Predictive factors of adverse clinical outcomes among patients receiving emergency hemodialysis in emergency department

Introduction

Emergency hemodialysis (HD) is defined as an urgent initiation of dialysis within 24 hours of the patient’s presentation with a life-threatening condition (1). Factors associated with emergency HD are patient-related factors (e.g., non-compliance, previous refusal of dialysis, drug toxicity), systemic illness (e.g., sepsis, cardiovascular, peripheral vascular disease, autoimmunity, other systemic causes with secondary renal involvement), and accelerated decline of renal function associated with volume overload, symptoms of uremia or hyperkalemia and significantly elevated serum creatinine on routine blood testing (2).

Acute on chronic kidney disease (CKD) is one of the commonest condition attending emergency department (ED) requiring emergency HD. It has been reported that patients with CKD had higher rates of ED presentation as compared to the normal population. The burden of HD machine use for medically-indicated condition among CKD patients, either known or undiagnosed, is significant (3). In the hospital setting, emergency HD is commonly provided at intensive care unit (ICU), wards, or HD units in which the decision for HD is mainly made by nephrologist while the machine is operated by nurses specialised in dialysis (4,5). It is uncommon for tertiary hospital to provide HD service in ED. Increase in referrals and demands for emergency HD at the study center resulted in decision made by the hospital administrator to set up permanent emergency HD service that is located in a critical zone of the ED. Many of those cases attended ED requiring HD for variety of reasons were in critical conditions but however access block to wards and HD unit had resulted in delay in the provision of emergency HD (6). This has in turn lead to increase morbidity and mortality among this cohort of patients.

Despite logistic advantages of performing emergency HD in the ED, it is unknown whether there are significant differences in the clinical characteristics and HD outcomes between patients receiving emergency dialysis in ED and non-ED setting. This local data is important in future practice that may either support or reject the proposal of locating HD service in ED setting. This study looked at predictive factors that were associated with adverse clinical outcome such as 60-day mortality, ICU admission and need for mechanical ventilation. New data gained may contribute toward improving the care of acute renal failure patients requiring emergency HD in ED. We present this article in accordance with the STROBE reporting checklist (available at https://jeccm.amegroups.com/article/view/10.21037/jeccm-23-165/rc).

Methods

Study design and setting

This was a retrospective observational study for all cases of emergency HD carried out within 24 hours of ED attendance from of a purposively selected Emergency Department of Hospital Universiti Sains Malaysia, a tertiary university hospital with annual average ED attendance of 90,000 patients. The data was collected for patient attendance that met the inclusion criteria and sample calculation within a period of one year commencing January 2021 till December 2021.

Selection of participants

The source population in this study was all medical based patients who were admitted through the ED and underwent emergency HD either in ED or in non-ED setting within 24 hours of admission (Figure 1). We took consideration for all medical related cases that required emergency HD such as acute on CKD, toxicology, sepsis, shock and others. The exclusion criteria were those who received HD treatment before attending to ED and trauma cases. The sample size calculation is based on the primary and secondary outcomes by using the two-proportion estimation using Power & Sample size (PS) software. The required sample size was between 76 per group (ED & non-ED group settings). Since the number of eligible populations was small, we conducted total population sampling, in which, all 177 patients were taken as participants. The ratio between participants who underwent emergent HD at ED (n=91) and non-ED (n=86) settings is about 1:1.

Figure 1 Hemodialysis set up in critical bay of emergency department (consent taken from relative AA for photo use in publication).

Measurement

The emergency HD patients electronic registry and record were retrieved from the HD unit of study location. Data was collected from the patient’s folder using the data collection form and was performed by the first author and verified by co-authors. The study variables include sociodemographic profile (age, gender, ethnicity), health profile (hypertension status, diabetes mellitus status, history of HD), presenting complaints (shortness of breath, lethargy, chest pain, high potassium, vomiting, fever), diagnosis in ED (fluid overload, acute kidney injury, hyperkalemia, severe metabolic acidosis), indication for HD (refractory fluid overload, severe hyperkalemia, hyperuricemia, pH less than 7.1), location of HD (ED, non-ED). Bias in data recording was minimized by appointing one investigator to locate and gain variables from folders of patients from medical record and verified by the principal investigator.

