Introduction:
Good morning. On behalf of Dr. Freireich and Institutional Grand Rounds, I'd like to welcome you all and I'd like to introduce the chair for this morning's session, Dr. Gregory Botz. Dr. Botz is a professor of Critical Care Medicine in the Department of Critical Care. I would like to welcome all of our speakers and thank you all for attending, and with that, Dr. Botz.
Dr. Gregory Botz:
Good afternoon. It's a pleasure to be here this afternoon to bring this topic to you about improving patient safety, the impact of healthcare associated infections. As you know, we've had tremendous advances in technology and science for medical science over the last 50 to 100 years, but it's still right on top of the nineteenth century medical framework. Hospitals can be dangerous places at times. Even the safest industries have mishaps every now and then, and medicine is no different.
Almost 10 years ago, the Institute of Medicine presented this report on To Err is Human, looking at medical injuries in the United States. It estimated conservatively that somewhere between 44,000 and 98,000 patients died per year from medical injuries or medical errors. And the direct health costs at that time were estimated to be somewhere between nine and 15 billion dollars per year. That's more deaths per year than breast cancer or AIDS or motor vehicle accidents. So if I look at this correctly, one of the leading causes of death in the United States could be being a patient in a hospital in United States. In a followup report, Crossing the Quality Chiasm, the main theme was that trying harder does not work. You have to change the system of care. There have been a number of national initiatives looking at trying to change the delivery of care in our hospitals to reduce unnecessary injuries or errors.
And this is from the Institute for Healthcare Improvement, the 100,000 lives campaign, which aimed at trying to reduce unnecessary injuries or errors in our patient population. There were six main planks looking at strategies to reduce errors and increase patient safety and three of the planks had to do with healthcare associated infections. It's my pleasure today to introduce three speakers who are going to talk about the impact of healthcare associated infection. Our first speaker is Dr. Dennis Maki. He is the Ovid O. Meyer Professor and head of Infectious Disease Medicine Center. He is an attending physician at the Center of Trauma and Life Support at the University of Wisconsin. Please help me welcome Dr. Maki.
[ Applause ]
Dr. Dennis Maki:
Thank you very much. It's a great honor to participate in this symposium this morning and particularly to be able to address an area that I think has long needed regress in modern day healthcare. My charge this morning is to define the problem of line sepsis. My goals this morning are to address four issues very quickly. I want to tell you about the types of devices that we use for vascular access because this is fundamental to developing more effective strategies for prevention and it turns out that the risk of infection differs tremendously with different types of devices and strategies that may work for one may not work for another.
I want to talk about the magnitude of risk with specific types of devices, the impact of line sepsis both medically and economically, and then finish with talking about pathogenesis. If there is anything we have learned in infectious diseases but also in almost all areas of medicine, understanding pathogenesis is a cornerstone of developing more effective strategies for prevention and for treatment.
So left's talk about the types of devices that we have available to use in clinical medicine. There are two major categories of devices available to us. The oldest are the short-term devices and these are devices that are rarely used for longer than ten days. We are talking about peripheral venous catheters, probably 100 to 150 million used in the United States every year, needles or small catheters, non-cuffed, non-tunneled central venous catheters, conventionally placed in the internal jugular, subclavian or femoral veins, NR20 catheters used for hemodynamic moderate. These were the first devices that were really developed and used widely.
In the last 10 to 20 years, there has been an explosion in the use of long-term vascular access, devices that are intended to be used for months or even indefinitely. Cuffed and tunneled catheters shown here schematically, subcutaneous central ports, which I know you are very familiar with, with the huge numbers of patients you give chemotherapy to for cancer. Then the peripherally inserted central catheter, which has increasingly supplanted cuffed and tunneled catheters and even short-term central catheters in the acute care hospital.
The risk of infection differs for these different types of devices as does pathogenesis. So what do we know about the magnitude of risk with different devices? These are data from a meta-analysis recently published, which focused on perspective studies in which every catheter and cohort was scrutinized for evidence of infection. These are research type data. They're much more rigorous than surveillance data, and in most of these studies we have, we can feel reasonably confident that every single bacteremia that might well have been related to the catheter was identified. And as you can see here, the risk of infection ranges widely. We used to express risk for hundred devices, but it's far more meaningful to express it for thousand days.
For instance, a short-term central venous catheter used in an ICU patient has a risk for thousand days of tube, but the risk in percentage is about four. If you look at a cuffed and tunneled catheter, the risk looks much higher. When you realize it is in place many fold longer, the risk for thousand days is actually less. All types of devices pose a risk of infection. It is important also to point out that these data might suggest that PICCs have a very low risk of infection and that's certainly true in the outpatient studies. These are mostly outpatient studies.
Inpatient PICCs have come into very wide use and it was our impression that they posed a much higher risk of infection than the literature might have suggested. And in this perspective study we found the risk of infection much, much higher than outpatient PICC studies. If we do a meta-analysis of the perspective studies of PICCs used in inpatients, we see a risk of infection here not that much different than conventional short-term central catheters and considerably higher than we would see with Hickmans or Broviacs. So it's very important to define the population that the device is being used in.
