Dept Of Anesthesiology & Critical Care, University Of Pennsylvania School Of Medicine

Poorly maintained anesthesia machine associated with awareness under anesthesia

Four hospitals in California were recently fined $25,000 each for serious mistakes resulting in patient harm or death.  Among the incidents was the use of a malfunctioning anesthesia machine which resulted in three patients being partially awake during their surgery.   With respect to the anesthesia machine the California Health and Human services Statement of Deficiencies notes that on 3/28/08 an Anesthesiologist expressed some question about the functionality of the anesthesia machine during a surgical procedure.  The same anesthesia machine was used again on 3/31/08 for three more surgical procedures.”  Source: San Diego Union Tribune, August 16, 2008.  In at least one of the three cases of awareness there were hemodynamic signs suggesting that too little anesthetic drug was being given.  None of the articles commented on the availability of or use of agent analyzers.  Also unclear is why it took 3 cases of awareness on the same day in the same OR before the machine was taken out of service.  It is cases such as these that fuel the national call by some groups and individuals for mandatory use of brain electrical activity monitors during general anesthesia.  As I have noted in past posts, there is considerable difficulty in interpreting and abstracting newspaper articles that deal with legal or potental legal issues.  I may have erred in some of my interpretation.

 

David S. Smith, M.D., Ph.D. 

Inhalation anesthetics produce damage by disrupting intracellular calcium homeostasis

Huafeng Wei continues his investigations into inhaled agent toxicity.  Using chicken lymphocytes he demonstrated that isoflurane, desflurane and sevoflurane at 2 MAC for 24 hours produced cell damage by increasing cytosolic calcium.  The calcium came from the endoplasmic reticulum.  In IP3 deficient lymphocytes this damage calcium translocation and cell damage did not occur (IP3 is a calcium channel located on the endoplasmic reticulum).  Isoflurane had a much greater effect than did desflurane or sevoflurane.  The clinical significance of these findings are as yet not clear.  The doses and duration of exposure was high for all three agents.  However from a mechanistic viewpoint these findings add to our understanding of the effect of inhaled anesthetics on intracellular calcium control (Yang H et al: Inhalational anesthetics induce cell damage by disruption of intracellular calcium homeostasis with different potencies.  Anesthesiology 2008;109: 1 – 8).

 

David S. Smith, M.D., Ph.D.

Cost of potentially preventable complications in surgical patients

The cumulative additional cost of potentially preventable complications in surgical patients may be as high as 1.5 billion dollars a year.  Encinosa and Hellinger (Health Services Research, early view July 2008, DOI: 10.1111/j.1475-6773.2008.00882.x) examined 5.6 million enrollees in the 2001- 2002 MarketScan Commercial Claims and Encounters Database (a data base of medical care claims data for 45 very large employer-sponsored benefit plans) and identified 161,004 patients with a major surgery admission who did not have another admission for major surgery within the prior 90 days.  The authors examined 14 patient safety indicators (PSI) which were grouped into technical problems (anesthesia complications, accidental lacerations, foreign body left in, transfusion reaction and similar events), infections (related to medical care, sepsis), pulmonary and vascular problems (pulmonary embolism and DVT), acute respiratory failure, metabolic problems, wound problems (such as hematoma or dehiscence) and nursing sensitive events (post op hip fracture, decubitus ulcers for example).  They identified 4,140 pts with an identified PSI and matched them to patients without an identified PSI.  They found that 2.6% of the patients who underwent major surgery had at least one of the 14 PSIs.  About 5.6% of this group had more than one PSI.  Patients with PSIs had a 90 day death rate of 6.3% compared to 0.6 in those without a PSI.  The 90 day readmission rate was 15% compared to 5.5%.  The 90 day cost for surgery was $66,879 on the average in those with PSIs compared to $18,284 for those without.  Post operative acute respiratory failure was the most expensive of the PSI groups costing $106,370 over the 90 day period and it had the highest 90 day death rate (12%).  Infection however had the highest readmission rate.  This is one of the first papers to determine complication rates and their costs beyond the period of hospitalization.  Costs from PSI continue after discharge.  The database represents a non medicare population and includes 4% of the commercially insured individuals in the United States.  The 14 PSIs examined account for 11% of the 90 days deaths, 2% of 90 day readmissions, and 2% of all 90 day expenditures after major surgery.  Decreasing the incidence of these PSIs could greatly decrease medical costs and improve the “bottom” line of acute care institutions.

