Hypotension during chronic hemodialysis
Intra-dialytic hypotension (IDH) is one of the biggest problems of chronic hemodialysis (CH) treatment. The incidence is especially high in elderly patients and patients with cardiovascular disorders. IDH is a complication of CH that decreases patient comfort and leads to shortening of treatment time (1). Although IDH is one of the most common complications of CH, data on its frequency vary significantly. It is estimated that IDH is present in 25% of all dialysis patients in the US. The percentage is somewhat lower in France with 19% (2).
Hypotension in hemodialysis patients can be acute or chronic. Acute episodic IDH is registered in approximately one quarter of all CH treatments and certain groups of patients are particularly sensitive to this complication. Chronic or recurrent hypotension usually appears in about 5% of CH patients and is more common in patients who have undergone treatment over longer periods (usually, more than 5 years). Such patients are registered at the hemodialysis centre with a systolic blood pressure lower than 90 or 100 mmHg. Due to chronic IDH, the patient’s general condition may deteriorate, i.e. already low values of blood pressure (BP) can suddenly drop to even lower values.
In CH treatment, which removes more than just fluid (isovolemic dialysis), hypotension is very rare. The first association with IDH is the removal of excess fluid from the body. The normal volume of plasma is 40 ml/kg of body weight, or about 2.8 L. Of that amount, the water content in the plasma is about 2.5 L, and the remaining part is cells and dissolved substances. Although 2-5 L are removed during hemodialysis by ultrafiltration that takes app. 4-5 hours, the research shows that the patient’s plasma volume is reduced only by 10-25% at the end of the procedure (3).
Due to kidney failure, uremic toxins that accumulate in the body also affect the hemodynamic stability of the patient because they have the ability to osmotically attract water like a dry substance. As soon as CH treatment starts, substances that have a positive role in maintaining the intravascular volume are removed abruptly. A decrease in plasma osmolality advances the shift of water from the extracellular to intracellular fluid. The increased levels of nitric oxide are registered in patients prone to IDH (4). As the most common complication of CH, IDH can be quickly cured by usual therapeutic procedures, but sometimes it can lead to further complications. The complications of IDH are many. Clinically, the most urgent are the changes that may occur in the heart or brain (5).
Every hypotensive episode also implies a more or less pronounced hypoxemia and asphyxia of heart or brain cells. In a patient with CH, it is necessary to accurately determine the optimal dry body weight (DBW; body mass at the end of CH, at which the patient has not developed symptoms of hypo- or hypertension, does not have leg swelling, or inter-dialytic complications), correctly choose the dialysis set, adjust the temperature and the composition of the dialysis fluid, prescribe the appropriate dose and type of hemodialysis treatment, and apply intra-dialysis on-line control of ultrafiltration according to the monitored values of AT, blood volume and temperature. The objective is to reduce the UF rate so that the patient can use compensatory mechanisms to tolerate premature ultrafiltration more easily (6). Hemodynamic instability during CH results in myocardial and brain ischemia/reperfusion. Recurrent episodes of myocardial and cerebral ischemia and reperfusion result in myocardial stunning (7). Myocardial stunning disturbs the regional mobility of the left ventricular wall and can lead to systolic heart failure while brain damage causes disturbed cognition. Prevention of IDH should include restriction of salt intake, inter-dialytic weight gain in patients with weight less than 5%, ultrafiltration rate less than 10 ml/kg/h, adjustment of sodium and calcium concentration in the solution, cold solution for CH (35-37 °C), use of vasopressors, hemodiafiltration and new dialysis modalities based on biofeedback systems. Preservation of hemodynamic stability of the patient during CH is the primary goal (8).