I would also point out that children have a risk of infection at least as high, probably a little higher than do adults. Now in the intensive care unit, the risk is highest by far. The data suggests that probably three-quarters, half the three-quarters of all bacteremias and candidemias that occurred in the ICU derived from an intravascular device and these are based on perspective studies where an effort was made to identify the source of every hospital acquired or ICU acquired blood stream infection.
Now these are data we have recently put together from a consortium of hospitals in the developing world. Dr. Rosenthal is a former research fellow and this is an effort to try to bring modern day infection control to the hospitals of the developing world. These are the best hospitals in these countries that have made a commitment to infection control and are doing surveillance following CDC criteria and methods and these data suggest that the risk of a hospital acquired infection in an ICU in the developing world is many folds higher than we would consider acceptable in North America. If we look at line sepsis, central line associated blood stream infection. This is at least four to six times as high as we would consider acceptable in the North American ICU. People talk about the terrible burden of infectious disease in the developing world due to malaria and tuberculosis and childhood diarrhea, but I submit to you that hospital acquired infections probably pose at least as great a burden, but one that is much more easily regressed than finding an effective vaccine for malaria or tuberculosis.
Now what is the impact of line sepsis? These data suggest that somewhere around two hundred million devices are used in United States every year. We said there is a one in one thousand risk of bacteremia with two hundred million devices, that is two hundred thousand blood stream infections, but there is no device that has a risk that low. By a variety of techniques, I think one can legitimately estimate there are at least five hundred thousand blood stream infections due to vascular access in this country every year. This is not a terrible trouble. What is the impact medically and economically? Well in these clearly prolonged hospitalizations, the best evidence suggests they have at least 30,000 dollars to the cost of healthcare, they cause loss of access sites for patients with long-term devices who may need critical access or cancer or dialysis and there is attributable mortality.
There have been papers recently written that question does line sepsis really kill people? Now let me show you some data that I think it does. This is a perspective study we did several years ago of a novel technology a little chlorhexidine impregnated disk to prevent infection with short-term central and arterial catheters and the study showed unequivocally it was very beneficial. It reduced the risk of infection about 60%.
In this study, we followed patient's raw hospitalizations, as long as they were in the ICU, they stayed in the study. What we found at the end of the day was that patients who had had a cath related blood stream infection had an adjusted three-fold increased risk of not surviving hospitalization. This they randomized to the novel strategy for prevention of those effected they were half as likely to die during hospitalization. Line sepsis kills. If I look at the data from the developing world that we've developed the attributable mortality of these infections is astronomical.
You have heard about the impact of the National Academy of Sciences, Institute of Medicine report. It has infections in hospitals on their radar screen, but let me show you something else that is even going to be more aggressive. Recent legislation that now mandates that the federal government is not going to pay for the added costs of line sepsis or sternotomy infection after heart surgery or catheter associated urinary tract infections. On the drawing board is all MRSA infections in the hospitals, C. difficile infections and probably most surgical infections.
Let me finish by saying a few things about pathogenesis, because pathogenesis drives the system and in terms of pathogenesis this is a schematic of a percutaneous intravascular device. It has been placed through the skin into the patient's blood stream. Infection occurs when microorganisms from some source invade this system and form a biofilm on the surface. This is simply a scanning electron micrograph of a biofilm. It is shown here schematically, once we have a rich biofilm, we are going to have a risk of shedding and causing a cath-related bacteremia.
Our goals of infection control are very simple. We would like to prevent planktonic organisms from gaining access to the implanted device and forming a biofilm, but we would also like to, if they do gain access, perhaps have strategies to delay their capacity to adhere and form a biofilm. You are going to be hearing about these strategies shortly.
What is the source of organisms that cause device-related bloodstream infection? Is it skin? Is it luminal contaminants from manipulation of these connections? Is it contaminated infusive? We find those answers by doing perspective studies of large cohorts of a single type of a device where all these sites are cultured at the time of catheter removal and we then do DNA subtyping to show where is the concordance. Here is a catheter that caused bacteremia and the blood eye slit is identical to the tip eye slit and the hub, but is different from the skin and the fluid.
This was an intraluminal contaminant. What we now know is that with short-term devices, the vast majority of infections originate from the patient's skin and strategies aimed to deny organisms on the skin from gaining access to the insertion site clearly reduced risk and often very substantially. On the other hand, with long-term devices, it is luminal contaminants that gain access during manipulation of the system. I'd like to tell you a little bit, and this is certainly true when we're talking about things like Hickmans and other cuffed and tunneled catheters or ports. I'd like to tell you about something you are going to be hearing about shortly, and that is the evolution of mechanical valved needleless systems that are developed to try to prevent sharp instruments. There are many of these systems available. They avoid the need to use a needle to make a connection. There have been early reports suggesting maybe they can become contaminated and cause bacteremia.
Recently, we've had multiple reports of increased rates of blood stream infection after hospitals bring a Neuwelt system into use. Look at these increases in these individual institutions. This is a problem that we have not identified the magnitude of adequately, but in these hospitals when they go back to an old system, rates often drop. This is an area that certainly is in need of much greater study.