 

David S. Smith, M.D., Ph.D.

Dead after knee surgery, a story of apparent multiple little failures

A 76 year old male entered a hospital for elective knee surgery.  Roughly twenty four hours after the surgery he was dead.  If the article, which appeared in the News and Observer, Raleigh, N.C. (June 8, 2008) , is at all accurate, the patient died because his doctors and nurses did not communicate with each other and no one was willing to say “stop.”  The article describes a series of “little” failures the cumulative effect of which was a fatality.  Note the following:  1) the apparent failure of those in the pre operative area to investigate what appeared to be a marked change in the patient’s medical condition, 2) an anesthesiologist who initiated treatment for angina but who did not appear to have communicated either orally or in writing about his observations or therapy (no progress note), 3) there appeared to be no one who willing to “stop” the operation for additional clincical investigation, 4) surgeons who appeared to deny any responsibility for knowing the patient’s medical condition just prior to surgery, 5) there were apparently multiple failures at many levels to document clinical findings and therapy, 6) an apparent failure to inform those responsible for the patient’s care post operatively about the suspected CHF and angina, and 7) delays in obtaining laboratory results.  Whatever time was “saved” by staying on schedule was lost many times over in the time required to be present at the multiple investigations and answer many hours of questioning.  Does the problem of communication and documentation sound familiar? (link to article: http://www.newsobserver.com/news/story/1100507.html )

 

David S. Smith, M.D., Ph.D.

No payment for events deemed avoidable

In the push to force “medicine” to reduce or eliminate errors Medicare, some private insurance companies and many states are creating lists of events for which payment will not be made.  The lists of events for which payment will be denied varies between payers.  Here is the list recently published by the State of New York for Medicaid payments:

1. Surgery performed on the wrong body part;   
2. Surgery performed on the wrong patient;   
3. Wrong surgical procedure on a patient;   
4. Foreign object inadvertently left in patient after  surgery;   
5. Medication error;   
6. Air embolism;   
7. Blood incompatibility;   
8. Patient disability from electric shock;   
9. Patient disability from use of contaminated drugs;   
10. Patient disability from wrong function of a device;   
11. Incidents whereby a line designated for oxygen  intended for patient is wrong item or contaminated;   
12. Patient disability from burns;   
13. Patient disability from use of restraints or  bedrails; and   
14. Patient disability from failure to identify and treat hyperbilirubinemia (bilirubin in blood) in newborns.

David S. Smith, M.D., Ph.D.

Bad news for tight intraoperative glucose control

Dr. Kofke calls our attention to this prospective randomized study of tight intraoperative glucose control using insulin infusion compared to “conventional” management with tight control in both groups during the post operative period.  Four hundred cardiac surgery patients were randomly assigned to tight glycemic control (blood glucose levels between 80 – 100 mg/dl) during surgery or conventional glucose control.    Patients who did not become hyperglycemic during surgery were not included in the analysis.  Pre operatively the glucose levels were similar in both groups.  At the conclusion of cardiopulmonary bypass the mean blood glucose level in the tight control group was significantly lower (123 mg/dl) compared to 148 mg/dl in the conventional glucose control group.  All patients in the intensive treatment group received insulin during surgery and 15% of the patients in the conventional therapy group received insulin.  At the end of 24 hours in the ICU the mean glucose levels were the same in both groups (about 106 mg/dl, mean).  The two groups did not differ in the primary composite endpoint of sternal infection, death, prolonged ventilation, cardiac arrhythmias, stroke or renal failure.  Nor did investigators find a direct benefit from intraoperative from intensive insulin therapy for and of the individual components of the composite end point.  In fact the reverse result was obtained.  The intensive treatment group had significantly more strokes (8 vs. 1) and deaths (4 vs. 0) than the conventional treatment group.  There was no treatment effect for length of stay in the ICU or hospital.  The authors note that this does not directly contradict prior findings of benefits from tight glucose control as the key studies involved tight control vs. conventional control during the entire period from OR through ICU.  It is possible that tight glucose control during the relatively brief period of surgery is not as important as tight glucose control during the longer post operative period.  Gandhi GY et al: Intensive intraoperative insulin therapy versus conventional glucose management during cardiac surgery.  Ann Intern Med 2007;146:233-243

Dr. WA Kofke is Professor of Anesthesiology at UPENN

David Smith, M.D., Ph.D.