Outcomes

The clinical characteristics include presenting complaints, diagnosis in ED, and indication for HD, while non-clinical characteristics include socio-demographic profile and health profile. The primary outcome was 60-day mortality and secondary outcomes were any adverse event following the emergency HD such as length of hospital and ICU stay, nosocomial infection and need for mechanical ventilation. The null hypothesis predicts similar prevalence of 60-day mortality and other adverse events between the two groups receiving HD (ED & non-ED groups),

Statistical analysis

Data was analyzed by using Statistical Package for Social Sciences (SPSS) software version 24.0. Univariable data were descriptively presented in terms of frequency or percentage. Hypothesis testing was conducted using Pearson’s chi-squared test to statistically compare the clinical and non-clinical characteristics and HD outcomes between patients who underwent emergency HD at ED and non-ED settings. Binary logistic regression was used to analyze the association factors for the primary and secondary outcomes for selected independent variables whilst controlling the confounders. The independent variables include socio-demographic profiles (gender, age, ethnicity), the health profile (hypertension, diabetes mellitus, history of HD), presenting complaints (shortness of breath, lethargy, chest pain, high potassium, vomiting, fever) and ED diagnosis (fluid overload, acute kidney injury, hyperkalemia, severe metabolic acidosis). The outcomes dependent variables were mortality within 60 days, ICU admission and need for mechanical ventilation. Statistical significance was set at P<0.05. Missing data of patient management will be double checked with the primary team record keeping at the HD unit and it will be excluded from study if crucial data missing involving primary and secondary outcomes.

Ethical considerations

This study was approved by the human research ethics committee of study center (JEPeM/18050236). The study protocol conformed to the ethical guidelines of the revised version 2013 Declaration of Helsinki as reflected in a prior approval by the human research committees. No written consent is needed in this study. The data was obtained from medical record unit and was approved by the director of the hospital of the study center.

Results

The total number of patients was 177. The majority of them aged between 40 to 65 years old (56.50%) followed by elderly aged more than 65 years old (36.72%). Male (55.37%) predominates the study samples. 63.28% of participants reported having both diabetes mellitus and hypertension, while 27.68% only had hypertension without diabetes. Most of them were new patients without a history of HD (60.45%). The majority of them complained of having shortness of breath (59.32%), lethargy (72.88%), vomiting (75.14%) and fever (67.23%). No missing data for outcomes for all the selected participants.

Table 1 demonstrates a baseline comparison of the sociodemographic and health profile of the patients between the group of emergency HD that was performed in the ED (“ED group”) and in the ward/HDU/ICU/HD unit (“non-ED group”). Most of the patients’ baseline characteristics were similar between the groups, except for the history of HD. ED group had a significantly higher proportion of new cases (i.e., no known history of HD) as compared to the non-ED group. Tables 2,3 demonstrate the clinical characteristics and outcomes comparison of emergency HD between the ED and non-ED groups. There is a significant difference in two outcome variables between ED and non-ED group. The proportions of 60-days mortality and of those who received mechanical ventilation are significantly higher in the ED group as compared to the non-ED group.