I would also finish up by saying that when you have an outbreak of bacteremias, it usually means contamination with a fluid and almost anyway you can think that multiple organisms gained access to a patient's infusions, particularly in an ICU has probably caused an epidermic at some time in the world. So in sum, we've learned a great deal about the magnitude and pathogenesis of infection, but most importantly, this is now being applied very effectively to be able to be prevent line sepsis. Thank you very much.
[ Applause ]
Dr. Gregory Botz:
Thank you, Dr. Maki. Our next speaker is Dr. Michael Gropper. Dr. Gropper is professor and vice chair of anesthesiology and preoperative care, director of critical care medicine and chair of medical quality at the University of California, San Francisco. Please join me in welcoming Dr. Gropper.
Dr. Michael Gropper:
Well I thank you all for inviting me. I think it's the second time in a couple of months I've had to follow Dr. Maki, so I should be asked to go first next time, but... I'm going to talk to you today about my perspective as an intensivist and I'll show you why this is becoming increasingly important to us in the ICU, but it should be important to all of you as well.
If you look at the safety of various endeavors and this is data from the British Medical Journal, you can see that there are things that we do that are quite dangerous like climbing Mount Everest, I think that's about, I don't know, one in fifteen mortality rate, yet when I came here yesterday to Houston I certainly didn't worry about getting on a plane, it's an incredibly safe thing to do. You don't worry about the training of the pilot, plane crashes are extremely rare. But there are other things that are in between that maybe not quite so safe.
Now as anesthesiologists we have done a good job in providing good outcomes in patients that are undergoing surgery, yet if you look at total medical risks, in other words the average patient who is hospitalized is about one in 104 risk of a complication that is significant, in other words mortality or morbidity. And this in catheter related bloodstream infection is one of the major causes of this morbidity, and therefore, that's why we're here today. You've already heard some background on the impact of this. I think when the report from the Institute of Medicine first came out, myself and other physicians felt it was a little overblown and that these people had an agenda. It can't possibly be that big a problem. Now in hindsight it is clear that it very much underestimated the magnitude of medical errors.
Because as we've started to look and pick up the carpet and look underneath, we found a lot of bad things. Major healthcare expenditures spent on this is at least five billion dollars a year and a large percentage of those are related to nosocomial infections in the ICU, in particular catheter related bloodstream infection. And the major infections that we worry about in the ICU are listed here. Whether you are in a medical ICU or surgical ICU, ventilator associated pneumonia and catheter related bloodstream infection are the two major factors.
This is data from the CDC. As you'll see, we have made progress as we have started to pay attention to it. It is heartwarming that we have done that. It's also a little horrifying that we paid so little attention to it before. It was so easy to make improvements.
If you look at the attributable mortality and cost, Andy Shore has done a nice analysis published in 2001. He found about a 56,000 dollar increase in cost per infection. At our, now we are in San Francisco, very high cost of living, high personnel costs, high hospitalization costs and our own analysis it is about 80,000 dollars per infection. Now what is important, the good news is that if I go to the hospital administration for support to try to prevent these, they're usually fairly willing to do so because it is such a major impact on the bottom line of the hospital. But you can see that if you look at risk adjusted patients, the increase in length of stay in the ICU and the hospital overall we can make a major impact with some simple measures that we will talk about today.
Well nothing moves medical practice like public opinion, so we have, you know, the joint commission of people like that that certainly changed off every practice, mostly for the better, but nothing does it like articles in the New York Times. Like you'll see here, and you heard a little bit about pay for performance, I will tell you a little bit more.
This is an article in the New York Times and of course compelling stories about people that had complications in the hospital, catheter related bloodstream infections, pressure ulcers, etc., and so when the public sees this, they take a simple view of this. If these are easier to prevent, there is no reason to have these things and I'll admit that when I talked to various business leaders and others and tell them that doctors only wash their hands about 70 to 80% of the time, in my intensive care unit, they are horrified. So public reporting is going to drive this.
This is a map of the United States with states that have passed legislation for required public reporting and so in fact you can see that this has been submitted in Texas, I don't know the status now, but in California now, we've gone to this purple, which is required public reporting. You can get onto a website and see the infection rates in my hospital and in my competitor hospitals in the area. So we need to pay attention. And the way you get any one to pay attention, whether a physician or anyone else, is to make it affect the bottom line. Pay for performance is a system that rewards quality with financial incentives large enough to affect change. Meaning that offer somebody 100 dollars for hard work of a year probably isn't going to change their behavior. You offer them maybe five thousand dollars it might. While the federal government has taken this on, and pay for performance is of course a misnomer, they will not pay for under performance. No one is going to get any bonuses for this, but you will be penalized if you don't perform.
So if you look at this extract, which is from the Federal Register and the law that was recently passed looking at payment for medical errors, I forgot a little piece here to highlight, and that is that payments will be withheld from hospitals for care associated with treating certain catheter related urinary tract and vascular catheter associated infections. Other things as you heard about as well mediastinitis or pressure ulcers. This can have a major impact if you saw the length of stay that can happen with a patient with catheter related bloodstream infections and nothing hits home like your own pockets.
