BIS monitors-- evidence and practice

[sluggs]

Hi all-- IM/CCM here but I get a lot of good information from your threads. Every place I have worked before by current one, we have used BIS monitoring in our patients that we paralyze and prone for ARDS.
Typically we would target BIS of 40 and then paralyze for 24-48 hours. My current place does not have them in the ICU, so we just sedate deeply, assess and then paralyze. Seems like a BIS would be good for longitudinal monitoring. I have been looking at the "evidence" for BIS and it is mostly equivocal small studies. Can anyone point me to better data, or some insight into acceptance in practice?
Much obliged!

---

Members don't see this ad.

 

[Psai]

Bis really helps us titrate meds, prevents awareness (pretty rare), gives us an idea of where the patient is at during tiva because there's no mac value to follow. It makes me more comfortable shutting off infusions at the end because the timing is sometimes difficult for long neuro cases for a timely wakeup. It also helps get a sense of how much of an infusion to use when you're not using bis. I generally don't like using it though outside of cardiac and neuro cases so I just go with what the attending wants.

Here are some studies

Bispectral Index Monitoring Allows Faster Emergence and Improved Recovery from Propofol, Alfentanil, and Nitrous Oxide Anesthesia | Anesthesiology | ASA Publications

http://www.sarasotaanesthesia.com/reading/literature/BIS index Declines durin NMB.pdf

Prevention of Intraoperative Awareness with Explicit Recall in an Unselected Surgical Population: A Randomized Comparative Effectiveness Trial

https://www.sciencedirect.com/science/article/pii/S0007091217310152

 

[Ronin786]

I'm generally not a fan of the BIS outside of the very elderly population where I use the outliers as a means of decreasing the amount of anesthetic required (a practice I actually learned here on SDN).

That said, I think they serve a purpose in the ICU when paralytics are utilized. Many ICU nurses are quite terrible at sedating patients, as evidenced by how frequently they actually manage to meet their RASS goals, or how often they actually check TOF on paralyzed patients. Having the BIS as a simple number they can target is important, because it usually will show some sort of derangement when the patient is starting to stir, before they actually become aware. I.e. I consider it a built in margin of error despite the paltry evidence to suggest it prevents recall. With the caveat that most studies are in the OR population, a population that isn't paralyzed for 24-48 hours and monitored by a nurse that's busy charting and limits their propofol to 83 mcg/kg/min.

The issue with deeply sedating then paralyzing is you're usually not assessing the patient as they're being turned/probed/cleaned/having their ETT juggled by RT. And depending on the inflammatory response associated with some of these critically ill patients (particularly the younger ones), tolerance can rapidly develop to benzodiazepines and opioids. The BIS won't prevent that from happening, but I think it's an extra monitor as long as you build in a margin of error (<40 as opposed to <60).

 

[Urzuz]

Bis really helps us titrate meds, prevents awareness (pretty rare), gives us an idea of where the patient is at during tiva because there's no mac value to follow. It makes me more comfortable shutting off infusions at the end because the timing is sometimes difficult for long neuro cases for a timely wakeup. It also helps get a sense of how much of an infusion to use when you're not using bis. I generally don't like using it though outside of cardiac and neuro cases so I just go with what the attending wants.

May not be helpful to you but here are two of the important studies that I'm aware of.

Bispectral Index Monitoring Allows Faster Emergence and Improved Recovery from Propofol, Alfentanil, and Nitrous Oxide Anesthesia | Anesthesiology | ASA Publications
Prevention of Intraoperative Awareness with Explicit Recall in an Unselected Surgical Population: A Randomized Comparative Effectiveness Trial

These studies aren't at all applicable to the setting he is discussing.

Here are some studies that may be of more use to you sluggs: Comparing BIS monitoring with clinical assessment for determining the level of sedation of mechanically ventilated adults in intensive care units

"We found insufficient evidence about the effects of BIS monitoring for sedation in critically ill mechanically ventilated adults on clinical outcomes or resource utilization. The findings are uncertain due to the low- and very low-quality evidence derived from a limited number of studies."

Bispectral index (BIS) monitoring of ICU patients on continuous infusion of sedatives and paralytics reduces sedative drug utilization and cost

"Bispectral index monitoring of ICU patients on continuous infusions of paralytics and sedatives reduces sedative drug cost as well as the recall phenomenon. Significant under-sedation may occur using subjective analysis of sedation in the ICU."

