>

MEETINGS / EVENTS

RSS
>

ASA FEATURED PRODUCT

Add this product to your shopping cart

Anesthesiology Continuing Education (ACE) Program

SKU: 30702-14CE

... Read more »

Single Copies, Member Price: $360

ASA Comments to the FDA regarding SEDASYS

May 14, 2009

Mr. Neel Patel

Center for Devices and Radiological Health (HFZ-480)

Food and Drug Administration

9200 Corporate Blvd.

Rockville, MD 20850

Re: Application for Premarket Approval – SEDASYS TM System by Ethicon Endo- Surgery, Inc. Docket No FDA-2009-N-0664; Meeting ID 2009-4438 before FDA’s Anesthesiology and Respiratory Therapy Devices Panel of the Medical Devices Advisory Committee

Dear Mr. Patel:

On behalf of the American Society of Anesthesiologists (ASA) and its over 43,000 members, whose primary responsibility is the safety of patients under sedation or anesthesia, thank you for the opportunity to comment on the premarket approval application for the SEDASYS TM Computer-Assisted Personalized Sedation System sponsored by Ethicon Endo-Surgery, Inc. Submitted for premarket approval on March 27, 2008, the SEDASYS device is indicated “for the intravenous administration of 1% (10 milligrams/milliliters) propofol injectable emulsion for the initiation and maintenance of minimal to moderate sedation, as identified by the American Society of Anesthesiologists Continuum of Depth of Sedation, in adult patients (American Society of Anesthesiologists physical status I or II) undergoing colonoscopy and esophagoduodenoscopy procedures.”

According to the manufacturer, the SEDASYS TM device has been designed to use patient sensors that measure blood pressure, EKG, hemoglobin oxygen saturation (SpO 2 ), exhaled carbon dioxide levels (ETCO 2 ), and response to verbal commands to control, via computer, the rate of propofol and oxygen administration to patients requiring moderate or “conscious” sedation. While we appreciate the attempts of the manufacturer to incorporate patient respiratory and cardiovascular monitors into the device, we still have several serious concerns regarding the proposed use of SEDASYS TM , including:

1. The lack of data to demonstrate the level of safety required of a complex device such as this, which monitors critical life support functions and determines the delivery rate of a potent intravenous anesthetic;

2. The limitations of oxygen saturation (SpO 2 ) and end tidal CO 2 (ETCO 2 ) measurements when used for the purpose of detecting respiratory depression in sedated patients;

3. The inability of SEDASYS TM to prevent or manage loss of consciousness;

4. The properties, inherent risks and current labeling of the anesthetic drug propofol itself; and

5. Safety requirements for closed loop control devices such as SEDASYS TM .

I. Lack of safety data

Our first concern lies with the limited number of published clinical trials that form the basis of the device’s safety record. We know of only three published abstracts and one published paper evaluating the SEDASYS TM device. All of these articles have been published by one research group and, with the exception of one abstract, all come from a single institution in the United States and one in Belgium (1,2,3,4). More important, each of these trials report that the complications of hypoxemia, apnea, and over-sedation occurred in a significant number of patients. While no “physician intervention” was reported to be required for SEDASYS TM patients in these trials, the studies do not define what constituted a “physician intervention” or the threshold for such an intervention. For instance, would interventions provided by a nurse, such as lifting a patient’s jaw, inserting an oral airway or stimulating the patient, qualify as a “physician intervention?” Further, the studies that recorded apneic episodes lasting longer than 30 seconds reported an incidence of 36% (1,2,3). Similarly, the studies that recorded hypoxemia, defined as an oxygen saturation less than 90%, found hypoxemia in 5.8% of patients (1,2,3), a rate we believe is a high for any anesthetic treatment.

