- Email: [email protected]
Discussion of: “Low-grade blunt hepatic injury and benefits of intensive care unit monitoring”
Accepted Manuscript Low-grade blunt hepatic injury and benefits of intensive care unit monitoring Jeffrey C. Perumean, Marco Martinez, Rachel Neal, James Lee, Taofeek Olajire-Aro, Jon B. Imran, Brian H. Williams, Herb A. Phelan PII:
S0002-9610(17)31425-3
DOI:
10.1016/j.amjsurg.2017.10.021
Reference:
AJS 12573
To appear in:
The American Journal of Surgery
Please cite this article as: Perumean JC, Martinez M, Neal R, Lee J, Olajire-Aro T, Imran JB, Williams BH, Phelan HA, Low-grade blunt hepatic injury and benefits of intensive care unit monitoring, The American Journal of Surgery (2017), doi: 10.1016/j.amjsurg.2017.10.021. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
ACCEPTED MANUSCRIPT Low-grade blunt hepatic injury and benefits of intensive care unit monitoring
RI PT
DISCUSSANT
DR. RONALD M. STEWART (San Antonio, TX): ill.
He is doing okay but could not travel.
So I am going to read
SC
his questions.
Dr. Eastridge is
TX):
M AN U
(Reading comments by DR. BRIAN EASTRIDGE of San Antonio, I would like to thank the Program Committee of the Southwestern
Surgical Congress for the invitation to discuss this paper.
I would
also like to thank the authors for a thoughtful and well‑written
TE D
manuscript.
The principal aim of Dr. Perumean and colleagues' research was to identify subpopulations of patients with blunt hepatic injury The premise
EP
that could safely be observed in a non‑ICU environment.
for this research is extremely well founded in the contemporary
AC C
climate of medicine with the ever present quest for efficiency, cost containment, and value while optimizing outcomes and preserving patient safety.
In the analysis, the metrics evaluated were deemed not sensitive or specific enough to safely predict which patients could safely be monitored in a non‑ICU environment.
Though the study did
not produce the expected results, the fundamental question still has
ACCEPTED MANUSCRIPT significant merit and suggest further investigation of your data is warranted.
I have a few questions for the authors.
1)
Though an admission SBP > 110 mmHg is associated with
a low rate of shock, why did you choose a static metric to assess For instance, why not look at those in the
RI PT
hemodynamic instability?
Group A cohort that remained stable (SBP > 110 mmHg) in the trauma resuscitation unit without PRBC or necessity for intervention? Group A would intuitively be the population
SC
2)
of angiography.
Group A had a 6.9% rate
M AN U
potentially amenable to non‑ICU observation.
How many of these procedures were therapeutic?
Were
these planned procedures or failed nonoperative management? 3)
Likewise, there was noted to be a 2.0% (Grade I&II)
TE D
and 4.3% (Grade III) operative rate for blunt hepatic injury and 9.2% all cause operative rate.
Were these planned procedures, evolution
of exam, or failed nonoperative management? The primary concern in patients with solid organ
EP
4)
injury is bleeding.
The ISS, length of stay, ICU length of stay, and
AC C
mortality suggest multiply injured patient population.
Would there
be value in excluding the moderate to severe traumatic brain injury from your cohort at least limit the confounding outcomes of neurological injury? 5)
Of the hemodynamically normal patients with grade
I/II blunt hepatic injury who underwent laparotomy, only 2.0% had a procedure for hepatic hemorrhage control.
The majority of those
ACCEPTED MANUSCRIPT patients, 4.8%, had splenic hemorrhage requiring splenectomy (46.9% of operative in grade I/II injuries).
As the original intent was to
evaluate the potential for non‑ICU monitoring of the low grade blunt hepatic injury, did the authors consider limiting their analysis to
RI PT
isolated blunt hepatic injury in order to limit the confounding effect of hemorrhage from other solid organ injury?
The focus of this research effort is a salient and very I applaud the authors
SC
important part of our practice of trauma care.
M AN U
on their preliminary efforts and suggest that further refinement of their metrics and analysis may yield remarkable results. Once again, thank you for the opportunity to review this
TE D
manuscript for the Southwestern Surgical Congress.
CLOSING DISCUSSANT
questions.
EP
DR. JEFFREY PERUMEAN:
Thank you for your comments and
Hopefully, I have them all written down here correctly.
AC C
The first one was why static metrics?
And the answer is that is an
inherent limitation of my retrospective review. If you will remember, the span of my study was from 1994 to
2014, and one of the issues with that is we transitioned from a written medical record to a electronic medical record.
So being to
obtain all of those data points in a more dynamic fashion was a limitation of this study, and I recognize that.
ACCEPTED MANUSCRIPT The second question was how many patients underwent therapeutic angios. it but I can't.
That's a great question.