A bunch of studies have been done...the above are just two examples. Take from it what you will, but the bottom line is that there is conflicting evidence. In my two cents, especially in the ICU where a critically ill patient ARDS patient "waking up" in a timely fashion is a non-issue, it is better to err on the side of sedating them a bit more than a bit less, especially if you are paralyzing them. Now if they are so critically ill that more sedation can compromise their physiology/hemodynamics, then they probably don't need much sedation to prevent recall to begin with. We never used BIS monitors in the ICU to guide sedation, even while paralyzing patients, but I also admit I don't know what the rate of recall was at my institution.

Most of the studies that we use to guide our practice in the OR aren't really applicable to your setting since the vast majority of OR cases employ the use of volatile gases, which have an inherent "BIS" in the MAC values that have been identified for them.

 

[Urzuz]

Many ICU nurses are quite terrible at sedating patients, as evidenced by how frequently they actually manage to meet their RASS goals, or how often they actually check TOF on paralyzed patients. Having the BIS as a simple number they can target is important...

Good point. This may be a reason to use them at your shop

 

[Psai]

These studies aren't at all applicable to the setting he is discussing.

Here are some studies that may be of more use to you sluggs: Comparing BIS monitoring with clinical assessment for determining the level of sedation of mechanically ventilated adults in intensive care units

"We found insufficient evidence about the effects of BIS monitoring for sedation in critically ill mechanically ventilated adults on clinical outcomes or resource utilization. The findings are uncertain due to the low- and very low-quality evidence derived from a limited number of studies."

Bispectral index (BIS) monitoring of ICU patients on continuous infusion of sedatives and paralytics reduces sedative drug utilization and cost

"Bispectral index monitoring of ICU patients on continuous infusions of paralytics and sedatives reduces sedative drug cost as well as the recall phenomenon. Significant under-sedation may occur using subjective analysis of sedation in the ICU."

A bunch of studies have been done...the above are just two examples. Take from it what you will, but the bottom line is that there is conflicting evidence. In my two cents, especially in the ICU where a critically ill patient ARDS patient "waking up" in a timely fashion is a non-issue, it is better to err on the side of sedating them a bit more than a bit less, especially if you are paralyzing them. Now if they are so critically ill that more sedation can compromise their physiology/hemodynamics, then they probably don't need much sedation to prevent recall to begin with. We never used BIS monitors in the ICU to guide sedation, even while paralyzing patients, but I also admit I don't know what the rate of recall was at my institution.

Most of the studies that we use to guide our practice in the OR aren't really applicable to your setting since the vast majority of OR cases employ the use of volatile gases, which have an inherent "BIS" in the MAC values that have been identified for them.

If BIS has been found to be helpful in decreasing amount of anesthetic use but also has been found to not be a reliable indicator of awareness in healthy paralyzed volunteers why would those results not be relevant?

 

[dhb]

I thought ICU recall was close to 0

 

[vector2]

I thought ICU recall was close to 0

This is actually a bad thing for most icu pts. Deep sedation without sedation holidays is associated with increased ptsd, longer vent times, longer stays, and increased delirium. Most icu pts don’t even need sedation when vented as long as adequate analgesia is on board.

 

[deleted875186]

Few other things not mentioned. Paralytic will decrease the BIS value themselves. Only propofol and Benzos will affect the BIS. Preceded, opioids, ketamine, will not effect your BIS value but will act synergistically and increase sedation. Often when we put a BIS on a sick ICU patient the value will be 40 without any anesthesia, whatever that means. They probably have a slim chance of recall anyway so long as some sort of sedation is on board.