The largest study to date, available only in abstract form (4), reported neither apnea nor hypoxemia independently, but rather reported a composite measure known as the “area under the curve,” reflecting the overall duration and severity of hypoxemia for each patient. The significance of the findings regarding “area under the curve” was not discussed or defined in this preliminary abstract, except to say that the “area under the curve” was lower for patients treated with SEDASYS TM than for patients sedated with midazolam combined with fentanyl or meperidine. The authors of this study did not provide reasons for measuring the “area under the curve” rather than the much simpler incidence of hypoxemia. It is apparent, though, that hypoxemia, no matter how it is described, occurred in some patients treated with SEDASYS TM in this series. Further, patients in this series were receiving supplemental oxygen and, as we will discuss later, SpO 2 can be a very late and insensitive indicator of hypoventilation in patients receiving oxygen.

In summary, hypoxemia, along with apnea and hypoventilation, did occur in a significant percentage of patients in each of the four studies published to support the safety of the device. Moderate sedation (the indication for which SEDASYS TM is being considered) is defined as “a drug-induced depression of consciousness during which patients respond purposefully to verbal commands, either alone or accompanied by light tactile stimulation.” However, some patients in each series, and as high as 58% in one series, (1) became unresponsive for at least some time during the procedure, and thus progressed to the states of deep sedation or general anesthesia. Therefore, not only the small number of studies, but the high incidence of apnea and hypoxemia in the existing studies do not adequately support the ability of the device to prevent unconsciousness and cardiorespiratory depression.

II. Limitations of the oxygen saturation (SpO 2 ) and end tidal CO 2 (ETCO 2 ) measurements when used for the purpose of detecting respiratory depression in sedated patients

Propofol, narcotics and other sedative drugs all affect patients’ breathing in the same way – by reducing the drive to breathe, causing a decrease in the rate and depth of respiration, or hypoventilation. Hypoventilation leads in turn to the accumulation of carbon dioxide and acid in the body, until they eventually rise to levels high enough to depress the heart and brain. The elevation in carbon dioxide, if not detected and readily reversed by ventilatory support, further increases sedation and causes a downward spiral that can ultimately lead to unconsciousness, and in extreme cases, death.

The SEDASYS TM system does indeed measure end-tidal carbon dioxide (ETCO 2 ), an established technique to detect

hypoventilation. However, SEDASYS TM uses open nasal cannulae, rather than a ASA FDA ARTD Comments May 14, 2009

3

closed connection with the lungs, such as an endotracheal tube, as the means to collect the exhaled gas from which carbon dioxide concentration is measured. A problem with the open nasal cannula technique is that it allows room air to mix with gas exhaled by the patient, often resulting in a falsely low carbon dioxide reading. Further, the accuracy of ETCO 2 in the detection of hypoventilation depends on the position of the cannulae, the structure of the patient’s nose and mouth, and the patient’s individual respiratory pattern (i.e. some patients are primarily nose breathers, some are primarily mouth breathers and some are a combination of both). With this technique, the ability of a computerized ETCO 2 monitor to detect hypoventilation is limited by its capacity to only detect slowing or complete cessation of respiration, which are late signs of respiratory depression. Earlier detection depends on a dedicated and experienced health care provider, who understands the limitations of the monitors and who is trained to recognize the patient variables that contribute to respiratory depression and hypoventilation.

The second respiratory monitor used by SEDASYS TM is the pulse oximeter, although it does not measure ventilation or breathing directly. Instead, it measures one of the changes that can accompany hypoventilation which is a reduction of oxygen that is brought into the lung. Reduced ventilation, caused by the patient being sedated, brings insufficient oxygen into the lungs to supply the blood, especially in patients with any pre-existing compromise in lung function. When patients are breathing room air, hypoventilation causes a rapid drop in the SpO 2 , which can be measured by the pulse oximeter. This early drop in SpO 2 , in patients breathing room air, can alert physicians quickly to hypoventilation. For example, Vargo et al. (5) found that 50% of the episodes of apnea or disordered breathing were detected by SpO 2 monitoring during upper endoscopic procedures. In their protocol, room air was used until the oxygen saturation fell, to maximize the chances for detection of the hypoventilation.