After controlling both clinical and non-clinical characteristics, HD performed in ED [adjusted odds ratio (OR) 2.7; 95% confidence interval (CI): 1.12, 6.40; P=0.03], severe metabolic acidosis [adjusted OR 3.01 (95% CI: 1.29, 7.00); P=0.10], severe hyperkalemia [adjusted OR 2.98 (95% CI: 1.16, 7.60; P=0.002] and mechanical ventilation in ED [adjusted OR 3.53 (95% CI: 1.59, 7.82); P=0.002] significantly predicted 60-day mortality [AUC 0.74; (95% CI: 0.64, 0.84); P<0.001]. Significant predictive factors for ICU admission include those presented with severe metabolic acidosis [adjusted OR 2.78 (95% CI: 1.10, 7.01); P=0.03] whereas performing HD in ED [adjusted OR 0.33 (95% CI: 0.14, 0.79); P=0.01] and known case of regular HD [adjusted OR 0.21 (95% CI: 0.06, 0.70); P=0.01] had lower odd for ICU admission [AUC 0.77 (95% CI: 0.68, 0.87); P<0.001]. Predictive factors for mechanical ventilation among were shortness of breath [adjusted OR 4.95 (95% CI: 2.25, 10.920); P=0.045] and severe hyperkalemia at presentation [adjusted OR 2.95 (95% CI: 1.16, 7.52); P=0.02] with [AUC 0.72 (95% CI: 0.64, 0.79); P=0.048] (Table 4).

Discussion

This study aimed to compare the clinical and non-clinical characteristics between patients who underwent emergency HD at ED and non-ED locations; and to analyze predictive factors of adverse clinical outcome for the whole emergency HD group. There were no significant differences in clinical and non-clinical characteristic between both groups except that those receiving HD in ED had a higher proportion of past history of without HD, being diagnosed as fluid overload and refractory fluid overload. As for the outcomes, there was no significant difference in most of the outcomes of acute HD between the two groups (ED and non-ED) except for mechanical ventilation and 60-days mortality, in which the proportions were significantly higher in the ED group as compared to the non-ED group. Many past studies looked into the significance of known CKD and maintenance dialysis groups of patients on the ED attendance for acute illnesses and its associated outcomes such as mortality, prolonged resuscitation and ICU care (7-9). One can postulate the fact that those patients with known maintenance dialysis tend to have poorer outcomes due to existing illness and its co-morbids such as acute coronary syndrome, diabetes mellitus, sepsis and many others. However very scarce data and study is available looking into the actual outcomes among those receiving emergency dialysis after attending ED especially among those with known end stage renal failure. Report by a study in Taiwan has shown that cases with known CKD with maintenance dialysis had higher ED utilization and a significantly higher risk of resuscitation during ED visits than those without HD. This report however did not specify the needs and logistics of urgent dialysis among these patients (10).

This study suggests that those cases with no history of HD and had a diagnosis of fluid overload tends to receive emergency HD at ED location. The real reason is unknown as no reliable information within the source population is available to make a valid deduction. The most plausible explanation was due to access block to general wards and intensive care units, or due to hemodynamic unstable patient which requiring immediate management and close monitoring in resuscitation zone in the ED. This is in line with the literature that shows undiagnosed CKD patients who presented to ED had worse conditions upon presentation. Similarly, patients with fluid overload are likely to be unstable and require immediate access to resuscitation facility (11,12). Both conditions are life threatening but potentially reversible by medical intervention such as emergency HD especially among patients with refractory fluid overload. It is thus justifiable to have HD facility in ED to intervene acutely on chronic kidney-related complications without compromising immediate access to life-saving resuscitation zone.

As for the HD outcomes, although fluid overload did not significantly influence the HD outcome, those new cases with no history of HD had lower odds of hospital stays more than seven days but higher odds of ICU stays and mechanical ventilation. As for fluid overload, Vaara et al. found that patients with fluid overload at the start of renal replacement therapy (RRT) initiation had a higher risk of 90-day mortality compared to those without (13). There is also growing evidence showing an association between fluid overload and increase risk of mortality (14). Regarding the history of HD, it is consistent with previous literature that reported that those emergency dialysis patients with no history of prior HD are associated with higher morbidity and mortality, which could probably support our finding (15,16). Nevertheless, further study is required to explore this aspect which could be beneficial in data analysis through controlling the confounder and research translation into policy.