 

[BLADEMDA]

Br J Anaesth. 2015 Jul;115 Suppl 1:i95-i103. doi: 10.1093/bja/aev072.
Response of bispectral index to neuromuscular block in awake volunteers.
Schuller PJ1, Newell S2, Strickland PA2, Barry JJ2.
Author information
Abstract
BACKGROUND:
The bispectral index (BIS) monitor is a quantitative electroencephalographic (EEG) device that is widely used to assess the hypnotic component of anaesthesia, especially when neuromuscular blocking drugs are used. It has been shown that the BIS is sensitive to changes in electromyogram (EMG) activity in anaesthetized patients. A single study using an earlier version of the BIS showed that decreased EMG activity caused the BIS to decrease even in awake subjects, to levels that suggested deep sedation and anaesthesia.
METHODS:
We administered suxamethonium and rocuronium to 10 volunteers who were fully awake, to determine whether the BIS decreased in response to neuromuscular block alone. An isolated forearm technique was used for communication during the experiment. Two versions of the BIS monitor were used, both of which are in current use. Sugammadex was used to antagonise the neuromuscular block attributable to rocuronium.
RESULTS:
The BIS decreased after the onset of neuromuscular block in both monitors, to values as low as 44 and 47, and did not return to pre-test levels until after the return of movement. The BIS showed a two-stage decrease, with an immediate reduction to values around 80, and then several minutes later, a sharp decrease to lower values. In some subjects, there were periods where the BIS was <60 for several minutes. The response was similar for both suxamethonium and rocuronium. Neither monitor was consistently superior in reporting the true state of awareness.
CONCLUSIONS:
These results suggest that the BIS monitor requires muscle activity, in addition to an awake EEG, in order to generate values indicating that the subject is awake. Consequently, BIS may be an unreliable indicator of awareness in patients who have received neuromuscular blocking drugs.

 

[BLADEMDA]

Anesth Analg. 2003 Aug;97(2):488-91, table of contents.
The bispectral index declines during neuromuscular block in fully awake persons.
Messner M1, Beese U, Romstöck J, Dinkel M, Tschaikowsky K.
Author information

Abstract
Bispectral index (BIS) is an electroencephalographic variable promoted for measuring depth of anesthesia. Electromyographic activity influences surface electroencephalography and the calculation of BIS. In this study, we sought to determine the effect of spontaneous electromyographic activity on BIS. BIS was monitored in three volunteers by using an Aspect A-1000 monitor. The experiment was repeated in one volunteer. Electromyographic activity was recorded. Alcuronium and succinylcholine were administered. No other drugs were used. In parallel with spontaneous electromyographic activity of the facial muscles, BIS decreased in response to muscle relaxation to a minimum value of 33 and, in the repeated measurement, to a minimum value of 9 when total neuromuscular block was achieved. In two volunteers, no total block was achieved. BIS decreased to a minimal value of 64 and 57, respectively. In turn, recovery of BIS coincided with the reappearance of spontaneous electromyographic activity. During the entire experiment, the volunteers had full consciousness. BIS, assessed by software Version 3.31, correlates with spontaneous electromyographic activity of the facial muscles. BIS failed to detect awareness in completely paralyzed subjects. Thus, in paralyzed patients, BIS monitoring may not reliably indicate a decline in sedation and imminent awareness.

IMPLICATIONS:
The bispectral index (BIS) is an electroencephalographic variable intended for measuring depth of anesthesia. Electromyographic activity influences the calculation of BIS. We found that the administration of a muscle relaxant to unanesthetized volunteers decreases the bispectral index value. Thus, awareness in totally paralyzed patients cannot be excluded.

 

[BLADEMDA]

Whenever I do a TIVA on a paralyzed patient I utilize a BIS if at all possible. While the device is far from perfect (as I have noted above) I believe that doing a TIVA with full paralysis minus a BIS is a disservice to most patients. Of course, other signs of "awareness" are essential to monitor as well as the BIS may NOT detect partial awareness in certain patients.

 

[BLADEMDA]

1. The volunteers did not receive noxious stimulation. The experimental set-up does not replicate a clinical situation. During surgery BIS usually increases to values associated with awareness if a surgical or other noxious stimulus is applied, irrespective of whether neuromuscular blockade is present.

2. This observation is inconsistent with most clinical and research knowledge and experience. A wealth of clinical experience and large clinical studies support the view that BIS monitoring does not increase the incidence of awareness in either at-risk patients or general patient populations.2,3 In fact, it may contribute to patient safety by reducing the incidence of awareness and/or allowing lower doses of anaesthetic drugs to be administered to sensitive patients.2–4 The authors and accompanying editorial focus on the possible risk of awareness, without mentioning the controversy over the repeated observational finding of increased mortality in vulnerable patients who are deeply anaesthetized.4,5 Preventing excessively deep anaesthesia is another potential use for these monitors.