However, when a patient is receiving supplemental oxygen, as in those treated with SEDASYS TM , even reduced ventilation can bring enough of the more concentrated oxygen into the lungs to prevent a decrease in blood oxygen concentration or SpO 2 . Fu et al. (6) demonstrated that the use of supplemental oxygen removes the ability of pulse oximetry to detect hypoventilation. After breathing at half the normal rate of ventilation, fifty percent of Fu’s patients who breathed room air had an oxygen saturation of less than 90%. However, almost all of Fu’s patients who received supplemental oxygen, even at inspired oxygen concentrations as low as 25-30%, maintained oxygen saturations greater than 90% after ten minutes of hypoventilation. Furthermore, the combination of hypoventilation and supplemental oxygen can cause the level of carbon dioxide and acid in the blood to rise to dangerous levels without a change in the oximeter reading. Davidson (7) reported profound hypoventilation with an arterial carbon dioxide level (PaCO 2 ) rising to 281 mmHg (normal of 40 mmHg) despite an oxygen saturation of 95% as measured by pulse oximetry in a sedated, spontaneously breathing patient receiving oxygen. It is clear that the monitors used by the SEDASYS TM system, such as pulse oximeter and ETCO 2 , have limitations and must be interpreted in context by experienced individuals with adequate training.

III. The inability of SEDASYS TM to prevent or manage loss of consciousness

What would appear to be the SEDASYS TM device’s most reliable monitor is the patient’s ability to purposefully respond to the device’s verbal or tactile commands by pushing a button. However, this monitor is limited to the device’s proposed use – the administration of minimal and moderate sedation, during which patients can reliably and purposefully respond to verbal stimulation. In current practice in the United States, many patients undergoing endoscopic procedures require, or at least desire, deeper levels of sedation than SEDASYS TM is designed to administer. It is not always possible to predict this need for deeper sedation before the procedure begins. Dr. Zuccaro describes this in his editorial published in Gastrointestinal Endoscopy in 2006: “there is no question that many patients undergoing endoscopy, particularly for long or complex procedures, are deeply sedated at some point during the procedure. This is ASA FDA ARTD Comments May 14, 2009

4 in part driven by the patient, who often expects or demands a painless experience. Because sedation is a continuum, increasing the frequency of deep sedation logically increases the frequency of inadequate ventilation or airway obstruction.” (8)

Further, propofol provides excellent sedation but is poor at relieving anxiety or pain. Withdrawal responses during colonoscopy and cough during EGD, in the opinion of Rex and VanNatta (9,10), makes moderate sedation, using propofol alone, inadequate for both the patient and the endoscopist. These authors (10) conclude that propofol combined with other agents, can be used effectively to provide moderate sedation, but that “propofol as a single agent is ineffective for targeting moderate sedation for endoscopic procedures.”

The limited anxiolytic and analgesic effects of propofol explain why many of the studies evaluating SEDASYS TM (1,2,3,4), and the use of propofol sedation during endoscopy without SEDASYS TM (10,11,12), include administration of narcotics or benzodiazepines. However, both the sedative effects and the respiratory depressant effects of these agents are additive with propofol. Therefore, co-administration of other agents, especially during the propofol infusion itself, can easily render ineffective the limitations placed by SEDASYS TM on the administration of propofol. In daily clinical practice, SEDASYS TM will frequently be used to administer propofol along with the simultaneous administration of other sedatives and analgesics by practitioners to provide the desired relief of anxiety and pain. This may explain why the four manufacturer-sponsored studies of SEDASYS TM found some patients in each trial whose sedation was deeper than intended, and in one study (1) found that 58% of patients became unresponsive.