I will show you something, a form I had to sign this year with the medical center who pays part of my salary as Head of Intensive Care. I had to give specific metrics of what I was going to do in the ICU and there would be a bonus or withholding of pay based on these metrics, based on the, a threshold target and outstanding infection rate determining the amount of money that would come to the ICU. This is actually money that goes towards educational programs and food and other things that help the medical staff. So this is obviously affecting my bottom line. Preventing catheter related bloodstream infections has technology as you will hear about particularly from Dr. Raad, but most of it is quite simple: Washing your hands, using a barrier that covers the body from head to toe, cleaning the skin as you've heard about, using chlorhexidine or trying to use a subclavian as a site, but it can't be done in all patients.
Ultrasound guidance has been recommended and then another thing, the easiest thing you can do is take out the lines that are not needed. What got every body's attention was when the group at Johns Hopkins led by Peter Pronovost and Sean Berenholtz did a nice study comparing two ICUs at their hospital: One was general surgical one was a cardiac surgical ICU, and they said we think we can train people and by training them reduce these complications. So before any house staff or physician could work in the ICU they had to complete a web based training module before they could insert a line. They put all the equipment together in one place so when tempted to break your sterile technique in the middle or look for something that you couldn't find, asking themselves everyday: Can we take out the catheter? And then have the bedside nurse play an important role in patient safety and that is to complete a checklist during insertion and say listen wait a minute, you need to stop because you didn't follow the sterile technique. Something that we're all used to as anesthesiologists in the operating room, but something that has never really been translated outside of the operating room.
So they had a checklist that the nurse would fill out: Did they wash their hands, use sterile gloves, gown etc., was this emergent or not, was this a new line or was it rewired. What they found was that in the study ICU over time the infection rate went to 0, and in fact, in the control ICU the rate fell as well because house staff go back and forth between the ICUs, nurses might work between two ICUs, and their final results, they had started out with a pretty poor rate at 11.3 per thousand and their final rate was 0. And in fact, as I said, it fell in the control ICU as well. They did what I will show you that I did as well, and that was to calculate how many infections you prevented, how many deaths you prevented, and of course, how much money you saved the hospital. So you can go to the hospital CEO and say look at this, and I need more ultrasound machines, I need some support for another fellow, etc.
That was impressive, but I think what was more impressive was when they went and did this on a much larger scale, published in the New England Journal at the end of last year. In the state of Michigan, in 108 ICUs where they put in basically the same programs getting people to wash their hands, use sterile barriers, etc., what they found was a substantial reduction in the number of infections in these ICUs. So here in their baseline period, you can see what their rates were, in fact they were not very high, their rates were around 3, 2.5 to 3, but after intervention, whether you are in a small hospital or large hospital, a teaching hospital, a non-teaching hospital, many of these hospitals got down to a rate of 0 just by these same interventions. So these are clearly preventable infections if you pay attention to them.
Now there's some criticism for this study. There is not a lot of detail there for us to look at, for example, well I mean, what was the compliance for these interventions? In other words, if the rate fell without compliance, then there must have been something else. What technique did they use for diagnosis? There are many different techniques. We maybe comparing apples and oranges. Dr. Maki I know is working on new guidelines that will come out next year that will hopefully unify this. Did they use any microbial catheters? And what was the cost on both mortality and length of stay? None of that data was reported.
Nevertheless, I think it gets your attention that a large group of hospitals, community academic medical centers can reduce their rates. So then the question is if that is so successful, what is the role of technology in our ability to prevent these infections? Well, we have antibiotic-coated catheters, and you'll hear from one of the pioneers following me, there's data that reports decreased colonization, decreased infection rates, mortality, morbidity, ICU length of stay, etc., and I think that if you look at how medicine is practiced and that you have human factors in medicine, that you should leverage all the technology that you have to minimize the risks to these patients, particularly since the added cost of this technology tends to be low.
There's a nice meta-analysis that was published in JAMA a few years ago from Sanjay Sen and his group which looked at the overall efficacy and using meta-analysis, efficacy of using antibiotic-impregnated catheters. What they found in their meta-analysis was a substantial reduction in catheter related bloodstream infection across all of these studies. In no study was the infection rate increased and in almost every study the infection rate was decreased. When it came the time for me to consider this in our hospital, we relied on evidence based medicine, and there really were primarily two products available at the time. It was nice that Dr. Rodden and his colleague Ravi Darwish did a study comparing these two catheters. Based on the results of this study which showed that the minocycline-rifampin coated catheter was more efficacious, we chose that as the catheter in our institution.
So what have we done at UCSF to reduce infections? So as I said, I have two hats, I wear one as the director of the Adult Intensive Care Units but also as the director of quality, and although I enjoy that, also there is a lot of scrutiny about what I do because I can't talk about quality if I can't achieve it in my own ICUs. So we took this on as a major program. We had the CDC recommendations, which I have shown you, but we also added a little bit more on top of that. We have a very aggressive hand hygiene campaign, which aside from the usual signage and things like that involves observations by nurses throughout the hospital, which are silent observations to measure compliance rates. We have our chief medical officer involved with some disciplinary actions, which if you are actually sighted up to three or four times, you can lose your privileges in the hospital. We haven't done that yet, but the threat is there and some people are getting close. Chlorhexidine skin antisepsis, etc., and then we added some additional things.