I wish I could answer
One of the other limitations of doing the chart
reviews, you don't really get the full sense of what happens in the
sometimes that is not translated in the chart.
RI PT
trauma bay, and what leads to the decisions for plan of care, and So you may think
that a patient is going to angio for one reason, but I wasn't able to
SC
determine whether the patient was going for a preemptive angiography,
M AN U
angioembolization, or whether they were going because they had continued hemorrhage that wasn't responsive to blood or was transiently responsive to blood product.
For your third question about planned procedures in the A lot of times, the
TE D
operating room, I ran into the same limitations.
OR reports wouldn't say if the patient was going to the operating room because they failed conservative management, or they had
EP
continuing hemorrhage suspected from the suspected splenic injury, or at times it was just trauma to the abdomen going to the OR.
This is
AC C
probably one of the limitations of maybe residents writing a lot of our operative reports.
I'm guilty of that myself.
Your fourth question, excluding neurological injury, we
didn't specifically look at this, but this is an excellent point. And I actually have the capabilities to go back and review this in my database, and I am going to do that in future research, because I do think the neurological injury does view the ICU length of stay, the
ACCEPTED MANUSCRIPT overall length of stay, and possibly cause of death.
So that's
something I will look at in further research. Lastly, an analysis to isolated blunt hepatic injury, one of the limitations of just analyzing patients with isolated blunt
solid conclusions.
RI PT
hepatic injury is that I don't have enough patients to really make Out of screening 1900 patients over a 10‑year
period, I only got a little under 200 that had isolated blunt hepatic So to making significant conclusions about that small of a Only 25 of those patients out of 188 that had
M AN U
database is difficult.
SC
injury.
isolated blunt hepatic injury required blood transfusion, laparotomy or angiography.
So it's a bit of a challenge to determine whether we
TE D
can make any conclusions from that.
Thank you for your comments.
DISCUSSANT
I applaud you for
EP
DR. SHARMILA DISSANAIKE (Lubbock, TX):
looking at an important topic, but I would say that you had applied
AC C
training to BIG data, and I would say we see a lot of studies these days from large databases that are true BIG data, multi‑center national databases.
And the problem is you don't have the
granularity to really question a lot of these things for the reasons you just described.
You can't tell why they went to the OR.
can't sort out whether it was the spleen or the liver.
You
You have got
a lot of information, and when you go back to 1996, I would caution
ACCEPTED MANUSCRIPT you that not only do you not have information because it was written on a scrap of paper and tucked away somewhere, but you also don't have information because the way we treat these patients has changed So I would say you are really muddying the waters there
because of the difference.
So what I would suggest is that maybe you
RI PT
a lot.
do need to go back to just those 200 or 188 who had isolated hepatic injury because that's the question you are really asking.
Almost every liver injury
SC
these patients do have combined injuries.
A lot of
And if I
M AN U
we see has either a kidney or a spleen lac of some sort.
have a liver and a spleen, it's almost always the spleen that's going to guide my management operatively or angiographically, usually not the liver.
That seems to be how it works.
TE D
So trying to tease out how to manage liver injuries from a cohort that includes all these other things, I think it's a little difficult.
But if you want to answer what to do with an isolated
EP
liver injury, I would say you do have enough numbers in this group to look at, and you have almost answered the question already, which is
well.
AC C
that almost all those patients with isolated liver injuries did very And I would say that liver injuries that need surgery usually
require enough blunt force trauma, that they are going to injure something else if they are going to need something done, and if it's an isolated liver injury, my suspicion is that a lot of times those are minor injuries ‑‑ not all but a lot ‑‑ and probably could do without ICU admission.
ACCEPTED MANUSCRIPT
CLOSING DISCUSSANT
Thank you for your suggestion.
RI PT
DR. JEFFREY PERUMEAN:
DISCUSSANT
I just want to
SC
DR. MICHAEL CORNEILLE (San Antonio, TX):
M AN U
make one quick comment and a quick question.
The comment is, rather than going back 20 years, just to echo the previous discussant there, consider multi‑center efforts so that your data is more contemporary, looking back 20 years looking
TE D
across institutions, and my question is, presumably, the clinical question is trying to offer your ICU with these patients, hemodynamically normal, so did you consider maybe just implementing Just
EP
or just looking at protocol where it was just grade 1 injuries?
start with the grade 1 injuries and see if it's feasible with just
AC C
grade 1 rather than grades 1, 2 and 3, or even 1 and 2.
CLOSING DISCUSSANT
DR. JEFFREY PERUMEAN:
One of the issues that I found with
that is that lower grade injury actually had an increased chance of having a blood transfusion and/or going to the operating room when it
ACCEPTED MANUSCRIPT was a hemodynamically stable patient.