IV. Properties of the anesthetic drug propofol

In an editorial in Gastrointestinal Endoscopy, Dr. Iravani, a UCLA anesthesiologist, states succinctly; “one of propofol’s attractive features, the quick onset of sedation, could be its downfall: The depth of sedation could change rapidly from moderate sedation to deep sedation or general anesthesia and result in respiratory depression and/or airway obstruction.” (13) As ASA indicated in recent comments to the FDA regarding fospropofol, a pro-drug of propofol (see the ASA comments dated April 23, 2008 regarding NDA22-244 fospropofol disodium injection), “propofol is a potent general anesthetic with a very steep dose-response curve.”

The FDA has already recognized, as part of the product labeling, that propofol and more recently fospropofol

should be administered only by persons trained in the administration of general anesthesia and not involved in the conduct of the surgical/diagnostic procedure. Patients should be continuously monitored, and facilities for maintenance of a patent airway, artificial ventilation, oxygen enrichment, and circulatory resuscitation must be immediately available.”

The FDA has maintained this safety standard despite requests from groups advocating for administration of propofol by nurses and physicians without such specialized training. However, recent articles demonstrate that the SEDASYS TM system is being proposed by some as a way to circumvent these established FDA labeling requirements. According to Dr. Pambianco and colleagues, the SEDASYS TM system is “intended to provide endoscopist/nurse teams an “on label” method to administer propofol sedation for colonoscopy and esophagogastroduodenoscopy.” (4) This position is being advanced even though the incidence of deep sedation and hypoxemia reported with SEDASYS TM is no lower, and sometimes higher, than that

5 reported in the past with propofol administered directly by registered nurses and endoscopists (10, 11) without the use of the device. Published studies demonstrate that the device offers little, if any, additional safety benefit, and no reason for the FDA to change its well-established labeling requirements for propofol with or without the use of SEDASYS TM .

Because of the variability of individual patient response to a given drug, especially a potent one, no electronic or mechanical device used to administer propofol can be expected to prevent adverse reactions entirely, and no device can substitute for the important requirement of human oversight. All the published trials of the SEDASYS TM system have reported the occurrence of unintended deep sedation. Under deep sedation or general anesthesia, patients may face several life threatening complications including airway occlusion, apnea, hypoxia, and cardiovascular collapse. Individuals not trained and experienced in the administration of general anesthesia may not be able to restore breathing or normal cardiac activity in time to prevent patient injury. This scenario is even more concerning if the physician responsible for detecting and treating complications of sedation has their attention directed to performing the procedure.

There are no reversal agents available for propofol, should too much of the drug be given or other complications arise. Therefore, the need remains for continuous patient monitoring by a trained and qualified individual without other responsibilities, who can maintain the patient’s ventilation and circulation until the drug wears off. We would expect that, as an essential part of a clinical study, such monitoring would be present in the clinical trials that are being submitted to establish the safety of the SEDASYS TM system. We are concerned whether, in the non-controlled clinical setting, patients will continue to receive an equivalent level of medical attention while the SEDASYS TM system is in use.

V. Safety requirements for closed loop controller devices such as SEDASYS TM

The SEDASYS TM system falls into the class of devices known as Physiologic Closed Loop Controllers (PCLC). The International Electrotechnical Commission/International Standard Organization is drafting comprehensive standards regarding the safe use of PCLC’s for medical devices (IEC 60601-1-10 Ed1). Therefore, should the SEDASYS TM system be approved, we assume and hope that it will be required to comply with these standards, once published. Specifically, these draft standards include requirements for the parameters such as the SpO 2 , ETCO 2 , EKG, and propofol infusion value to be displayed by the device, how the device responds to inoperable components (such as the pulse oximeter, capnometer, EKG, or noninvasive blood pressure monitor in the case of the SEDASYS TM system) and information to be included in the PCLC instructions for use.

Gillham et al. (14) published a peer-reviewed study of a similar type of closed loop infusion controller, using patient responses to voice commands to determine the appropriate propofol drug level. They studied 20 patients undergoing endoscopic retrograde cannulation of the pancreatic duct (ERCP), a somewhat more invasive procedure than colonoscopy. In these patients, they noted the need for manual override of the device in 4 of their patients, because of inadequate sedation in three patients and confusion in 1 patient. Four additional patients were deemed oversedated during the procedure. Therefore, 40% of their patients either were oversedated, or required manual override of the system for insufficient sedation.