It is the same kind of checklist that you saw from the Johns Hopkins study, we routinely use ultrasound guidance whenever we can, not always possible subclavian placement, an antibiotic stewardship program, which I saw that you have here as well, which I think is very important in trying to change pressure, daily review of the need to remove lines, and then a discharge checklist, which I'll show you, so we have daily goals. When we round in the ICU, we cover a number of things that we think improve patient safety. It also helps us focus our plans for that patient that day and this includes can we remove any catheters in this patient? And very importantly when a patient is ready to leave the ICU, we go through a checklist, both the nurse and the house staff and see can we take these catheters out? Can we start a peripheral IV and remove the PICC line or the central line? And we found that there were a lot of catheters that we could, and this included the urinary tract catheters as well.
In fact, in about a six-month period, we tracked the number of catheters we removed based on this trigger or the discharge checklist and you can see a large number of catheters removed. Well, these are catheters that can never cause an infection. This couldn't possibly be an easier way to prevent infection as we like to say if you don't have a central line, you can't have a central line infection.
I get monthly recording from these nurses that are doing the observations about compliance for physicians: Are they washing their hands, are they wearing a cap and gown when they put in central lines and in fact we have achieved very high compliance, you can see that this is in red where the patient wasn't draped entirely so that's something I need to look into and find out why is that occurring. Then daily, or excuse me, monthly reports from each ICU on infection rates and we use this data. In fact we will play one ICU off another, we'll buy pizza for the dayshift and nightshift for the ICU that is going to have the lowest infection rates. But you need to share this data with those nurses and physicians that work there because you need to give them feedback that their efforts are working. And they take this on as a source of pride and it is in their mind every day. In fact, our institution now has an incentive program where as a unit, for example, the intensive care units can take this on as a goal so not only was my salary affected by this but nurses would get a bonus if they were able to achieve a certain infection rate and it was significant. It was between 500 and 1500 dollars if we're able to achieve those incentives.
It is very inexpensive for the medical center to do that if you calculate the cost savings that occur. So I did the same kind of thing, I did a calculation, I looked at our initial rate of 3.7 for 1000 line days, calculated our cost at 80,000 dollars per infection and expected 16 deaths. We implemented that program in early 2005 and we were able to drop our rate in half down to 1.7. I calculated how many infections that was, and I estimated I saved eight lives and that I saved the hospital two million dollars. Immediately I went to see the CEO and the chief medical officer and was able to get some additional support for the ICUs based on that.
Now what is interesting is that in the past and if your savvy, you'd say well that's actually not quite true, and they didn't say that to me, I did save the lives, hopefully, but I didn't save the money, because in the past we actually were paid for those infections and that is no longer the case. As of 2008, with pay for performance, I really did save two million dollars. So it's the right thing to do for the patients, and in fact, it really does affect bottom line. So I'll conclude there saying this is a major cause of morbidity and mortality, yet it is preventable with relatively easy minor measures that are mostly behavioral in nature. You have to get people to change their behavior. I thank you.
[ Applause ]
Dr. Gregory Botz:
Thank you, Dr. Gropper. Our next speaker is Dr. Issam Raad. He is the professor and chair of the Department of Infectious Diseases. Please help me welcome Dr. Raad.
[ Applause ]
Dr. Issam Raad:
It is a special honor to be a speaker after two distinguished colleagues, Drs. Dennis Maki and Michael Gropper. What I would like to do is outline the basics that they have highlighted related to central venous catheters. Central venous catheters are widely used in critically ill patients, hemodialysis and cancer patients with an estimate of, as Dr. Maki presented, more than 10 million central venous catheters in United States. Central venous catheters have become the leading source of bloodstream infections especially in those associated with MRSA and also staphylococcal bloodstream infections. CRBSI, that is, catheter related bloodstream infections, are associated with high morbidity and mortality and our opinion at the end of the day these are preventable.
I think it is very appropriate to view the CMS indications that they will not reimburse a hospital for a catheter associated bloodstream infection from the standpoint that there should be a zero tolerance for these infections because they are indeed preventable. From that point, the discussion that I would like and the presentation I would like to have is the challenge of how to make them preventable.
Now, I think we are 40 years late as far as the Time magazine. On February 25, 1966, Time magazine in one of the editorials stated that nearly all experts agree that by the year 2000 all bacterial and viral infections would be eliminated. We are far from it at this point, but we can do something, and the emphasis to do something in terms of prevention should really take a certain magnitude at this point, especially since we are forty years late.
This is a study that recently appeared in JAMA showing the impact of invasive MRSA, it's mortality, the fact that as you well know is becoming major now, the community is becoming a major source, but also highlighting the MRSA bloodstream infections are a major problem. And these as I mentioned, like many other bacterial bloodstream infections are mainly line related. Now this study by Richards and colleagues highlights that.
This is a study by the CDC and that involved more than fourteen thousand nosocomial infections. Again, the leading causes of these nosocomial infections are: Number one, urinary tract infections. Number two is ventilator associated pneumonias. Number three is bloodstream infections. And as you can see here, 87% of the bloodstream infections in the intensive care unit in a large number of ICUs here that included 97 hospitals, 87% of those were line related. So devices are really the driving force for these nosocomial infections, the same is true for ventilator associated pneumonia, the endotracheal tubeing the major source, and 95% of urinary tract infections are urinary catheters. So really from the perspective of the intensivist and very appropriately and also the internist, these devices, particularly central venous catheters have become essential in the care of the intensive care unit patient and the cancer patients as well as the hemodialysis patients.
