioners tend to report airway management complications because there's a widespread belief that there's room to improve upon these problems. Some time ago, here in our facility, we worked on creating extubation guidelines because our reintubation rate was 1 in 300. It's not a simple issue, because leaving the endotracheal tube in for a prolonged period also creates risk. However, we've since risen to the current standard of 1 in 1,000 thanks to this effort.

Question: Your review article states that anesthesia-related mortality hasn't improved by any order of magnitude over the years. This is counter to the general perception that safety has improved greatly. Can you explain?

Answer: Current data suggest that the overall perioperative mortality rate for patients with ASA physical status 1 through 5 is about 1 per 500 anesthetics, and that the anesthesia-related mortality rate, in particular, is 1 per 13,000 anesthetics. These data are consistent with other reports and haven't changed during the past decade. We must dispel the myth that anesthesia-related mortality has improved by any order of magnitude.

There may be several reasons for this. For one, safety systems and technologies create risk as they become more sophisticated. This is the theory posed by systems behavior expert Charles Perrow, who showed that, while the safety of airline travel increased dramatically with the development of radar, airline fatality rates really haven't improved since the 1960s. In effect, technology made it possible to allow more takeoffs, landings and traffic density, adding risk and offsetting improvements in safety. Anesthesia has undergone a similar change. Improved monitoring equipment, better anesthetic agents and new surgical approaches are allowing us to perform more surgeries faster and on older, sicker and heavier patients ' factors that, in and of themselves, elevate risk. In fact, the risk of death related to an anesthesia practitioner's human error increases exponentially with ASA status. When we talk about human errors and adverse outcomes, we may need to consider the number of clinical judgments that go into the case, not just the number of procedures or anesthetics.

Question: How can practitioners and facility managers work to reduce these errors?

Answer: Reporting of adverse events is extremely important. To encourage reporting, we shouldn't identify the practitioner in written reports and we should make sure the reporting is done solely for the benefit of patient safety. We have 300 to 400 adverse outcome reviews in our facility every year out of about 30,000 anesthetics. With reporting, our human error rate decreased from 11.5 percent to 3 percent. Mere discussion goes a long way in preventing practitioners from making the same error twice. When your peers agree it's a human error, the practitioner won't make the same mistake twice.

When you're just starting out with a reporting system, I recommend selecting an initiative that's likely to succeed. It's important to demonstrate success from the get-go and show everyone that your intent is patient safety ' not finger-pointing. You can tackle the harder issues later. If you can't create your own reporting system, consider cooperating with one of the national databases now underway, like the Veterans Health Administration's National Surgical Quality Improvement Project (www1.va.gov/surgery/) or the Surgical Care Improvement Project (www.medqic.org/scip). These databases are excellent and are helping us develop risk models and identify best practices so we can ultimately make anesthesia a model of safety.

Anesthesia for the Obese Patient
Anthony N. Passannante, MD, is an associate professor of anesthesiology at the University of North Carolina School of Medicine in Chapel Hill. He co-authored "Anesthetic Management of Patients with Obesity and Sleep Apnea," which appeared last year in Anesthesiology Clinics of North America.

Question: Nearly one-third of the U.S. population is considered obese and nearly 5 percent of Americans are morbidly obese. How well are surgical facilities responding to this challenge?

Answer: Overall, facilities are adapting with a thoughtful, step-wise approach. This is smart because these patients carry more risk. Many facilities, for example, are gravitating toward limb and certain breast procedures in heavier patients but may still shy away from more invasive, intra-abdominal procedures that involve more blood loss, more pain and hospital admission rates as high as 10 percent. Many facilities have also set a weight limit for patients, typically 300 pounds. We'll have to raise this if current trends continue. Surgical facilities won't be able to shut out 40 percent of the population and survive.

I believe, though, that it would make more sense to use BMI as the criterion for accepting surgical patients in facilities that aren't otherwise equipped to manage obese patients. There is a great difference between a person who weighs a lot because he has big bones and substantial muscle mass and a person who weighs a lot because he's obese. The former may be very healthy; the latter is more likely to have comorbidities that complicate anesthesia.

Question: What are the risks of anesthesia in the obese patient?

Answer: Some of the risks include altered pharmacokinetics; increased risk of hypoxia during general anesthesia; and impaired respiration related to obstructive sleep apnea, the obesity itself or a pneumoperitoneum. There are also higher rates of peripheral nerve block failures and acute PNB complications in obese patients.

