Alactic Base Excess, New Potential Marker Associated for Circulatory Stress Due to Hemodialysis, a Pilot Study

Indroduction: Circulatory stress increases mortality in patients with chronic kidney disease in renal replacement therapy by hemodialysis, the measurement of central venous saturation has been proposed as a useful tool for diagnosis but with limitations. We wanted to evaluate a new marker, the alactic base excess, to be applied in all types of hemodialysis patients to help diagnose this clinical condition. Methodology: An observational, analytical, prospective and longitudinal study was carried out in hemodialysis of the Instituto Mexicano del Seguro Social, in León, Guanajuato from May 2020 to June 2021 by determining the alactic base excess. The association of alactic base excess as a marker of circulatory stress in hemodialysis was proposed as a primary end point and mortality at 12 months was evaluated as a secondary end point. Results: An inverse association was found between the alactic base excess with the initial pH (r= -0.303, p= <0.05) and the nal lactate (r= -0.297, p= <0.05), in addition to bicarbonate (r= 0.593, p= <0.05) and central venous saturation variability (r= 0.304, p= <0.05). In the analysis by subgroups, both lower tertiles had a higher risk of presenting the adverse event (HR= 0.817, [95% CI 0.21 to 3.05], p= 0.763). No association of mortality was found with the rst (HR= 0.95, [95% CI= 0.73 to 1.2], p= 0.687) or nal determination (HR= 1, [95% CI= 0.758 to 1.3], p= 0.99). Discussion: In this study we found that the alactic base excess proposed to diagnostic different types of acidosis is capable of identifying small changes related to circulatory stress, regardless of the chronicity of kidney failure, hemodynamic status or hemoglobin concentration, but not related to mortality. Conclusions: Alactic base excess is potentially useful to evaluate circulatory stress in conjunction with other tissue hypoperfusion markers, however it


Background
The patient with chronic kidney disease has a high mortality (1), especially in patients with renal replacement therapy through hemodialysis (2), both due to the presence of multiple comorbidities and the high cardiovascular risk that de ne each of them, as well as variables typical of dialysis therapy such as ultra ltrate volume and intradialysis hypotension (3). Circulatory stress associated with renal replacement therapy by hemodialysis was proposed as a result of the imbalance between the supply and consumption of oxygen in the tissues caused by the hemodialysis cycle; however, its pathophysiological mechanism has not been fully elucidated (4)(5)(6). Central venous saturation measurement has been the only marker studied with good predictive values for circulatory stress (7)(8)(9). The limitations of the use of central venous saturation are that its evaluation requires a correctly placed central vascular access in the vena cava and that the patient does not have signi cant anemia, which is uncommon in hemodialysis patients.
Just like in sepsis, circulatory stress from hemodialysis could cause damage to the microcirculation, activation of anaerobic cellular metabolism and release of free radicals or pro-in ammatory cytosines, resulting in increased lactate production (10,11). In multiple clinical trials of critical ill patients, has been proposed the usefulness of lactate as an indirect marker of tissue hypoperfusion, being a therapeutic target for resuscitation with good results (12)(13)(14). In the patient with chronic kidney disease, a state of hyperlactatemia and chronic acidosis is recognized due to renal dysfunction for the elimination of nonmeasurable strong anions, so the evaluation of circulatory stress with these biochemical determinants is di cult. In 2019, Gattinoni et al. (15) in the only study published on alactic base excess, they propose this marker to discriminate between metabolic acidosis secondary to lactate or other xed acids such as phosphates and sulfates, with results dependent on renal function. We study the behavior of lactate, acidosis and alactic base excess in patients with chronic kidney disease, undergoing hemodialysis therapy to assess whether this biochemical determinant can be used as a marker of circulatory stress in hemodialysis, being the rst trial of its kind Worldwide.

Study design and approval
An observational, analytical, prospective and longitudinal study was carried out in hemodialysis of the Instituto Mexicano del Seguro Social, in León, Guanajuato from May 2020 to June 2021. The study was carried out with the approval of the local committee for health research 1001 (COFEPRIS 17 CI 11 020 146, CONNBIOETICA 11 CEI 0032018080) and the regulations of the 1964 Helsinki declaration; Prior to the inclusion of the study, an informed consent signature was requested for all participants.