So I think one of the issues
is what we are actually looking for as far as deciding what patients we are going to put in the ICU, and I think we have to go away from the grading system and towards maybe hemodynamic parameters or how And that's one of the
RI PT
the patient trends over the first 24 hours.
limitations of using the grading system that has been described
AC C
EP
TE D
M AN U
SC
previously, not just in our paper.
S0002-9610(17)31425-3
DOI:
10.1016/j.amjsurg.2017.10.021
Reference:
AJS 12573
To appear in:
The American Journal of Surgery
Please cite this article as: Perumean JC, Martinez M, Neal R, Lee J, Olajire-Aro T, Imran JB, Williams BH, Phelan HA, Low-grade blunt hepatic injury and benefits of intensive care unit monitoring, The American Journal of Surgery (2017), doi: 10.1016/j.amjsurg.2017.10.021. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
ACCEPTED MANUSCRIPT Low-grade blunt hepatic injury and benefits of intensive care unit monitoring
RI PT
DISCUSSANT
DR. RONALD M. STEWART (San Antonio, TX): ill.
He is doing okay but could not travel.
So I am going to read
SC
his questions.
Dr. Eastridge is
TX):
M AN U
(Reading comments by DR. BRIAN EASTRIDGE of San Antonio, I would like to thank the Program Committee of the Southwestern
Surgical Congress for the invitation to discuss this paper.
I would
also like to thank the authors for a thoughtful and well‑written
TE D
manuscript.
The principal aim of Dr. Perumean and colleagues' research was to identify subpopulations of patients with blunt hepatic injury The premise
EP
that could safely be observed in a non‑ICU environment.
for this research is extremely well founded in the contemporary
AC C
climate of medicine with the ever present quest for efficiency, cost containment, and value while optimizing outcomes and preserving patient safety.
In the analysis, the metrics evaluated were deemed not sensitive or specific enough to safely predict which patients could safely be monitored in a non‑ICU environment.
Though the study did
not produce the expected results, the fundamental question still has
ACCEPTED MANUSCRIPT significant merit and suggest further investigation of your data is warranted.
I have a few questions for the authors.
1)
Though an admission SBP > 110 mmHg is associated with
a low rate of shock, why did you choose a static metric to assess For instance, why not look at those in the
RI PT
hemodynamic instability?
Group A cohort that remained stable (SBP > 110 mmHg) in the trauma resuscitation unit without PRBC or necessity for intervention? Group A would intuitively be the population
SC
2)
of angiography.
Group A had a 6.9% rate
M AN U
potentially amenable to non‑ICU observation.
How many of these procedures were therapeutic?
Were
these planned procedures or failed nonoperative management? 3)
Likewise, there was noted to be a 2.0% (Grade I&II)
TE D
and 4.3% (Grade III) operative rate for blunt hepatic injury and 9.2% all cause operative rate.
Were these planned procedures, evolution
of exam, or failed nonoperative management? The primary concern in patients with solid organ
EP
4)
injury is bleeding.
The ISS, length of stay, ICU length of stay, and
AC C
mortality suggest multiply injured patient population.
Would there
be value in excluding the moderate to severe traumatic brain injury from your cohort at least limit the confounding outcomes of neurological injury? 5)
Of the hemodynamically normal patients with grade
I/II blunt hepatic injury who underwent laparotomy, only 2.0% had a procedure for hepatic hemorrhage control.
The majority of those
ACCEPTED MANUSCRIPT patients, 4.8%, had splenic hemorrhage requiring splenectomy (46.9% of operative in grade I/II injuries).
As the original intent was to
evaluate the potential for non‑ICU monitoring of the low grade blunt hepatic injury, did the authors consider limiting their analysis to
RI PT
isolated blunt hepatic injury in order to limit the confounding effect of hemorrhage from other solid organ injury?
The focus of this research effort is a salient and very I applaud the authors
SC
important part of our practice of trauma care.
M AN U
on their preliminary efforts and suggest that further refinement of their metrics and analysis may yield remarkable results. Once again, thank you for the opportunity to review this
TE D
manuscript for the Southwestern Surgical Congress.
CLOSING DISCUSSANT
questions.
EP
DR. JEFFREY PERUMEAN:
Thank you for your comments and
Hopefully, I have them all written down here correctly.
AC C
The first one was why static metrics?
And the answer is that is an
inherent limitation of my retrospective review. If you will remember, the span of my study was from 1994 to
2014, and one of the issues with that is we transitioned from a written medical record to a electronic medical record.
So being to
obtain all of those data points in a more dynamic fashion was a limitation of this study, and I recognize that.
ACCEPTED MANUSCRIPT The second question was how many patients underwent therapeutic angios. it but I can't.
That's a great question.
I wish I could answer
One of the other limitations of doing the chart
reviews, you don't really get the full sense of what happens in the
sometimes that is not translated in the chart.