Although not directly related to anesthesia, patient positioning is also very challenging and very important, especially during longer procedures, to prevent complications from pressure points and even rhadbomyolysis of the gluteal muscles leading to renal failure. Standard equipment is designed for patients with BMIs of about 23. Large patients don't have the same shape as average-sized patients, and it's not reasonable to expect you can position them in the same way. Doing so jeopardizes their safety.

Question: How does obesity alter the pharmacokinetics of anesthetic agents?

Answer: Plasma protein binding, body composition and regional blood flow all differ in obese patients. We don't have a lot of pharmacokinetic studies in obese patients, but we know that certain drugs shouldn't be dosed based on total body weight. To avoid overdose, opioids like fentanyl and remifentanil should be dosed based on ideal body weight in obese patients. Nondepolarizing neuromuscular agents should be dosed on IBW, too, to avoid prolonged duration of action. However, studies do show that propofol can be dosed in the usual fashion (based on TBW) without any known adverse effects.

Question: What should be done to address the increased risk of hypoxia in obese patients during general anesthesia?

Answer: Obese patients can have a myriad of respiratory dysfunctions, and they can desaturate fast. They use more oxygen per minute, have less oxygen stored in the functional residual capacity of the lung and don't tolerate anesthesia-induced apnea. It's critical to do everything possible to avoid hypoxemia. First, pre-oxygenation is a basic and simple tenet. Second, I can't emphasize strongly enough the importance of proper positioning and paying extreme attention to detail during laryngoscopy. The patient must be in the ramped-up position with a lot of blankets or pillows under the shoulders, neck and head. This makes it so much easier for the practitioner to visualize anatomy and minimizes the chance of difficult intubation. We need to keep in mind how difficult it is to reposition the obese patient when intubation fails. I also recommend awake intubation if there are any safety concerns and having an LMA immediately available.

Question: How can facilities ensure safety of the patient with obstructive sleep apnea?

Answer: Foremost, thorough pre-op screening is critical ' including careful evaluation of intubating conditions and thorough assessment of cardiopulmonary morbidity ' because many obese people have OSA. If you know about OSA beforehand, you can be prepared to manage it and patients can use their CPAP equipment post-operatively. The more difficult situation, however, is the patient with undiagnosed OSA. Unfortunately, the overwhelming majority of OSA patients aren't sufficiently diagnosed, so we must have a very high index of suspicion for every single obese patient. In 2005, the American Society of Anesthesiologists released practice parameters for screening patients ("Practice Guidelines for the Perioperative Management of Patients with Obstructive Sleep Apnea"), and the American Sleep Apnea Association also offers a screening tool.

During general anesthesia, we should be certain that we don't extubate obese patients until they're conscious. By conscious, I don't mean the patient is just flailing around. The patient must be awake and responsive.

We must also be very careful about the level of post-op sedation we allow these patients, because even standard doses of opioids can exaggerate OSA and lead to airway obstruction during recovery. This is a challenge, because we don't want to deprive these patients of needed pain control.

Before sending these patients home, we must also ensure that they fully return to their pre-op level of consciousness. The patient should no longer require supplemental oxygen, must be fully conscious with no airway obstruction and have pain adequately controlled. The more opioids on board and the more invasive the procedure, the harder this is to achieve. I highly recommend using sedation scores and ensuring that your recovery nurses are experienced with obese patients. If pain requires sedation to a level that impairs consciousness, don't send these patients home.

Question: Is regional anesthesia a good alternative for obese patients?

Answer: There is no simple answer, but neuraxial blocks and PNBs should definitely be considered for obese patients. Regional blocks can help contain pain and reduce or eliminate the need for opioids after surgery, and this is a real advantage because opioids put these patients at real risk after they leave the facility. But this decision should depend on how much sedation the patient is going to need intraoperatively, because significant sedation in the obese patient with an unsecured airway is a very bad idea. Practitioner expertise is also a consideration, because patient positioning and depth of needle penetration differ in obese patients and require an excellent working knowledge of anatomy.

Question: What is the facility manager's role in ensuring the safety of obese patients?