Population
Adult patients in need of renal replacement therapy by hemodialysis were included, without discriminating between the presence of comorbid diseases or indications for renal replacement therapy. A central venous blood sample was taken through the vascular access for hemodialysis, from which a determination of lactate, pH, base de cit, bicarbonate and central venous saturation was obtained. The exclusion criterion was the diagnosis of respiratory failure due to COVID 19 and the elimination criteria included patients who could not complete a hemodialysis session without being possible to reprogram it.
Procedure for sampling.
Before being included in the study, the procedure to be carried out was explained and a request for informed consent was provided to all participants. Venous blood was taken from the central vascular access for hemodialysis, prior to the start of the session, at 90 minutes and at the end of the procedure. The lactate determination was carried out by a colorimetric test (Vitros 5600 from Ortho Clinical Diagnostics) and the analysis of gases in venous blood was carried out in the gasometer (GEM Premier 3000), later the determination of alactic base excess was obtained with the formula proposed by During the procedure, demographic information was obtained from the participants, clinical data included blood pressure, heart rate, oxygen saturation, respiratory rate, temperature, weight, height and body surface area, indications for hemodialysis like ultra ltrate volume, time of therapy, blood ow, dialysate ow, and lter size.

Statistical analysis
The results of the baseline characteristics of the study population are described in proportions for the categorical variables, means and standard deviation for continuous variables. As a primary end point, the association of alactic base excess with circulatory stress in hemodialysis was proposed, for which a hypothesis test was performed using Pearson's correlation, with an alpha value of 0.05, for the general analysis and by subgroups, which were classi ed into tertiles according to the central venous saturation level at the start of hemodialysis. To evaluate the variables by subgroups, a mean difference test (Studet's T test) was performed, grouping the population into the upper tertile and the lower two tertiles in order to preserve the same proportion of patients. As a secondary end point, all-cause mortality was evaluated with a 12-month follow-up which was analyzed using the COX correlation, expressing results in Kaplan meier graphs. The statistical analysis was performed with the IBM SPSS Statics v23.0 Statistical Software.

Results
Baseline characteristics of the study population.
We enrolled 176 patients in hemodialysis of the Instituto Mexicano del Seguro Social from June to December 2020, in which it was not possible to determine central venous saturation in 96 of them secondary to vascular access other than the central one; 18 patients did not agree to participate in the study and 7 more were excluded due to additional peritoneal dialysis. Of the total of 55 patients who met the inclusion criteria, 5 patients from whom complete laboratory data could not be obtained were eliminated. Correlation analysis was performed on 50 of the remaining participants. For the mortality analysis, 11 participants were lost to follow-up.
In the demographic variables, a higher proportion of male users was found, with a mean age of 49 ± 16.7 years. 84% were on subsequent hemodialysis and only 10% presented an emergency for dialysis. The proportion of comorbid diseases and the indications for hemodialysis are exempli ed in Table 1. Association between hemoglobin and alactic base excess The mean hemoglobin level in our population was 9.7 ± 2.28 (95% CI 8.9 to 10.45) and the alactic base excess initial and nal of hemodialysis are shown in Abbreviation SD (standard deviation), IC (con dence interval), Δ Delta (variability) Association between pH, bicarbonate and alactic base excess.
The pH and bicarbonate quantity were evaluated to obtain global information on the acid-base equilibrium state in the study participants, whose means are shown in table 2. A signi cant difference in pH was found between the tertiles, being less in both lower tertiles of central venous saturation (Table 3), however, the state of metabolic acidosis was not re ected in the pH level in most of the participants. On the other hand, the initial bicarbonate levels showed variability between patients, nding a negative association with the nal lactate (r= -0.326, p= <0.05), but it was not possible to determine the cause responsible for lower levels of bicarbonate in each participant. The Pearson test showed an inverse association of the initial pH with the nal lactate levels (r= -0.303, p= <0.05), association with the initial bicarbonate (r= 0.593, p= <0.05) and both values of alactic base excess (initial r= 0.481 and nal r= 0.548, p= <0.05) (Figure 1a, b, c). Final pH and alactic base excess also had an association (r= 0.382, p= <0.05) (Figure 1d, e). The initial alactic base excess values were found to be negative (-1.43 ± 4.09) as well as positive at the end of therapy (table 2), this variable had no signi cant difference between the subgroups. The association study of the alactic base excess yielded signi cant results between the initial pH, the initial bicarbonate (r= 0.973, p <0.05) and at the end of hemodialysis (r= 0.660, p <0.05), as well as the association of the alactic based excess with central venous saturation variability (r= 0.394, p= <0.05) ( Figure 1f) and nal lactate (r= -0.297, p = <0.05).