3. No (depth of anaesthesia) monitor is perfect. The BIS monitor does not provide a perfect number that can be interpreted in isolation from its clinical context.6 Learning to interpret the basic EEG trace and the effects of commonly used anaesthetic drugs can improve the utility of BIS monitoring, but no monitor based on the spontaneous EEG will reveal the full spectrum of anaesthetic drug effects on the central nervous system.7 The experimental EEGs in the article by Schuller and collegues1 do not look typical of normal general anaesthesia, largely because they lack recognizable sleep spindles. These would be expected at a BIS of 50–60 during standard volatile or propofolbased anaesthesia. The index number should not be looked at in isolation from the EEG waveform and the clinical context.

Response of bispectral index to neuromuscular block in awake volunteers

 

[BLADEMDA]

Summary: is BIS useful in the operating theatre?
There now seems to be consensus that a depth of anaesthesia monitor should be employed when using TIVA coupled with NMB, including for periods of patient transfer. Which monitor to use (or whether to employ IFT) is a secondary question, and amenable to further research. But there seems little other way by which to monitor the drug or the effect of drug in this context.

However, considerably more debate exists around whether to use depth of anaesthesia monitoring during inhalational anaesthesia. Detractors point to the many limitations of BIS 82, whereas proponents point to the many trials and advice from national bodies supporting its use 14, 15, 18, 21. The main argument making specific monitoring unnecessary is that the end‐tidal concentration is an accurate reflection of anaesthetic dose and, therefore, of likely brain effect 30. The counter‐argument is, however, that this does not reflect an individual patient's requirements and the effect of supplementary regional anaesthesia or large‐dose opoids.


Nevertheless, until the results of studies looking at the effect of depth of anaesthesia on outcome are published, a balanced approach might be to use BIS when: (a) the patient falls into a ‘high‐risk’ category; (b) the patient has had an experience of AAGA or expresses specific fears of such; (c) other measures such as heart rate or blood pressure indicate that very high levels of agent are required; (d) other measures such as blood pressure are unusually low despite low levels of agent (i.e. danger of ‘triple low’); or (e) when the patient has brain injury or neurological state impairing consciousness.

https://onlinelibrary.wiley.com/doi/full/10.1111/anae.13739

 

[CCM-MD]

If you are using the ACURASYS trial as evidence for paralysis in ARDS, then the question you need to ask is: Were BIS monitors utilized in this trial? Answer is no.

In my experience, BIS monitors in critically ill patients are extremely unreliable.

A pet peeve of mine is when physicians bundle "paralyze and prone". One can be proned without being paralyzed - which in many situations is the right thing to do. It is often forgotten that a majority of the patients proned in PROSEVA were not paralyzed.

 

[Psai]

lol blade I already posted the first two studies and I was told "These studies aren't at all applicable to the setting he is discussing."

 

[Hork Bajir]

I like Sedline more than BIS, because it gives you the option to see the raw frontal EEG data (or a spectrogram of the same). It takes a little practice to learn how to read the EEG, but once you get the hang of it you quickly realize that the PSI number spit out by the machine (the equivalent of the BIS value) is total bull****.

Another thing you can start to see, though, is that "sick" patients with some degree of pre-existing encephalopathy often have lower amplitude high frequency tracings, which show up as low enough power on the EEG spectrogram that the device doesn't pick it up (I suspect the same thing happens with the bis). We actually had a big bad case of awareness at our institution which was attributed in part to this phenomenon. As a result, any time I'm worried about awareness and want to spare my anesthetic dose in a sick patient, I'll try to put the Sedline on BEFORE giving any drug to get a sense of what the baseline looks like. If they're awake and talking to me with a tracing that looks like they should be anesthetized, that's good to know.

@sluggs I guess this doesn't apply so much to your question, other than to say that BIS values in critically ill patients haven't specifically studied, and I would guess they are somewhat less reliable.

 

[Urzuz]

Bis really helps us titrate meds, prevents awareness (pretty rare), gives us an idea of where the patient is at during tiva because there's no mac value to follow. It makes me more comfortable shutting off infusions at the end because the timing is sometimes difficult for long neuro cases for a timely wakeup. It also helps get a sense of how much of an infusion to use when you're not using bis. I generally don't like using it though outside of cardiac and neuro cases so I just go with what the attending wants.