Answer: The facility manager's role in patient selection can't be understated. These patients require specialized equipment and experienced personnel, and only the facility manager has the perspective on what the facility can and can't support. You need adequate OR tables, large transport beds, proper airway management equipment, and qualified personnel in the OR and in recovery. The manager should also have perspective on what makes sense for the facility given local demographics. If you serve a significant obese population, for example, the facility manager may work to equip the facility to ensure the needed revenue stream. But, if you have a surgeon who wants to bring in an occasional obese patient for a lap chole, for example, the manager can exercise discretion and make it understood that this probably doesn't make sense.

Sedation During MAC
Lars E. Helgeson, MD, is an assistant professor in the department of anesthesiology at the Yale University School of Medicine in New Haven, Conn. He authored "Sedation During Regional Anesthesia: Inhalation vs Intravenous," which appeared in Current Opinion in Anesthesiology last year.

Question: Patients who receive monitored anesthesia care can lapse into deep sedation or transient general anesthesia, which increases the risk of respiratory depression, obstruction and aspiration. How common is this and why does it occur?

Answer: It's more common than we realize. In fact, I think it happens fairly frequently. A primary reason is production pressure, which can come down on the anesthesia practitioner from many sources, including late-arriving surgeons. The practitioner may too rapidly administer sequential doses of the drugs in an effort to achieve rapid sedation. The doses stack up, potentially resulting in an unexpected induction of general anesthesia. In patients with a low cardiac output or slow circulation, this can happen even when administering the right doses at the right rate. Or the practitioner may push a single large dose. You can induce anesthesia at a moment's notice, but not sedation.

Excess sedation can also occur when the practitioner tries to meet unrealistic patient or surgeon expectations. That is, patients might perceive they're not sedated enough if they feel pressure. Or surgeons might expect a non-moving, quiet patient. This is pseudo-MAC and instead describes general anesthesia ' not MAC. It's important to understand that MAC means an arousable patient, not an unresponsive one. The anesthesia practitioner must understand pre-op what surgeons want and expect.

Finally, some anesthesia practitioners may mistakenly administer MAC for major procedures that really require general anesthesia ' like substantial inguinal hernia repairs or breast biopsies. In these cases, it's all too easy to push the envelope in an effort to make patients comfortable. This is a mistake, because the practitioner can induce general anesthesia in a patient with an unsecured airway. It's much more difficult to perform a well-balanced, safe MAC than general anesthesia.

Question: You recommend against relying solely on pulse oximetry to monitor respiratory status during MAC. How do you use capnography and the precordial stethoscope?

Answer: I use capnography for every MAC patient who receives any sedation. The stethoscope is useful for patients who receive heavier levels of sedation, because capnography doesn't give any sense of the adequacy or volume of respiration. With capnography, you can see a perfectly shaped waveform at the normal respiratory rate yet still have a patient who's not ventilating adequately.

Question: You say that pre-op education and intra-op communication with the patient are instrumental in ensuring patient safety. How so?

Answer: Pre-op patient education is our most potent pre-medication because it helps ensure reasonable patient expectations. When patients are prepared to be sedated but aware they'll feel pushing, pulling and tugging during surgery, their anxiety levels decline. They typically don't request deeper sedation during surgery. Communicating with patients intraoperatively diverts their attention from the stress of the operation and soothes them. The anesthesia practitioner can evaluate comfort and sedation levels and better titrate the drug dosage as desired. Pre-op surgeon education is important, too, so we can meet his needs without risking the patient. If the surgeon requires an immobile, unconscious patient, we should perform general anesthesia.

Question: How can the facility manager help ensure patient safety during MAC cases?

Answer: No matter what type of case it is, emergency airway equipment must always be readily available. Support personnel can also help educate and soothe patients before surgery. They can also work with patients and the anesthesia team to determine which patients are good candidates for MAC. I've seen firsthand how this kind of support can virtually eliminate same-day cancellations.

Office-based Anesthesia
Rebecca S. Twersky, MD, MPH, is professor of clinical anesthesiology at the State University of New York Health Science Center in New York, N.Y., and past chair of the American Society of Anesthes-iologists' Committee on Ambulatory Surgical Care and Task Force on Office-based Anesthesia. She has written extensively on the topic.

Question: Is surgery and/or the administration of anesthesia less safe in the office setting?

Answer: There are very limited data comparing mortality rates for the same procedures and same anesthetic approaches between settings. Nevertheless, there's no question that some dramatic and even shocking mortalities have resulted from surgeries and anesthesia performed in the office setting. I think it's clear that the risks are greater when inappropriate procedures are performed in office settings that lack proper equipment, qualified staff or solid policies and procedures. That said, well-qualified anesthesia practitioners can manage patients in nontraditional settings without compromising safety when the setting meets professional standards.