Association between central venous saturation and alactic base excess.
Although the central venous saturation levels were signi cant for the study and showed a signi cant difference between the subgroups, no signi cant association was found, except for the variability (delta) of central venous saturation (r= 0.304, p= <0.05).

Survival analysis
As a secondary end point, we evaluated all-cause mortality over a 12-month period, nding mortality in 5% of the participants. In the analysis by subgroups, those classi ed in both lower tertiles limited by central venous saturation level ≤ 74% had a greater risk of presenting the adverse event (HR= 0.817, 95% CI= 0.21 to 3.05, p=0.763) in which, the signi cance was clinical ( gure 2). No signi cant relationship between mortality was found with the determination of initial alactic base excess (HR=0.95, 95% CI= 0.73 to 1.2, p=0.687) and nal alactic base excess (HR= 1, 95% CI=0.758 a 1.3, p=0.99) ( gure 3).

Discussion
In the search for a simple and effective marker for the timely recognition of circulatory stress due to hemodialysis, other authors have proposed the measurement of indirect markers of tissue hypoperfusion such as central venous saturation and lactate (7,8). Conversely, the determination of lactate has been evaluated in multiple trials of patients in a critical environment (16,17), especially as a therapeutic objective of resuscitation (18-20), but its usefulness in the hemodialysis patient is unknown.
One of the limitations for the use of central venous saturation and lactate in the hemodialysis patient is the baseline state of metabolic acidosis and chronic hyperlactatemia due to renal dysfunction to achieve physiological balance, for which we wanted to evaluate the usefulness of the determination of alactic base excess rst proposed by Gattinoni et al. (15) in 2019, in which the excess of alactic base, is a useful marker to differentiate a state of acidosis secondary to lactate from that of other types of acidosis due to non-measurable strong ions, in this study lactate levels were independently elevated of the central venous saturation level and were directly proportional to renal function, more negative values were associated with the development of acute kidney injury. As proposed by these authors, in our study we found more negative levels of alactic base excess prior to the start of hemodialysis, being related to a state of acidosis secondary to non-measurable organic ions other than lactate, which means of the applicability of this marker, not only in the context of acute kidney injury.
The levels of excess alactic base were not related to the concentration of creatinine, bicarbonate or lactate, but they were related to the determination of pH, despite the fact that the mean pH in the population of our study was found within normality, which enhances the ability of excess alactic base to identify the origin of the acid base state. It should be noted that there was a signi cant difference between the tertiles, nding a lower pH in those patients with lower central venous saturation of 74%, so the utility of the alactic base excess could be even greater in patients with central venous saturation levels lower than this range. After hemodialysis therapy and due to its compensatory effect, the levels of base excess became very positive in both tertiles.
Another limitation of the applicability of central venous saturation in previous studies (7,8) is the dependence on hemoglobin concentration, losing its prognostic value in patients with anemia (11), one of the strengths of our study is that we consider essential the inclusion of all patients regardless of their hemoglobin levels, because the real clinical environment is that the patient with kidney disease has anemia, often severe (21). In our study we found that the determination of alactic base excess is not related to hemoglobin concentration but is related to central venous saturation, especially with saturation delta, so that together they could be used for the diagnosis of circulatory stress even in patients with severe anemia.
Another application that we found for alactic base excess is its ability to identify circulatory stress early in hemodialysis when the patient develops minimal changes in pH or bicarbonate, despite the severity of the metabolic acidosis with which hemodialysis therapy has been initiated. Hemodialysis, since it was not related to the concentration of these laboratory determinants, so it could be applied for the early detection of circulatory compromise in hemodialysis.

Limitations:
To avoid making a type II error, the sample size should be improved, obtaining the same variables, but this was not possible due to the epidemiological contingency due to COVID 19 and it was considered unethical to expose patients to infection by going to the hospital instance. In our study, no interventions were performed in the hemodialysis settings, because we cannot issue a recommendation on which indications would be best for each type of patient. Due to our methodological process, we cannot draw conclusions about the risk of mortality using this marker. Strengths: In our study, we consider all patients regardless of hemodynamic status, urgency for dialysis or baseline hemoglobin levels in order to nd a marker applicable in all types of population and is feasible in daily clinical practice.

Conclusion
The determination of alactic base excess can favor the diagnosis of circulatory stress by hemodialysis together with other markers such as central venous saturation, even in patients with anemia or severe acidosis.
Declarations * Declaration of con ict of interest: There is no con ict of interest on the part of any of the authors.
* Financing: This study did not receive any type of nancing or sources of support.