From the perspective of the infectious disease and the infection control healthcare workers, these devices are basically an open wound. They are a bridge to these microorganisms to migrate from the skin insertion site into the sterile bloodstream, which is a major challenge and a problem. Now I presented some data in critically ill patients. Now let's move to the chronic catheters - the chronic central venous catheters used in cancer patients.
This is a study we published in the Annals of Internal Medicine on differential time to positivity. We used also quantitative blood cultures and we have demonstrated that again. There is another study that recently appeared in the European Journal of Infectious Disease, a clinical microbiology infectious disease and the same finding is that 70% of blood Gram positive bacteremias in cancer patients were definitely catheter related based on differential time to positivity and simultaneous quantitative blood cultures and in hematologic malignancy patients more than 50% of all Gram positive bacteremias were definitely catheter related.
Now I would say definitely because there is another group of probably catheter related not accounted for here. Now when it comes to prevention, if you read the guidelines, I think the new guidelines that Dr. Maki and I are proud of, would probably highlight three major aspects of the 21st century in terms of prevention. In addition to hand hygiene, and looking at the site, there are probably three major interventions that sort of were a by-products of several studies in the 90s. And again Dr. Maki's group and our group were involved. The number one is the maximal sterile barrier precautions during insertion and its impact, and its based on this one study that I will highlight, compared infection between hospital and immunology. There are some other supportive data. The catheter site and the sepsis moving from povidone iodine to the chlorhexidine and Dr. Maki has pioneered this work with his milestone study in Lancet where they are all of antimicrobial catheters. And somehow you look at this kind of tryant, this is a combination of good antiseptic techniques, during insertion and maintenance plus technology. And I think the two compliment each other.
These are the two wings of this eagle, if you may, to fly high. The initial study that we have done was conducted here at M. D. Anderson Cancer Center. I was telling Dr. Maki yesterday, the challenge I had as an assistant professor to conduct this study and many others who were senior thought that it will not show any difference but none-the-less inserting a catheter using maximum sterile precautions with a mask, a cap, a sterile guard and a sterile gown did show a difference in terms of catheter related blood stream infections, colonization and also a strong trend to the prevention of catheter related blood stream infections. This is a factor in terms of antiseptic technique during insertion. Now during insertion and maintenance, the use of chlorhexidine versus povidone iodine, there are eight prospective randomized clinical trials. Again Dr. Maki initiated the larger study and the initial study in Lancet with several others. And then if you look at the meta-analysis there is no doubt that chlorhexidine is superior to povidone iodine the 1-2%. Now we're using the 2% in terms of a prevention of catheter related blood stream infections. So good antiseptic techniques are important. They are essential, they are the foundational.
However on their own they are necessary but not sufficient. They are not easily enforceable, we have to admit that. They are not very durable and you have to follow up in terms of compliance, continuously, not completely prevent infections. You will not be able, in my opinion, to maintain a zero level and zero tolerance just simply with antiseptic techniques. They are important to move you downward, and can save some but not enough lives. And having said this, then is there all of a technology that would compliment, supplement and move you to the zero tolerance.
I mentioned this several times that Dennis Macki has been an inspiration to me and many other investigators. This statement that he put in his study on the silver cuff in 1988 attracted our attention. I joined joint M. D. Anderson in 1989. I was finishing my fellowship then, and I remember clearly this statement that I highlighted on when I read that article: binding antimicrobial catheters may ultimately prove to be most effective technologic innovation for reducing the risk of catheter related infections. And having said this, it is important to realize that what you are doing basically is surface modifying. The pathogenesis of the infection involves at the adherence of these organisms to the catheter surface and forming a strong hold which is a microbial biofilm, you're changing the whole kind of a table around in this kind of warfare against these microorganisms.
You are changing this surface from becoming the focus of infection or adherence to become a fatal bed for these organisms, while they are trying to land on the catheter surface. Not only on the external surface but the lumen. Probably this technology as far as central venous catheter as or intervention, is more study than any other area of prevention that had an impact on CRBSI. They are probably on the spot as chlorhexidine silver sulfadiazine. A total of around 19 studies, most of them are perspective randomized, almost majority of them which show a significant difference in colonization. Several of them shows significant difference in catheter related bloodstream infections.
As far as the immunocycline rifampin there are now nine studies and again I would you show you meta-analysis basically most of them would show a difference in catheter related bloodstream infections. The initial study and again a milestone study on the chlorhexidine silver sulfadiazine was done by Maki and colleagues, This study shows significant decrease in colonization as well as significant decrease in catheter related bloodstream infections by more than four fourth.
Now subsequently, and actually during the same issue of the analysis of [ inaudible ] medicine, we evaluated the minocycline rifampin. And this was based on data that we realize that is easy through a simple technique to impregnate catheters with a combination of antibiotic minocycline rifampin, each of then independently has a potent activity against MRSA and staphylococci, but also it would give you some gram negative coverage against stenotrophomonas and thorough bacter and acinetobacter. We realize that also we can through that the simple technique impregnate the external and the internal surface of the catheter with better zones of inhibition compared to chlorhexidine silver sulfadiazine.
