Hi everyone, my name is Jessie Briscoe. I'm a PGY2 critical care pharmacy resident at Erlinger Health System in Chattanooga, Tennessee, and today we'll be discussing my research on the incidence of AKI in TBI patients treated with hypertonic saline. Before we get started, I'd like to note that neither I nor any of my investigators have anything to disclose. The objective of this presentation is to determine the incidence of AKI in traumatic brain injury patients treated with hypertonic saline. To give you some background on this topic, hypertonic saline is the mainstay of pharmacologic treatment for severe TBI because of its ability to mitigate secondary injury by decreasing intracranial pressure and improving cerebral perfusion pressure. Superphysiologic hyperchloramic solutions have been associated with hyperchloremia and clinically significant events like hyperchloramic metabolic acidosis and acute kidney injury. Previous literature examining the association between acute kidney injury and hypertonic saline has failed to examine non-hypertonic saline chloride sources and has not specifically focused on traumatic brain injury patients. This population is unique due to its demographics, concomitant traumatic injuries, and augmented renal clearance. The primary outcome of this research was to determine the incidence of AKI in TBI patients receiving hypertonic saline and secondary outcomes were the incidence of hyperchloremic metabolic acidosis and the correlation between chloride load and development of AKI. In terms of how we define different outcomes throughout this research, AKI was defined utilizing KDGO serum creatinine based criteria. Hyperchloremia was defined as a serum chloride greater than 115, which aligns with previous literature that's assessed this. And then hyperchloremic metabolic acidosis was defined as an arterial pH less than 7.35 or a venous pH less than 7.31. Bicarb less than 20, serum chloride greater than 115, and an anion gap between 3 and 14. This research was conducted at Erlanger Health System, which is a level 1 trauma academic medical center where our TBI patients are managed exclusively by our surgical critical care service, not a separate neurocritical care service. Additionally, we utilize 3% hypertonic saline that is not routinely transitioned to buffered solution or sodium acetate. With regard to methods for this study, we started by obtaining IRB approval. We then identified patients with a diagnosis of TBI between November 2017 and August 2021 via our institutional trauma registry. We then screened patients who received at least one dose of hypertonic saline for inclusion in the study. We collected data via retrospective chart review and then completed statistical analysis, which we'll talk a little bit more about in a minute. We included patients who are at least 12 years old, admitted to the adult surgical critical care service with a diagnosis of traumatic brain injury, and who received, who had an ICU length of stay of at least 72 hours, and who received at least 24 hours of continuous hypertonic saline, or at least 500 milliliters of cumulative hypertonic saline boluses in a 24 hour period. We excluded patients who had baseline renal dysfunction, which we defined as a creatinine complex, and post-dysfunction, which we defined as a creatinine clearance less than 15, or in-stage renal disease requiring hemodialysis prior to admission. Additionally, we excluded patients who received hypertonic saline solely for hyponatremia and not for management of a traumatic brain injury itself. We looked at over 1,400 patients and excluded about 1,300 of those, with most patients being excluded due to insufficient ICU length of stay or insufficient quantity of hypertonic saline administered. In total, we were able to include 130 patients, and 110 of those did not develop acute kidney injury, that's about 85%, and 20 did develop acute kidney injury, which is about 15% of our population. In terms of statistics, numeric data that was normally distributed was analyzed with a two-sample independent t-test. Non-normally distributed numeric data was analyzed with Wilcoxon rank sum. Categorical data was either assessed with Chi-square or Fisher's exact test. And we completed a logistic regression analysis to analyze the association between maximum serum chloride level and development of acute kidney injury. Now we'll move on to talking about our demographics. These will be broken down into our no AKI and AKI groups. And if you look at age, sex, and body mass index, this was statistically similar between groups. If you take a closer look at age, though, you'll note that the AKI group was about 10 years older than the no AKI group. And while this wasn't statistically significant, it very well may be clinically significant. Moving on to demographics based on injury severity and hospital and ICU length of stay, if we look at initial injury severity, between Glasgow Coma Scale injury severity score and abbreviated injury scale of the head, there were no differences between groups. If we assess Glasgow Coma Scale on discharge, you'll see that the AKI group did have a statistically significantly lower GCS at discharge than the no AKI group, which indicates that there may be a correlation between development of acute kidney injury and worse neurologic outcomes in this population. There were no differences between