Acute atrial fibrillation
Heart rhythm disorders, especially acute atrial fibrillation (AAF), occur in 7-27% of all patients and are more frequent in patients undergoing the CH treatment over a longer period of time. AFA can cause hemodynamic changes and myocardial ischemia, which leads to an unfavorable prognosis. The development of AAF in patients on CH is associated with worse survival outcomes and the need for longer hospital treatment. If the appropriate dry body weight and ultrafiltration are underestimated, there is an even greater decrease in systolic and diastolic pressure and the occurrence of hypotension. A drop in BP leads to a decrease in blood flow through the valves and disturbances in the conduction system of the heart, which can lead to AAF (9). The majority of patients who have completed 5 years of CH treatment have a moderate to severe burden on the circulatory system. Cardiac remodelling is the dominant manifestation and may mechanically contribute to hemodynamic instability. Moreover, cardiac remodelling may actively contribute to the development of complications during CH, and, most importantly, to IDH. In the patients on CH, a combination of volume overload (fluid retention, presence of arteriovenous fistula, anemia) and pressure overload occur (Figure 1) (10).
The objectives of this research were to determine the frequency of AAF in cases of IDH and to emphasize the importance of correctly determined dry body weight in patients on CH.
Patients and Methods
At the Hemodialysis Centre in Živinice Health Centre, a randomized study was conducted in the period May 1, 2021 - March 31, 2022, which examined the frequency of IDH and AAF in patients on CH. The total number of dialysis patients in that period was 125, out of which 96 were included in the study. The remaining 29 patients were not included in the study as they had other associated diseases, such as diabetes, complications related to COVID-19 and dementia. All patients included in the study were older than 18 years and on CH treatment three times a week.
The patients were divided into four groups according to gender and incidence of IDH or normotensivity:
Group 1 included 20 men with IDH
Group 2 included 26 women with IDH
Group 3 included 23 men who were normotensive during CH
Group 4 included 27 women who were normotensive during CH.
A 12-lead electrocardiogram (ECG) was performed on all patients included in the study during the last hour of CH, once a week. Patients who had symptoms/clinical signs of cardiac arrhythmia or anginal pain had an ECG performed immediately after the onset of symptoms/clinical signs. All patients were monitored for signs of ischemia on the ECG (changes in elevation and depression on the ST-segment and T wave), regardless of the presence of clinical signs/symptoms of acute coronary syndrome.
All data were processed using the methods of descriptive statistics. The numerical data are presented as appropriate measures of central tendency and variance and displayed in appropriate tables and figures. Non-parametric methods and tests were used to calculate statistical significance. The χ2 test was used to calculate the differences within the groups while the Kruskall-Wallis test was used to calculate the differences between the groups. In case there is a statistically significant difference between the groups, the Mann-Whitney U test was used as an additional testing tool. For parametric data, the differences between the groups were calculated using a one-way analysis of variance (ANOVA), with post-hoc Tukey’s HSD test in case there were differences between the groups. Student’s t test was used for dependent samples. Statistical hypotheses were tested at the α = 0.05 level, that is, the difference between the groups in the sample was considered significant if p <0,05. Statistical processing was performed with the support of the biomedical application software “MedCalc for Windows version 12.4.0”, Copyright © 1993-2013 and “SPSS Statistics 17.0”, Copyright © 1993-2007.
Results
A total of 96 subjects were included in the study, of which 43 were men and 53 were women. This difference in the frequency of male and female respondents is statistically significant in favour of female respondents (χ2= 11.025; df=1; p=0,001).
The analysis of ECG recordings in patients with IDH indicated that 46.5% of patients had AAF. The observed differences are statistically significant (χ2 = 3.974; df = 3; p = 0,26). As part of IDH, 16.6% of patients with AAF also had ischemic changes on the ECG. According to the Kruskal-Wallis test, this difference is statistically significant (χ2= 30,869; df=9; p<0,001). In patients who did not have AAF, only one case of ischemia was recorded on the ECG during the study, which is statistically significant (χ2= 0.775; df = 4; p = 0,001). As part of IDH, 16.6% of patients with AAF also had ischemic changes on the ECG. According to the Kruskal-Wallius test, the difference is statistically significant (χ2= 17.6; df = 3; p = 0,001) (Table 1).
In patients who did not have AAF as a rhythm disorder, only one case of ischemia was recorded on the ECG during the study, which is statistically significant (χ2= 0.775; df = 4; p = 0,001) (Figure 2).
Errors in the underestimated dry body weight in the range of 0.5 - 1 kg result in IDH, and are often associated with the occurrence of AAF during haemodialysis, which can be seen inFigure 2. The Kruskal-Wallis test indicated that the underestimated value of dry body weight in patients has statistical significance (χ2= 1.813; df=2; p=0,4) (Figure 3).