Now zones of inhibitions although they are crude measure, they ultimately do reflect a certain level of efficacy. That study when we compared in perspective multicenter randomized trail, where we compared minocycline rifampin, coated centeral venous catheter is to uncoated catheters, there was a significant decrease in colonization but also a prevention of catheter related bacteremia, a significant prevention in catheter related bacteremia. This subsequent study led by Ravi Darwish from Baylor and our group that appeared in [inaudible] medicine compared the minocycline rifampin coated catheters to the first generation chlorhexidine silver sulfadiazine.
And this first generation coats only the external surface as you will know that is the second generation were both external and internal surface are coated with chlorhexidine. Now in this initial study there was a significant difference in favor of minocycline rifampin, in decreasing colonization and almost 12-fold decrease in bloodstream infections. Initially when Zell and colleagues in their editorial commenting on the study, attributed this difference to the fact that this catheter, the enthemodical catheter coats not only the external surface unlikely the first generation antiseptic catheter, but also in the internal surface.
Subsequently I think there are multiple factors that accounted for this difference. I think number one factor is understanding this consult of biofilm and the potency against biofilm forming organisms. When organisms are trying to land on the catheter surface, they are coming as free floating organisms. Soon they try to attach themselves to fibrin, fibrinetic. Soon after that, within hours to days, they start forming this matrix, extracellular polysaccharide matrix, that is rich in calcium and iron and becomes like a city wall, like a bullet proof jacket were these organism are well embedded, well protected behind their barracks. Basically antibodies cannot penetrate very well and cannot eradicate these organisms in that setting. So before they build their own stronghold it is prudent to find that more defined the surface of the catheters with antibiotics without active against this biofilm forming organisms.
This is for example some of the data that now we presented in 2007 in a biofilm colonization, heavy biofilm colonization model. When you exposure catheters that are uncoated there was a low activity to a high colony count of MRSA, VRSA, Stenotrophomonas, multidrug resistant Stenotrophomonas maltiphilia, Acenatobacter. You build a large number of organisms in biofilm on the catheter surface. When you expose them to clotted catheters you decrease this kind of adherence significantly. With aminorifampin they are so potent and you what you end up zero adherence. And you find the same trend actually you find a more significant difference between these two Stenotrophomonas, Acenatobacter and so forth.
So these are the advantages of surface modifications because this allow this organism from forming the biofilm and adhering to a catheter surfaces, and you protect that surface and make it anti-infective. The other reason for this difference is this a prolonged antimicrobial durability of the minocycline rifampin coated catheters compared to not only the first generation but also the second generation antiseptic catheters. That is why 30 days you maintain a large zone of inhibition and strong antiadherence activity even in serum of these dwelling catheters.
Now this is reflected clinically and the studies, for example, Logi and colleagues when they compared uncoated catheters to the first generation antiseptic catheters and leukemia and lymphoma patients with a mean duration of 20 days. There was no difference. The reason is because the mean dwell time was 20 days rather than seven days which is the, where these catheters do very well. Where as in our cancer patient population when the minocycline rifampin silicone catheters were compared in a prospective randomized study to the uncoated catheters, there was a significant difference in bloodstream infections by almost five-fold. And this significant difference occurred in the catheters that had an in dwelling time of around two months. So this, we sort of played to the strength of these catheter, and we realized that these catheters have a prolonged antimicrobial durability. Why not use them in these long-term settings where we know that most of the infections occur in long-term catheters within the first two months.
And hence in an elegant study by Ravi Darwish and again our group where they compared in a prospective randomized trial tunneled uncoated central venous catheters these are typical Hickman catchers to just percutaneous nontunneled but minocycline rifampin coated catheters. Again this was another study that we expected equivalence. And to our surprise from a significant reduction in catheter related bloodstream infections and which correlates very well with antimicrobial durability of two months. If you look at a meta-analysis by Falgas and colleagues of all the prospective randomized a clinical trials there is no doubt that these catheters decrease colonization as well as decreasing the risk of catheter related bloodstream infections and most of these studies are independently positive by multi-variant analysis. Now we come back to the Pronovost study which I think it is very interesting and very insightful.
What they did in the Pronovost study that was discussed by Dr. Gropper, they looked at these basic antiseptic techniques, the five criteria particularly had in there the maximum sterile barrier and the chlorhexidine at the insertion site. And what it did is they moved. It is a crossover study evaluating the baseline rate before they initiated these criteria and then after 18 months or after three months and then 12 months, and finally after 18 months there was a significant decrease in the mean catheter related bloodstream infections from 7.7 per 1000 catheter days to 1.4 per 1000 catheter days, which is impressive.
Now I agree with Dr. Gropper as to the fact that the study leaves a lot of questions unanswered. There is no data on compliance of the definitions of a catheter related bloodstream infections. Some are and includes one positive blood culture of dye negative staff. There is some data doubt there could be antimicrobial catheter introduced during this time period. There are many other issues are needed to be addressed. But I personally believe this data, because this reflects our experience. Since we were pioneers in introducing the maximum cell barrier and we adhere to the chlorhexidine very early, we implemented these measures in our institution at a very early stage.