groups in terms of hospital or ICU length of stay. And there were also no differences between placement of an intracranial pressure monitor or in-hospital mortality. Now, breaking this down even further into our subgroup analysis, both groups received about three days of hypertonic saline therapy. And in the group that did develop acute kidney injury, this typically occurred around day three. Only one of the 20 patients who developed AKI went on to require renal replacement therapy. And 19 of the 20 patients actually were classified as KDGO Stage 1 AKI, which is the most mild form of AKI on the KDGO scale. If we move on to mannitol administration, there were no differences between the two groups in terms of total number of patients who received mannitol, total days of mannitol therapy, or daily mannitol dose. When we looked at concomitant nephrotoxic medications, for this we looked at vasopressors, vancomycin, administration of piperacillin and tezobactam, trimethoprim and sulpamethoxazole, as well as diuretics, there were no differences between groups. And then there were also no differences in terms of the development of diabetes insipidus requiring DDAVP. Now we'll look at baseline and maximum serum laboratory values between groups, starting with serum sodium at the top. There were no differences in terms of serum sodium, and then as we move down you can also see that there were no differences in serum chloride or baseline serum creatinine either. When we look at maximum serum sodium, there was a statistically significant difference with the AKI group having a higher maximum serum sodium level. However, both of the numbers that you see up there are numbers that we sometimes target when utilizing hypertonic saline therapy, so I don't think that this is clinically significant. Moving on to serum chloride, you'll see that the AKI group had a higher maximum serum chloride level and a higher change in serum chloride level, indicating that both maximum and change in serum chloride may be associated with the development of AKI in this population. Finally, ending with serum creatinine, this was higher in the AKI group in terms of maximum and change in serum creatinine, which was to be expected since that was the group that developed AKI. Now when we look at our logistic regression analysis, we found that for every one unit increase in maximum serum chloride level, the odds of having an AKI were 1.076 times higher. Now we have a graph that indicates serum chloride throughout hypertonic saline administration, and just to orient you to that, you have serum chloride on the y-axis, hospital day on the x-axis, and then the orange dots representing AKI, and the blue diamonds representing the group that did not develop AKI, and you'll see that that orange trend line for the AKI group goes up throughout hypertonic saline administration, whereas the blue trend line representing the no AKI group remains relatively flat. throughout hypertonic saline administration. Now we'll look at average daily chloride load. I think this is really interesting because chloride load has been something that has been associated with the development of AKI in other populations. So on the y-axis, you have milliequivalents of chloride, and then we'll start from the left with total chloride load. You can see between the two different groups that this was not statistically significant, and this did include chloride from hypertonic saline and the non-hypertonic saline sources. When we move on to hypertonic saline chloride load, you'll see that this was statistically significant, but it was actually the group that did not develop AKI that received more chloride from hypertonic saline than the group that did not, which does not align with previous literature. Also, when we look at non-hypertonic saline chloride load, this was not statistically different between groups, but I think it is important to note that non-hypertonic saline chloride load, which included things like other fluids, medication diluents, and salt tabs, provided more than 40% of total chloride for both groups, which is significant because this is something that hasn't been considered when looking at chloride load and development of AKI in studies like this in the past. Now, we'll finish with chloride-associated abnormalities. All of these are listed in terms of percent of patients, the actual number of patients developing these things above the bars on the graph. And when we look at hyperchloremia, this was not different between groups, and neither was hyperchloremic metabolic acidosis. So, in conclusion, we find that the incidence of AKI in traumatic brain injury patients treated with hypertonic saline was similar to that seen in other populations, but tended to be relatively mild. We also found that maximum chloride level, rather than chloride load, may be associated with the development of AKI in traumatic brain injury patients treated with hypertonic saline. Additionally, I think that this research shows that it's important to include non-hypertonic saline sources of chloride when analyzing the impact of chloride load, both in our research and when we're treating our patients as well. I'd like to acknowledge my co-investigators, and that's all that I have for you guys. Thank you for coming to my talk.

acute kidney injury

traumatic brain injury

hypertonic saline

hyperchloremia

chloride load