Here are some studies

Bispectral Index Monitoring Allows Faster Emergence and Improved Recovery from Propofol, Alfentanil, and Nitrous Oxide Anesthesia | Anesthesiology | ASA Publications

http://www.sarasotaanesthesia.com/reading/literature/BIS index Declines durin NMB.pdf

Prevention of Intraoperative Awareness with Explicit Recall in an Unselected Surgical Population: A Randomized Comparative Effectiveness Trial

https://www.sciencedirect.com/science/article/pii/S0007091217310152

I encourage you to critically analyze those studies you posted and think about why they aren’t applicable to the question that was originally asked. Maybe it’s because my medical education drilled into me the importance of critiquing evidence, but you will find that the vast majority of literature is complete garbage research, and furthermore, doctors will apply those garbage conclusions to completely unrelated settings and patient populations.

Again, if you actually do go through this exercise, keep in mind that one of the biggest pitfalls in analyzing literature is drawing your own conclusions from things that aren’t explicitly being studied.

 

[Mman]

BIS was monitored in three volunteers by using an Aspect A-1000 monitor. The experiment was repeated in one volunteer. Electromyographic activity was recorded. Alcuronium and succinylcholine were administered. No other drugs were used. In parallel with spontaneous electromyographic activity of the facial muscles, BIS decreased in response to muscle relaxation to a minimum value of 33 and, in the repeated measurement, to a minimum value of 9 when total neuromuscular block was achieved. In two volunteers, no total block was achieved.

So how much do you have to pay someone to let you paralyze them but not sedate them? Also pretty sure that wouldn't get through an IRB here in the US.

 

[Psai]

I encourage you to critically analyze those studies you posted and think about why they aren’t applicable to the question that was originally asked. Maybe it’s because my medical education drilled into me the importance of critiquing evidence, but you will find that the vast majority of literature is complete garbage research, and furthermore, doctors will apply those garbage conclusions to completely unrelated settings and patient populations.

Again, if you actually do go through this exercise, keep in mind that one of the biggest pitfalls in analyzing literature is drawing your own conclusions from things that aren’t explicitly being studied.

Question: is bis useful for monitoring awareness in paralysis for icu patients?

Answer: you can have patients paralyzed with no sedation that show bis numbers that would falsely suggest that there is adequate level of sedation in healthy volunteers.

 

[Ronin786]

So how much do you have to pay someone to let you paralyze them but not sedate them? Also pretty sure that wouldn't get through an IRB here in the US.

Nothing. They were all volunteers from the anesthesia department.

 

[sluggs]

If you are using the ACURASYS trial as evidence for paralysis in ARDS, then the question you need to ask is: Were BIS monitors utilized in this trial? Answer is no.

In my experience, BIS monitors in critically ill patients are extremely unreliable.

A pet peeve of mine is when physicians bundle "paralyze and prone". One can be proned without being paralyzed - which in many situations is the right thing to do. It is often forgotten that a majority of the patients proned in PROSEVA were not paralyzed.

I beg to differ-- 91% of proned patients in PROSEVA were on paralytics; 82% supine were on paralytics.
Sorry... not trying to be a jerk... just saying. I don't paralyze for > 48 hours, but for those hours I want deep sedation.... ARDS in general (P/F < 150) and paralytics (obviously) are exceptions to "awake and breathe"

 

[sevoflurane]

I haven’t used a BIS in probably 7 years.
I don’t find them useful and generally don’t like treating a number to target sedation.

 

[sluggs]

Gestalt that I am getting from all of this plus the literature is that in a paralyzed proned ICU patient a low Bis (even 40) may not represent adequate sedation because of the paralytic, but a higher BIS would likely indicate need to crank the sedation higher???

 

[Hork Bajir]

Gestalt that I am getting from all of this plus the literature is that in a paralyzed proned ICU patient a low Bis (even 40) may not represent adequate sedation because of the paralytic, but a higher BIS would likely indicate need to crank the sedation higher???

...Or a high BIS represents EMG activity, which means you need more paralytic

...Or it could represent seizure activity, in which case... More sedation may still be the answer lol

 

[nimbus]

So how much do you have to pay someone to let you paralyze them but not sedate them? Also pretty sure that wouldn't get through an IRB here in the US.

Nothing. They were all volunteers from the anesthesia department.

Here’s a firsthand account.