Possibly this comparatively might give you some explanation. The studies in white are this study lead by M. D. Anderson. Most of them are a multicenter trials except for this one, and this is a prospective randomized trial. When we compared the uncoated catheters to coated minocycline coated catheter in the mid 90s and published it in 1997, we were not using routinely the maximal sterile barrier. It was a multicenter trial. We would not using chlorhexidine, and the base line rate of 7.3 per 1000 catheter days was almost an identical to the baseline rate of the Pronovost study.
Now what happened here is following in the year 2000 and beyond we started applying in the uncoated group in these studies, we started applying the maximal sterile barrier routinely and the chlorhexidine application, and what you find is our rate in the control uncoated group with a antiseptic technique being applied is almost identical to where they have the endpoint of the intervention in the Pronovost study. This is identical because these are the measures that were taken under controlled arm. But if you look all through in all of this situation or all of our studies where we used the minocycline rifampin catethers, we were even going further than 1.4 per 1000 days or 1.3 to a level of less then 0.5.
The message following is that if you combined the two good antiseptic techniques plus a novel technology such as coating the catheter, then you are able to aim more and come closer to the zero tolerance level, which we should aim at. Now having said this, it is important to kind of highlight it from our experience, this is based on the data by Hannah and colleagues. Just on these two years where after introducing the antimicrobial catheter in the ICU, there was a significant decrease that continued to go down then we stopped for few years, and just reinforced basically some of the other measures and antiseptic techniques now and then, it has started going upward. Then what he enforced were these measures and maintained the level of around one per 1000 catheter days, and most of the infection in this group are with uncoated catheters that end up being used in the ICU such as the pick line.
So again what you see here and the trend is that when you use both the antiseptic techniques and also the antimicrobial catheters, a novel technology, you achieve a very well level and you can even bring it further. In the study by Hannah and colleagues, she has shown in the intensive care unit and in collaboration without critical care department that there was not only decrease in the days of ICUs significant decease of morbidity and the days of ICU and [inaudible] infections but also significant degrees and multi drug resistant organism such as VRE, because these antimicrobial catheters are active against VRE as well as VRSA. There was also a significant cost saving of all most 1.4 million dollars annually and the cost saving has been shown by Andy Shore, in his analysis where use of such catheters would save your patient compared to the uncoated catheters of around 277 dollars.
Now this morning I was looking with Elizabeth Ramuz, a second to your follow, at the rates of resistance occurring to rifampin. We looked over 10-year period. At the baseline data that we had and the susceptibility of all staphylococci. We are talking about thousands of organisms, actually tens of thousands of organisms. Staphylococci, MRSA and negative staff tested for rifampin and tetracycline to the time when we introduced these catheters at M. D. Anderson and now almost eight years later, and there is basically no difference. In fact if anything the susceptibility is better to rifampin and almost equivalent to tetracycline above 90%.
After 10 years of using these catheters at M. D. Anderson extensively and almost using them intensely in the intensive care unit, I think the previous guidelines by CDC will highlight this issue very well that you have to do your best to emphasis infection good infection control measures. Good hand hygiene and implement good antiseptic techniques maximal sterile barrier, chlorhexidine at the insertion site. And at the end of the day compliment this in high risk patient such as ICU and the patient were the catheter is going to stay more than five days with antimicrobial catheters and novel technology.
I did not touch base on the lock solution here because of lack of time, but I would like to say that the future is in combining the technology with good antiseptic techniques and I've quoted Dr. Maki here who has been inspiration to us. But I think the trend in the future is to think of how we can improve these technologies the same way like the airline in this improved technologies in order to make a plane crash a very uncommon and rare undesired complication. Now where can be go better I think minocycline rifampin we have to have add antifungal agent to prevent not only catheter related bacteremia but catheter related candidemia. With chlorhexidine we have to think of methods and we have developed certain methods of combining it for example to junction violet to prolong to say antimicrobial durability because a durability is an issue of the antiseptic catheter and this will have also important implications in impregnating endotracheal tube and the urinary catheter. I would like to stop here and thank you very much.
[ Applause ]
Dr. Gregory Botz:
Thank you Dr. Raad. We have a few moments for questions if anyone might like ask questions to one of our speakers. Yes?
[ inaudible question ]
Dr. Raad:
Yes, in a way. I hate to use the term, but at least antimicrobial catheters could be sort of a fool proof it may in other words when there is a failure to adhere to good antiseptic behavior, there is a shock absorber at the end the day. But I think ultimately if you do not use good antiseptic techniques you are going to be bombarding them with high inoculum of organisms and then you going to allow more for emergence of resistance. So this is why I think in a bigger sense you have to kind of fuse the two technique together.
Dr. Gregory Botz:
Any other questions? I think with the lack luster success we have in hygiene in this institution we are going to have to find a way to impregnate hands of healthcare workers [inaudible] as a strategy. I'd like to thank all of you for attending our Grand Rounds, and thank our speakers for some delightful presentations. Thank you.
[ Applause ]
©2007 The University of Texas M. D. Anderson Cancer Center
1515 Holcombe Blvd, Houston, TX 77030
1-800-392-1611 (USA) / 1-713-792-6161