Location significantly influences the HD outcomes in terms of 60-day mortality and ICU stays. Those who underwent emergency HD in ED have three times higher odds of 60-day mortality. A study carried out by Sacchetti et al. described the role of emergent HD in ED and stated that ED HD is an option of treatment after initial stabilization (17). From his study, 68% of patients could be discharged from ED after completion of HD in ED. There was no follow up on mortality after discharge or admission to wards and there was no data to suggest high mortality if HD were to be done in ED. On the contrary, they reported three times lower odds of staying in ICU. We may postulate that after stabilization and emergency HD in ED, patients could be weaned off from mechanical ventilator and thus does not require ICU admission (18,19). However, from multivariate analysis, location for HD does not significantly affect the outcome of hospital acquired pneumonia or ventilator acquired pneumonia, duration of hospital stay, and days on mechanical ventilator.

Limitations

This study was one of very few that reported the clinical application of HD in ED setting in comparison with other settings. At least this study had contributed to further insight on the applicability of HD set up in ED for emergency use. Despite its strength to infer causation, this study was subject to several limitations. There was no randomization to the subjects that could lead to selection bias and a threat to external validity. This was unavoidable due to the retrospective design of the study, and therefore, results should be cautiously interpreted. Additionally, retrospective design data verification was difficult, as we had to rely on patients’ records. A retrospective data collection is another limitation for case series detection of 60-day mortality. However, the decision was made by investigators to obtain data which was systematically kept in electronic medical record specifically established by the CKD unit of the study center that contain reliable variables required for the purpose of the study.

Conclusions

Severe metabolic acidosis, hyperkalemia and mechanically ventilated patients were significantly associated with adverse outcomes among patients attended ED with acute renal failure. Hence more precautions should be taken in managing cases with these clinical characteristics requiring emergency HD in the ED.

Acknowledgments

This study was carried out in collaboration with hemodialysis and record unit of the study center. We would like to thank them for their support and assistance given throughout in this research. In particular, to the Consultant Renal Physician (Associate Professor ASA) who had given full support for the conduct of the study.

Funding: None.

Footnote

Reporting Checklist: The authors have completed the STROBE reporting checklist. Available at https://jeccm.amegroups.com/article/view/10.21037/jeccm-23-165/rc

Data Sharing Statement: Available at https://jeccm.amegroups.com/article/view/10.21037/jeccm-23-165/dss

Peer Review File: Available at https://jeccm.amegroups.com/article/view/10.21037/jeccm-23-165/prf

Conflicts of Interest: All authors have completed the ICMJE uniform disclosure form (available at https://jeccm.amegroups.com/article/view/10.21037/jeccm-23-165/coif). The authors have no conflicts of interest to declare.

Ethical Statement: The authors are accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved. This study was approved by the human research ethics committee of study center (JEPeM/18050236). The study protocol conformed to the ethical guidelines of the revised version 2013 Declaration of Helsinki as reflected in a prior approval by the human research committees. No written consent is needed in this study. The data were obtained from medical record unit and was approved by the director of the hospital of the study center.

Open Access Statement: This is an Open Access article distributed in accordance with the Creative Commons Attribution-NonCommercial-NoDerivs 4.0 International License (CC BY-NC-ND 4.0), which permits the non-commercial replication and distribution of the article with the strict proviso that no changes or edits are made and the original work is properly cited (including links to both the formal publication through the relevant DOI and the license). See: https://creativecommons.org/licenses/by-nc-nd/4.0/.