What is six sigma? How does it apply to health care?

There are two meanings 1) Sigma ( σ ) is the accepted symbol for 1 standard deviation from the mean of the “normal” distribution (bell curve).  Since 1 sigma would exclude 31.8% of a normally distributed event under consideration and three sigma would exclude 0.3%, six sigma implies events of very low frequency (on the order of 3 per million). 2) Six sigma is also commonly used as the name of a process of quality control originally developed at Motorola Corporation and later adopted by many including General Electric Company (a six sigma error rate is considered to be virtually flawless).  In many industries application of the methodology has reduced defects, improved customer satisfaction, and decreased costs.  With respect to medicine, Parker et al used these techniques to develop a process that improved adherence for antibiotic prophylaxis in patients undergoing cardiac surgery (1); Women and Children’s hospital of Charleston used six sigma to enhance access to care by decreasing time to appointment and waiting time to see a doctor, increase patient satisfaction, and improve revenues in a resident run obstetrics and gynecology clinic (2); and Massachusetts General Hospital avoided the addition of an additional shift for its proton beam facility by changing how it scheduled cases requiring anesthesiologists (3).  Key to the success of this methodology is the fact that it is data driven.  It focuses on process improvement and variation reduction using the paradigm of define, measure, analyze, improve, and control (4).  It can be applied to both existing and planned processes.  We do not like to consider our practice of medicine to be the same as manufacturing super soaker recreational water guns.  However for success both require the correct application of a sequence of steps.  In the case of medicine  failure of any one of those steps may lead to patient injury, patient or family dissatisfaction or increased costs in the form of a cancelled case, increased hospital stay, etc.

1)      Parker BM et al: Six sigma methodology can be used to improve adherence for antibiotic prophylaxis in patients undergoing cardiac surgery.  Anesth Analg 2997:104:140-6

2)      Calhoun BC: Six-sigma inspired workflow redesign enhances access to care and increases patient satisfaction, visits, and revenues in obstetrics and gynecology residency clinic.  (http://www.innovations.ahrq.gov/content.aspx?id=1868 )

3)      Krasner J: New medicine for what ails hospitals.  Boston Globe January 28, 2008 (http://www.boston.com/business/healthcare/articles/2008/01/28/new_medicine_for_what_ails_hospitals? )

4)      Sivy J: Six Sigma.  Carnegie Mellon Software Engineering Institute Software Technology Roadmap (http://www.sei.cmu.edu/str/descriptions/sigma6_body.html )

David S. Smith, M.D., Ph.D.

Anoxic Encephalopathy; anesthesiologists still involved

The Pennsylvania Patient Safety Authority summarizes 3 years of submitted reports related to anoxic encephalopathy in Pennsylvania Hospitals (http://www.psa.state.pa.us/psa/lib/psa/advisories/v5n1march_2008/mar_2008_v5_n1.pdf ), pages 34 - 35.  Thirty one reports were identified.  Seventeen of the 31 patients died.  Nineteen appeared to be associated with the perioperative period with 4 occurring in the OR.  The PSA emphasizes that anoxic encephalopathy can be associated with more then just airway and ventilation; three occurred in association with blood loss.  The following causes were noted: BIPAP came off, arrest following sedation on the floor, arrest during OR intubation, arrest on induction of anesthesia, post extubation arrest at the end of operation, arrest following sedation in MRI, difficult reintubation in an ICU, and propofol syndrome, among others.  Eight of the 31 incidents were related, in some way, to airway management.  When one considers the total number of patients cared for in Pennsylvania hospitals during this time period, the incidence is very low, however each one had a devastating effect on the families of these patients and most likely on the caregivers involved.  Dr. Fleisher notes that “while some have argued that we approach six sigma in terms of cardiac arrests and deaths directly due to medical error, we must continue to learn from these events in order to provide the best possible care for our patients.”

David S. Smith, M.D., Ph.D.

Dr. Fleisher is Chair of Anesthesiology and Critical Care, University of Pennsylvania

There is an increased methadone use for pain therapy; there is a marked increase in methadone related deaths. Are they related?

According to one of my pain therapy sources, methadone, which is longer acting and does not cause euphoria compared to other narcotics, is increasingly being used for the treatment of chronic pain.”  However there now appears to be an increase death associated with methadone use.  According to a recent Philadelphia Inquirer article (April 18, 2008), death of people taking methadone is increasing at a very rapid rate.  According to the National Center for Health Statistics the number of methadone deaths across the United States rose from 786 in 1999 to 4,462 in 2005.The Inquirer notes that the drug is easily diverted to the black market.  They state that even though methadone does not produce a “high” it is often combined with other drugs.