At the time of admission, the mean value of systolic pressure was 107 mmHg, and the value of diastolic pressure was 83 mmHg. By testing the correlation between systolic and diastolic pressure, a one-way correlation was determined (Table 2) (Pearson’s r=0.136; p=0,043).
Discussion
This research study shows that lower values of systolic and diastolic blood pressure are associated with the occurrence of AAF in CH patients. In female patients on CH with IDH, slightly lower values of systolic and diastolic pressure were found, in contrast to male patients. This is at variance with the findings indicating the correlation between BP and AAF in the general population. In a study covering 4.3 million adults in the United Kingdom, each increase of 20 mmHg in the systolic pressure was associated with a higher risk of incident AAF, as was the case with each increase in diastolic pressure of 10 mmHg that is higher than the normal (11). This, generally linear, relationship between systolic and diastolic blood pressure was observed among subjects aged 30 to 60 years, but was much less pronounced in patients over 60 years of age. Similar findings indicating a positive correlation between BP and AAF life expectancy in the general population were reported in a longitudinal study using data from the Framingham Study (11).
In the study conducted by Chang, in the group comprising 17,003 patients, 3,785 developed AAF. Lower systolic pressure before dialysis was associated with a higher risk of AAF, while higher systolic pressure was associated with a lower risk of AAF. Every decrease of 10 mmHg in comparison with the normal systolic blood pressure was associated with the risk of developing AAF. Lower diastolic pressure was associated with a higher risk of AAF (12).
In this study, a higher number of cases of AAF and myocardial ischemia during hemodialysis were also observed in IDH patients, and more frequently in female subjects. A possible explanation for the inconsistencies between the studies in CH patients compared to other groups of patients may be that patients with lower BP prior to dialysis maybe had difficulties with adequate ultrafiltration and achieving post-dialysis extracellular euvolemia (13). In turn, volume overload can lead to larger atrial size and increased pulmonary artery pressure, which is associated with AAF in non-haemodialysis patients. Moreover, there are certain risk factors related to AAF that are specific for the CH patient population, especially the changes in serum potassium and calcium concentrations that often occur during CHT (14).
This study proves that an underestimated dry body weight in the range of 0.5 to 1 kg plays the most important role in the development of IDH, and consequently AAF. Incorrect dry body weight was most common in male subjects. Normotensive patients rarely had AAF and ischemia, which was not statistically significant in this study.
IDH and myocardial remodelling, especially left ventricular hypertrophy (LVH), are closely correlated. LVH not only paves the way for the occurrence of IDH, but actively facilitates the drop in BP during CH, including the occurrence of arrhythmias and myocardial ischemia. There is also evidence to suggest that IDH enhances LVH, creating a vicious circle. It is indeed difficult to explain the interrelationship, but a clearer understanding of the complex interplay between IDH and LVH may help design useful strategies to avoid the occurrence of IDH and LVH. (15)
A decrease in DBW can cause decreased cardiac output and is associated with a number of risks, including IDH. Adequate DBW is of utmost importance in the prevention of IDH. A study by Sinha et al. included 269 patients on CH, 16 of whom had a fatal outcome of cardiovascular origin due to underestimated DBW and subsequent cardiac arrhythmias. This indicates a correlation between DBW and cardiovascular status (16).
The incidence of IDH and AAF in Japan is 24.7%, and in Belgium 27%. The incidence of newly diagnosed AAF during the CH period was 3.0 per 100 patients in Sweden (17). Clinical randomized trials indicate that lower systolic and diastolic pressures are associated with higher risks of cardiovascular events and death (18). This is consistent with this research. These findings emphasize the importance of randomized trials to establish the optimal target BP in CH patients in order to reduce the risks of cardiovascular events, including AAF (19).
Conclusion
The frequency of occurrence of AAF as part of IDH was investigated by analyzing four groups of subjects. Several conclusions were reached. Occurrences of AAF are common in IDH, especially in female patients. Ischemic changes in CH patients are more frequent in patients with AAF. Underestimated dry body weight is the main cause of IDH.