References

1. Chan CT, Blankestijn PJ, Dember LM, et al. Dialysis initiation, modality choice, access, and prescription: conclusions from a Kidney Disease: Improving Global Outcomes (KDIGO) Controversies Conference. Kidney Int 2019;96:37-47. [Crossref] [PubMed]
2. Arulkumaran N, Navaratnarajah A, Pillay C, et al. Causes and risk factors for acute dialysis initiation among patients with end-stage kidney disease-a large retrospective observational cohort study. Clin Kidney J 2018;12:550-8. [Crossref] [PubMed]
3. Komenda P, Tangri N, Klajncar E, et al. Patterns of emergency department utilization by patients on chronic dialysis: A population-based study. PLoS One 2018;13:e0195323. [Crossref] [PubMed]
4. Hooi LS. Dialysis in Malaysia. Med J Malaysia 2006;61:513-5. [PubMed]
5. Meng OL, Seong HL. The National Haemodialysis Quality Standards 2018. Ministry of Health Malaysia. 2018. Available online: https://www.moh.gov.my/moh/resources/Arkib/The_National_Buku_Haemodialysis_Quality_Standards_2018.pdf
6. Baharuddin KA, Abdull Wahab SF, Nik Ab Rahman NH, et al. The record-setting flood of 2014 in kelantan: challenges and recommendations from an emergency medicine perspective and why the medical campus stood dry. Malays J Med Sci 2015;22:1-7. [PubMed]
7. Chow E, Wong H, Hahn-Goldberg S, et al. Inpatient and emergent resource use of patients on dialysis at an academic medical center. Nephron Clin Pract 2014;126:124-7. [Crossref] [PubMed]
8. Harel Z, Wald R, McArthur E, et al. Rehospitalizations and Emergency Department Visits after Hospital Discharge in Patients Receiving Maintenance Hemodialysis. J Am Soc Nephrol 2015;26:3141-50. [Crossref] [PubMed]
9. Moe J, Kirkland S, Ospina MB, et al. Mortality, admission rates and outpatient use among frequent users of emergency departments: a systematic review. Emerg Med J 2016;33:230-6. [Crossref] [PubMed]
10. Lin YC, Hsu HK, Lai TS, et al. Emergency department utilization and resuscitation rate among patients receiving maintenance hemodialysis. J Formos Med Assoc 2019;118:1652-60. [Crossref] [PubMed]
11. Claure-Del Granado R, Macedo E. Indications and timing of renal replacement therapy. Gac Med Mex 2018;154:S15-21. [PubMed]
12. Bagshaw SM, Wald R. Indications and Timing of Continuous Renal Replacement Therapy Application. Contrib Nephrol 2018;194:25-37. [Crossref] [PubMed]
13. Vaara ST, Korhonen AM, Kaukonen KM, et al. Fluid overload is associated with an increased risk for 90-day mortality in critically ill patients with renal replacement therapy: data from the prospective FINNAKI study. Crit Care 2012;16:R197. [Crossref] [PubMed]
14. Vaara ST, Reinikainen M, Wald R, et al. Timing of RRT based on the presence of conventional indications. Clin J Am Soc Nephrol 2014;9:1577-85. [Crossref] [PubMed]
15. Susantitaphong P, Cruz DN, Cerda J, et al. World incidence of AKI: a meta-analysis. Clin J Am Soc Nephrol 2013;8:1482-93. [Crossref] [PubMed]
16. Baer G, Lameire N, Van Biesen W. Late referral of patients with end-stage renal disease: an in-depth review and suggestions for further actions. NDT Plus 2010;3:17-27. [PubMed]
17. Sacchetti A, Stuccio N, Panebianco P, et al. ED hemodialysis for treatment of renal failure emergencies. Am J Emerg Med 1999;17:305-7. [Crossref] [PubMed]
18. Chan MR, Dall AT, Fletcher KE, et al. Outcomes in patients with chronic kidney disease referred late to nephrologists: a meta-analysis. Am J Med 2007;120:1063-70. [Crossref] [PubMed]
19. Palevsky PM. Renal replacement therapy in acute kidney injury. Adv Chronic Kidney Dis 2013;20:76-84. [Crossref] [PubMed]