            Methadone has been recently associated with cardiac death in patients using this drug.  Chugh et al (1) over a four year period prospectively evaluated all patients who had sudden cardiac death and underwent investigation by the medical examiner in the metropolitan area of Portland.  Case subjects had a therapeutic blood level of methadone and these were compared to patients with no identified methadone.  Patients with recreational drug use or any drug overdose were excluded.  They found a total of 22 sudden cardiac death cases with therapeutic levels of methadone.  The most common indication for methadone use was pain control.  They found that significantly fewer of the patients taking methadone had a structural abnormality that would explain the cardiac death compared to the non methadone group.  They speculated that death in the methadone cases may have been related to an arrhythmia.  Others (2) have suggested that methadone may produce potassium ion channel blockade, prolonged QT interval and the potential for a Torsade de Pointes arrhythmia.

            The UPENN chief of pain medicine provides the following additional information: most experts believe that methadone-related deaths are attributable to 2 main issues.  First, methadone is used by relatively inexperienced clinicians who do not understand proper dosing.  In these patients, either the initial dose is too high, or dose changes are made too frequently, and the patient "over shoots" the proper dose and then experiences opioid-induced respiratory compromise.  Second is the rarer risk of Torsade de Pointes.  This appears to be an adverse event unique to methadone among the potent opioids.  Methadone appears to block the rapid component of the delayed rectifier potassium current in a dose-dependent fashion, and as a result may prolong the QT interval.  This effect is most commonly observed in patients taking high-dose methadone (> 100 mg/day), but has been reported at lower doses.  The good news is that QTc intervals of 500 msec or more are predictive of an increased risk for Torsade de Pointes.  One of our pain doctors obtains an ECG on patients on methadone if their daily dose is > 80 mg.

            With respect to methadone there is a large inter-individual variability and that reaching a steady state can take 7 days or more.

1)      Chugh SS et al: A community based evaluation of sudden death associated with therapeutic levels of methadone.  Am J Med 2008;121:66-71

2)      Maremmani I et al: QTc interval prolongation in patients on long-term methadone maintenance therapy.  Eur Addict Res 2005;11:44-49

David S. Smith, M.D., Ph.D.

An argument for NOT stopping antiplatelet drugs prior to surgery

The Anesthesia Pain Service Guidelines presented earlier call for stopping clopidogrel seven days prior to surgery if neuraxial anesthesia is desired.  It is important to note, however, there have been a number of reported cases of coronary artery occlusion in patients with drug eluting stents after temporary cessation of either aspirin or clopidogrel.  Bengeri (1) and Bolsin et al (2) argue that the risks of cessation outweigh the risk of increased surgical bleeding and that these antiplatelet drugs should be continued into the perioperative period.  Bolsin notes that they surveyed 24 patients with drug eluting stents who presented for a total of 43 non-cardiac surgery procedures.  On 15 occasions clopidogrel was stopped though aspirin was continued.  Three patients experienced myocardial infarction secondary to in-stent thrombosis.  Two of the three infarcts occurred prior to surgery.  Only one patient of the 24 experienced excessive bleeding.

            The American College of Cardiology/American Heart Association joint guidelines for reducing the risk of a perioperative cardiac event (3) emphasized the need to continue anti-platelet drugs perioperatively.  The cases presented by Bengeri and by Bolsin et al together with the recommendations found in the ACC/AHA guidelines force the conclusion that antiplatelet drugs should not be discontinued prior to surgery in patients receiving these drugs after coronary stenting.  The Anesthesia Pain Service notes that epidural analgesia can be safely provided to patients taking aspirin.  In patients on clopidogrel, the drug should be continued preoperatively but epidural analgesia should not be offered.

1)      Bengeri S: Successful management of patients with a drug-eluting coronary stent presenting for elective surgery (letter).  BJA 2007;98:841-8

2)      Bolsin S et al: Comment on Bengeri 2007 (letter): BJA 2007;98:842

3)      Fleisher LA et al: ACC/AHA 2007 guidelines on perioperative cardiovascular evaluation and care for non cardiac surgery. http://circ.ahajournals.org

David S. Smith, M.D., Ph.D.