The monitoring aspects of the SEDASYS TM could improve patient care currently available in some settings by using pulse oximetry, end tidal CO 2 , and patient responsiveness monitoring to alert physicians and nurses to needed changes in sedative dosing during endoscopic procedures. However, the computer driven, mechanized drug delivery system can never anticipate the varied levels of stimulation, and can only react “after the fact”; once the patient becomes hypoxic, apneic or unresponsive. And of course, the SEDASYS TM system cannot rescue the patient from deeper levels of sedation. Physicians and nurses with ASA FDA

6

high levels of training and experience in the treatment of patients before, during, and after sedative administration will continue to be required to perform at least the following functions during endoscopy:

1. Patient selection

A. The identification of patients who will need more sedation than available with the minimal or moderate sedation provided by SEDASYS TM ;

B. Identification of patients with increased risk of airway obstruction;

C. Identification of patients with increased risk of pulmonary aspiration; and

D. Identification of patients with increased risk of respiratory or cardiovascular depression.

2. Continuous patient monitoring

A. The current labeling of propofol includes the precaution that the individual who administers the drug should “not (be) involved in the conduct of the surgical/diagnostic procedure.” This reflects the well-established principle that there must be an independent practitioner whose sole responsibility is administering propofol and monitoring the patient to assess the level of consciousness and to identify early signs of hypertension, bradycardia, hypoventilation, apnea, airway obstruction, and/or oxygen desaturation.

B. Organizations that accredit hospitals, ambulatory health facilities and office practices, such as the Joint Commission, the Accreditation Association for Ambulatory Health Care and the American Association for the Accreditation of Ambulatory Surgical Facilities, require the immediate availability of a staff member with documented expertise in airway management and advanced cardiopulmonary resuscitation, as well as immediate access to emergency equipment, during procedures performed with propofol sedation.

C. The presence of an additional practitioner who devotes his or her full attention to monitoring the patient’s response to sedative drugs and who is able to accept complete responsibility for sedation provides additional patient benefit because it allows the practitioner conducting the surgical or diagnostic procedure to devote his or her full attention to the procedure.

3. Anticipation of the level of stimulation

The level of stimulation associated with each part of the surgical procedure must be assessed before the stimulus occurs, so that the level of the anesthetic or sedative drug can be increased in time to meet the anticipated needs of the surgical procedure or decreased to avoid over-sedation during periods of little stimulation. SEDASYS TM may be able to respond to change in patient stimulation, but it cannot anticipate these changes.

4. Rescue and the treatment of complications

In order to act quickly and effectively to prevent and treat any complications involving the respiratory or cardiovascular systems, personnel must be immediately available who have expertise in the assessment of respiration and circulation disturbances and have documented training and current experience in the management of these disorders.

In summary, the use of potent, rapidly acting general anesthetics, such as propofol and now fospropofol, continues to require patient care by personnel trained in the management of deep sedation and general anesthesia, and not involved in the surgical or diagnostic procedure itself. This remains true even if the intent is to provide minimal to moderate sedation, which we know from the literature may not be sufficient for many patients undergoing endoscopy procedures. A computer-assisted closed loop controller, such as SEDASYS TM , cannot substitute adequately for the essential human functions. Therefore, in order to maintain an acceptable level of safety, pre-market approval for the SEDASYS TM system should include:

  1. The requirement for trained, certified personnel dedicated to continuous monitoring of the patient and the anesthetic administration, identical to that currently in effect for the administration of propofol and fospropofol, be included in the pre-market approval;

  2. Displays on the device to provide to the operator continuous output information from all patient monitors, along with the EKG, capnogram, and pulse oximeter waveforms, as well as propofol and oxygen infusion rates;

  3. The activation of audible and visual notifications when the propofol infusion rate is automatically changed by the device;

  4. The activation of audible and visual alarms when any of the SEDASYS TM monitors ceases normal function, loses its patient signal, or senses a pre-determined “alarm condition;”

  5. Warning labels regarding the potential harmful effects of sedative and analgesics if administered simultaneously with the SEDASYS TM infusion; and
  6. More extensive pre- and post-market surveillance to ascertain the safety of the device in the clinical setting.