RI PT
trauma bay, and what leads to the decisions for plan of care, and So you may think
that a patient is going to angio for one reason, but I wasn't able to
SC
determine whether the patient was going for a preemptive angiography,
M AN U
angioembolization, or whether they were going because they had continued hemorrhage that wasn't responsive to blood or was transiently responsive to blood product.
For your third question about planned procedures in the A lot of times, the
TE D
operating room, I ran into the same limitations.
OR reports wouldn't say if the patient was going to the operating room because they failed conservative management, or they had
EP
continuing hemorrhage suspected from the suspected splenic injury, or at times it was just trauma to the abdomen going to the OR.
This is
AC C
probably one of the limitations of maybe residents writing a lot of our operative reports.
I'm guilty of that myself.
Your fourth question, excluding neurological injury, we
didn't specifically look at this, but this is an excellent point. And I actually have the capabilities to go back and review this in my database, and I am going to do that in future research, because I do think the neurological injury does view the ICU length of stay, the
ACCEPTED MANUSCRIPT overall length of stay, and possibly cause of death.
So that's
something I will look at in further research. Lastly, an analysis to isolated blunt hepatic injury, one of the limitations of just analyzing patients with isolated blunt
solid conclusions.
RI PT
hepatic injury is that I don't have enough patients to really make Out of screening 1900 patients over a 10‑year
period, I only got a little under 200 that had isolated blunt hepatic So to making significant conclusions about that small of a Only 25 of those patients out of 188 that had
M AN U
database is difficult.
SC
injury.
isolated blunt hepatic injury required blood transfusion, laparotomy or angiography.
So it's a bit of a challenge to determine whether we
TE D
can make any conclusions from that.
Thank you for your comments.
DISCUSSANT
I applaud you for
EP
DR. SHARMILA DISSANAIKE (Lubbock, TX):
looking at an important topic, but I would say that you had applied
AC C
training to BIG data, and I would say we see a lot of studies these days from large databases that are true BIG data, multi‑center national databases.
And the problem is you don't have the
granularity to really question a lot of these things for the reasons you just described.
You can't tell why they went to the OR.
can't sort out whether it was the spleen or the liver.
You
You have got
a lot of information, and when you go back to 1996, I would caution
ACCEPTED MANUSCRIPT you that not only do you not have information because it was written on a scrap of paper and tucked away somewhere, but you also don't have information because the way we treat these patients has changed So I would say you are really muddying the waters there
because of the difference.
So what I would suggest is that maybe you
RI PT
a lot.
do need to go back to just those 200 or 188 who had isolated hepatic injury because that's the question you are really asking.
Almost every liver injury
SC
these patients do have combined injuries.
A lot of
And if I
M AN U
we see has either a kidney or a spleen lac of some sort.
have a liver and a spleen, it's almost always the spleen that's going to guide my management operatively or angiographically, usually not the liver.
That seems to be how it works.
TE D
So trying to tease out how to manage liver injuries from a cohort that includes all these other things, I think it's a little difficult.
But if you want to answer what to do with an isolated
EP
liver injury, I would say you do have enough numbers in this group to look at, and you have almost answered the question already, which is
well.
AC C
that almost all those patients with isolated liver injuries did very And I would say that liver injuries that need surgery usually
require enough blunt force trauma, that they are going to injure something else if they are going to need something done, and if it's an isolated liver injury, my suspicion is that a lot of times those are minor injuries ‑‑ not all but a lot ‑‑ and probably could do without ICU admission.
ACCEPTED MANUSCRIPT
CLOSING DISCUSSANT
Thank you for your suggestion.
RI PT
DR. JEFFREY PERUMEAN:
DISCUSSANT
I just want to
SC
DR. MICHAEL CORNEILLE (San Antonio, TX):
M AN U
make one quick comment and a quick question.
The comment is, rather than going back 20 years, just to echo the previous discussant there, consider multi‑center efforts so that your data is more contemporary, looking back 20 years looking
TE D
across institutions, and my question is, presumably, the clinical question is trying to offer your ICU with these patients, hemodynamically normal, so did you consider maybe just implementing Just
EP
or just looking at protocol where it was just grade 1 injuries?
start with the grade 1 injuries and see if it's feasible with just
AC C
grade 1 rather than grades 1, 2 and 3, or even 1 and 2.
CLOSING DISCUSSANT
DR. JEFFREY PERUMEAN:
One of the issues that I found with
that is that lower grade injury actually had an increased chance of having a blood transfusion and/or going to the operating room when it
ACCEPTED MANUSCRIPT was a hemodynamically stable patient.
So I think one of the issues
is what we are actually looking for as far as deciding what patients we are going to put in the ICU, and I think we have to go away from the grading system and towards maybe hemodynamic parameters or how And that's one of the
RI PT
the patient trends over the first 24 hours.
limitations of using the grading system that has been described
AC C
EP
TE D
M AN U
SC
previously, not just in our paper.