“Becoming a Reckless Resident

In the spring of 1951, I was considering postdoctoral research positions and visited four biophysics departments. Before making a decision, I met with Robert Dripps, M.D., at the University of Pennsylvania in Philadelphia. Within 5 min, he persuaded me that anesthesia would be the best field for me to apply electronics to medicine.
When I started the anesthesia residency, the only monitor was a blood pressure cuff. I teamed up with Peter Safar, M.D. (1924–2003), then in his second year of residency, and we tested a new relaxant, succinylcholine, on each other before using it in patients. Peter gave me 20 mg, causing instant apnea with my still mobile arm trying to grab the oxygen mask. I ached for a week from the fasciculations.”

Gadgeteering for Health Care:The John W. Severinghaus Lecture on Translational Science | Anesthesiology | ASA Publications


And another doozy....lol

“The manufacturers and I wanted to know how accurate pulse oximeters were at low saturations. In 1986, I got permission from the committee on human experimentation at UCSF to take volunteers down to an oxygen saturation as low as 40 % just long enough to obtain an arterial sample. My credentials showed that I had been doing similar studies at high altitude looking at respiratory control, and I had publishing altitude studies for over 20 years. This permission is reviewed yearly and still stands. The manufacturers were later told by the FDA that we do not need to provide any information below 70 %. That lab is still going strong although I am no longer involved.”


In conversation with…John Wendell Severinghaus

 

[Psai]

Here’s a firsthand account.

“Becoming a Reckless Resident

In the spring of 1951, I was considering postdoctoral research positions and visited four biophysics departments. Before making a decision, I met with Robert Dripps, M.D., at the University of Pennsylvania in Philadelphia. Within 5 min, he persuaded me that anesthesia would be the best field for me to apply electronics to medicine.
When I started the anesthesia residency, the only monitor was a blood pressure cuff. I teamed up with Peter Safar, M.D. (1924–2003), then in his second year of residency, and we tested a new relaxant, succinylcholine, on each other before using it in patients. Peter gave me 20 mg, causing instant apnea with my still mobile arm trying to grab the oxygen mask. I ached for a week from the fasciculations.”

Gadgeteering for Health Care:The John W. Severinghaus Lecture on Translational Science | Anesthesiology | ASA Publications


And another doozy....lol

“The manufacturers and I wanted to know how accurate pulse oximeters were at low saturations. In 1986, I got permission from the committee on human experimentation at UCSF to take volunteers down to an oxygen saturation as low as 40 % just long enough to obtain an arterial sample. My credentials showed that I had been doing similar studies at high altitude looking at respiratory control, and I had publishing altitude studies for over 20 years. This permission is reviewed yearly and still stands. The manufacturers were later told by the FDA that we do not need to provide any information below 70 %. That lab is still going strong although I am no longer involved.”


In conversation with…John Wendell Severinghaus

I'm a big fan of the sux race stories

 

[Chetamine]

Thank you for this post. I love reading anesthesia history.

Here’s a firsthand account.

“Becoming a Reckless Resident

In the spring of 1951, I was considering postdoctoral research positions and visited four biophysics departments. Before making a decision, I met with Robert Dripps, M.D., at the University of Pennsylvania in Philadelphia. Within 5 min, he persuaded me that anesthesia would be the best field for me to apply electronics to medicine.
When I started the anesthesia residency, the only monitor was a blood pressure cuff. I teamed up with Peter Safar, M.D. (1924–2003), then in his second year of residency, and we tested a new relaxant, succinylcholine, on each other before using it in patients. Peter gave me 20 mg, causing instant apnea with my still mobile arm trying to grab the oxygen mask. I ached for a week from the fasciculations.”

Gadgeteering for Health Care:The John W. Severinghaus Lecture on Translational Science | Anesthesiology | ASA Publications


And another doozy....lol

“The manufacturers and I wanted to know how accurate pulse oximeters were at low saturations. In 1986, I got permission from the committee on human experimentation at UCSF to take volunteers down to an oxygen saturation as low as 40 % just long enough to obtain an arterial sample. My credentials showed that I had been doing similar studies at high altitude looking at respiratory control, and I had publishing altitude studies for over 20 years. This permission is reviewed yearly and still stands. The manufacturers were later told by the FDA that we do not need to provide any information below 70 %. That lab is still going strong although I am no longer involved.”


In conversation with…John Wendell Severinghaus