Question: What procedures are and aren't appropriate for the office setting?

Answer: Appropriate office procedures include cosmetic surgeries (such as rhytidectomies, blephs, breast augmentation reduction and rhinoplasty), liposuction, laser surgery, urologic procedures, orthopedic procedures, endoscopies, colonoscopies, pregnancy terminations, dental procedures and micro-laparoscopies. Inappropriate procedures have the potential for significant intra-op blood loss; involve the intra-abdominal, intra-thoracic or intra-cranial cavity; have the ability to cause significant fluid shifts or hypothermia; or last longer than six hours.

Question: A 2001 study suggested general anesthesia is the culprit of some office surgery deaths. Is this true?

Answer: No. That study analyzed reportable surgical incidents and deaths in Florida in 2000, when dermatologists and cosmetic surgeons were arguing about who should do liposuction. General anesthesia was a surrogate for other issues. Although the six deaths involved general anesthesia, just two were directly related to it: one acute bronchospasm and the other unexplained bradycardia and asystole. You must ensure qualified practitioners, the right equipment and proper emergency preparedness.

Question: What can anesthesia practitioners do to ensure safety in the office setting and meet these standards you mention?

Answer: Often, the office environment is structured to accommodate the needs of the primary physician, not the anesthesia practitioner. Resources taken for granted in the OR may be unavailable in the office ' like a central oxygen source, wall suction, overhead lighting, gas evacuation systems, sufficient electrical outlets and even sometimes the anesthesia workstation. The layout is important, too, because the practitioner needs adequate space to work. I've created a checklist that can help ensure an adequate facility and equipment.

Top Anesthesia Safety Concerns
Robert K. Stoelting, MD, is the president of the Anesthesia Patient Safety Foundation. He served as a clinical fellow at the National Institutes of Health and as chairman of the depart-ment of anesthesia at Indiana University School of Medicine. He has authored and edited several anesthesia textbooks.

Question: What are the most important safety issues in anesthesia today?

Answer: Seven years ago, we surveyed 801 practicing anesthesiologists and found that difficult airway management was the most pressing concern. Production pressures ranked second and administration of anesthesia outside the OR ranked third. Other concerns in the top 10 list included stress, fatigue, medication errors and inadequate time available for pre-op evaluation. These top 10 concerns are no different today than they were back then.

Question: What does this tell you about the success, or lack thereof, in addressing these concerns?

Answer: It tells me that these are very real issues with no immediate or ready solutions and that we need to continue to work to resolve them.

Question: Production pressures seem to underlie many anesthesia safety concerns. What can be done about this, given that 'throughput' is paramount to ensuring economic viability in the ambulatory surgery environment?

Answer: Production pressure is a real problem in the OR, where signals to produce are often inherently stronger than signals about safety. The Anesthesia Patient Safety Foundation has started an initiative to ensure that these pressures don't overcome real safety concerns. It's known as the high reliability organization (HRO) model, which other high-risk, high-stress, high-tempo industries, such as aviation, use. We're working to apply this model to anesthesia practice and the entire perioperative process.

HROs have established mechanisms to ensure safety: formal missions, rules, checklists, milestones and a system whereby even the most junior person feels empowered and obligated to halt production for a credible safety threat. HROs flatten the hierarchy when it comes to responsibility for patient safety and ensures that time is available for training.

Intense training is another key to HROs, and we believe this can help us improve upon the current continuing medical education approach for physicians, which is a fairly weak and haphazard system that leaves too much to the discretion of individuals. HROs focus on simulation-based training for individuals, crews, and teams.

The HRO model can also help strengthen our system of detecting, assessing and responding to incidents and adverse events. In healthcare, our systems have been local and isolated and, thus, don't let us share lessons learned. They typically focus on individuals rather than systems, and they don't typically yield practical and effective changes in work practices. There's been too much attention to reporting incidents and accidents, and too little attention to the organizational learning and changes in work practices that should come from the analysis of these reports.

Finally, HRO also rewards personnel for honesty, offers positive sanctions and incentives, reimburses based on outcomes (not just production) and promotes standardization of practices. We are hoping our efforts will truly help improve anesthesia safety.