Thank you for your thoughtful consideration of our comments. If you have any questions or require additional information, please do not hesitate to contact Chip Amoe, J.D., M.P.A, Assistant Director – Federal Affairs, in our ASA Washington Office, (202) 289-2222.

Sincerely,

Roger A. Moore, M.D.

President

American Society of Anesthesiologists

REFERENCES

Pambianco DJ, McRorie J , Martin J, Poltilove R, Whitten CJ: Feasibility assessment of computer assisted personalized sedation: a sedation delivery system to administer propofol for gastrointestinal endoscopy (abstr) Gastrointestinal Endoscopy. 2006; 63:AB189.

2. Moerman A, Pambianco DJ, McRorie J, Martin J, Struys M: Feasibility Assessment of a sedation delivery system to administer propofol for GI endosopy. Anesthesiology 2006; 105:A1586

3. Pambianco DJ, Whitten CJ, Moerman A, Struys MM, Martin JF: An Assessment of Computer-Assisted Personalized Sedation: a sedation delivery system to administer propofol for gastrointestinal endoscopy. Gastrointestinal Endoscopy 2008; 68:542-547.

4. Pambianco DJ, Pruitt RE, Hardi R, Weinstein ML, Bray WC, Kodali VP, Vargo JJ, Schubert T: A computer-assisted personalized sedation system to administer propofol versus standard-of-care sedation for colonoscopy and esophagogastroduodenoscopy: a 1,000 subject randomized, controlled, multicenter, pivotal trial. Gastroenterology 2008; 135:294.

Vargo JJ, Zuccaro G, Dumot JA, Conwell DL, Morrow JB, Shay SS: Automated graphic assessment of respiratory activity is superior to pulse oximetry and visual assessment of the detection of early respiratory depression during therapeutic upper endoscopy. Gastrointestinal Endoscopy 2002; 55:826-831

Fu ES, Downs JB, Schweiger JW, Miguel RV, Smith RA: Supplemental Oxygen Impairs detection of hypoventilation by pulse oximetry. Chest 2004; 126:1552-1558 Davidson JAH, Hosie HE: Limitations of pulse oximetry: respiratory insufficiency – a failure of detection. British Medical Journal 1993; 307:372-373

Zuccaro G: Sedation and analgesia for GI endoscopy. Gastrointestinal Endoscopy 2006; 63:95-96.

Rex DK: Review article: moderate sedation for endoscopy: sedation regimens for non-anesthesiologists. Alimentary Pharmacology and Therapeutics 2006; 24:163-171.

VanNatta ME, Rex, DK: Propofol alone titrated to deep sedation versus propofol in combination with opioids and/or benzodiazepines and titrated to moderate sedation for colonoscopy. Am J Gastroenterology 2006; 101:2209-2217

Cohen LB, Hightower CD, Wood DA, Miller KM, Aisenberg J: Moderate level sedation during endoscopy: a prospective study using low-dose propofol, meperidine/fentanyl, and midazolam. 2004; Gastrointestinal Endoscopy 58:795-803.

McQuaid KR, Laine L: A systematic review and meta-analysis of randomized, controlled trials of moderate sedation for routine endoscopic procedures. Gastrointestinal Endoscopy 2008; 67: 910-923.

Iravani M: On computers, nurses, and propofol: further evidence for the jury. Gastrointestinal Endoscopy 2008; 68: 510-512

Gillham MJ, Hutchinson RC, Carter R, Kenny GNC: Patient-maintained sedation for ERCP with a target-controlled infusion of propofol: a pilot study. Gastrointestinal Endoscopy 2001; 54:14-17.