a b s t r a c t

An estimated 6000 patients who are undocumented immigrants have end-stage renal disease (ESRD) and rou- tinely present to public safety-net hospitals for life-saving emergent dialysis treatments. Because these patients lack a dialysis unit, they often do not have access to medication management consistently coordinated by a ne- phrologist, and this can result in more frequent emergency department (ED) utilization and cost of care. We hy- pothesized that patients who were taking Loop diuretics had fewer ED visits for Emergency dialysis. Loop diuretics can potentially take advantage of residual renal function and mitigate excess fluid gain that can induce heart failure and high potassium, the two most common indications for emergency dialysis. In our univariable analysis, patients on furosemide had 3.1 fewer ED visits on average compared with patients who are not on fu- rosemide. After adjusting for vintage and serum potassium measures, the average number of ED visits was about 1.1 visits less in furosemide-treated patients compared with patients not receiving furosemide (95% con- fidence interval, -4.4 to 2.1). These results suggest that loop diuretics may have an important role in undocu- mented patients with ESRD with residual renal function. Further study to develop practical approaches to the care of undocumented patients with ESRD is greatly needed.

(C) 2016

Introduction

According to formal surveys, nearly 11 million undocumented immi- grants reside in the United States [1]. The health of undocumented im- migrants, specifically those with renal disease, is an important Public health concern because of the pressure patients place on city and county health care systems that often care for the undocumented population. An estimated 6000 patients have a particularly Severe form of renal dis- ease called end-stage renal disease (ESRD) [2]. Patients with ESRD rou- tinely present to public safety-net hospitals for life-saving emergent dialysis treatments, often with severe symptoms or life-threatening electrolyte abnormalities. Because these patients lack a dialysis unit, they often do not have access to medication management consistently coordinated by a nephrologist. This situation can result in more frequent emergency department (ED) utilization and cost of care should they subsequently require hospitalization or ICU care.

As part of quality improvement initiative regarding the care of un- documented patients with ESRD, we collected data on the frequency of visits and the medications taken by undocumented patients with ESRD. We hypothesized that patients who were taking diuretics,

? This study was funded in part by the Selzman Institute for Kidney Health, Baylor Col- lege of Medicine, Houston, Texas.

* Corresponding author at: 1400 Pressler Ave.,FCT 13.6, Houston, Texas 77030. Tel.: +1 713 745 0105; fax: +1 713 798 3510.

E-mail address: [email protected] (B. Workeneh).

specifically a potent variety called loop diuretics (ie, furosemide or bu- metanide), had fewer ED visits for emergency dialysis. Loop diuretics can potentially take advantage of residual renal function that patients may have and control excess fluid volume that can induce heart failure and high potassium, the two most common indications for emergency dialysis. Patients often take diuretics before starting dialysis, but once they are on dialysis, these drugs may be discontinued because there is little data to support its use. For a 3-month period, we carefully reviewed charts of undocumented patients who presented for at least three emergent dialysis treatments to the Ben Taub General Hospital, a centrally located county facility in Houston, Texas, that cares for the bulk of undocumented patients in the city. To our knowledge, manage- ment of undocumented patients with ESRD and the effect of medica- tions have on ED utilization have not been studied.

Methods

The study was reviewed and approved by the Baylor College of Med- icine institutional review board. We reviewed charts of all undocument- ed patients who presented for at least three emergent dialysis treatments to the Ben Taub General Hospital between June 1, 2013, and August 31, 2013.

We abstracted prescription data for furosemide and also the follow- ing commonly prescribed medications: oral alkali solutions, ?-blockers, sevelamer, and erythropoietin stimulating agents (ESAs). We also

http://dx.doi.org/10.1016/j.ajem.2016.05.042

0735-6757/(C) 2016

S. Ahmed et al. / American Journal of Emergency Medicine 34 (2016) 1552-1555 1553

abstracted data regarding the number of ED visits, the number of hemo- dialysis sessions, and mean, median, and peak potassium values.

Data from all patients who were emergently dialyzed during a 3- month period were analyzed. Patients were dialyzed if they met clinical criteria for acute dialysis (ie, hyperkalemia, severe metabolic acidosis, symptomatic volume overload, and anemia requiring transfusion) as judged by the ED physician. Patient characteristics are summarized using mean and SD or frequency and percentage. Fisher exact test and independent, two-sample t test were used to determine differences in ED frequency. The effect of diuretics on ED frequency was assessed using a multiple regression model incorporating the concomitant use of medications including ESAs, ?-blockers, and oral alkali agents.

Results

The analysis included 88 patients; 4 patients were excluded because they had less than 3 visits during the 3-month observation period, and 1 patient was excluded because there was insufficient laboratory data available in the patient’s chart (Fig. 1). The patients studied were those who presented on multiple occasions seeking emergency dialysis treatment. Regarding ethnic background, the patients were over 90% Hispanic in origin. The etiologies of kidney disease are listed in Table 2. Referral to dialysis was made as a standard computer- generated order that read in part: “hyperkalemia, uremia, and volume overload” in all patients included in the final analysis.

All patients in this cohort were using furosemide in particular as the

loop diuretic; 71 patients received furosemide, whereas 17 patients did not. Patients had no significant difference in initial systolic or diastolic blood pressure; heart rate; potassium or bicarbonate values; or hemo- globin values (Table 1). Patients were categorized as having systolic congestive heart failure (CHF) if they had reduced ejection fraction noted on echocardiogram. Patients were categorized as having diastolic CHF either if the patient’s chart noted a diagnosis of CHF with preserved ejection fraction or if the patient’s chart noted a diagnosis of CHF (unspecified) and the echocardiogram report noted a normal ejection fraction.

In the univariable analysis, patients on furosemide had 3.1 fewer ED visits on average compared with patients who are not on furosemide (Fig. 2). After adjusting for dialysis vintage and serum potassium

Table 1

Demographics and clinical characteristics

Variable No Lasix (n = 17) Lasix (n = 71) P?

Demographics

Female 9 (53%) 35 (49%) 1.00

Age (y) 44.5 (13.5) 48.0 (14.2) .36

Systolic BP (mm Hg) 156.2 (16.7) 152.9 (17.2)

Diastolic BP (mm Hg) 87.7 (16.0) 81.8 (14.0)

Heart rate (bpm)	70.5 (7.4) 71.7 (10.9)
Potassium (mg/dL)	5.5 (0.8)	5.4 (0.6)
Bicarbonate (mg/dL)	21.5 (3.0)	21.5 (3.4)
Hemoglobin (mg/dL)	8.8 (0.7)	8.8 (0.6)
ED visits CHF diagnosis	20.4 (9.1)	17.3 (6.2)	.10
No	2/12 (17%)	2/63 (3%)	.18
Diastolic	5/12 (42%)	33/63 (52%)
Systolic Medications Erythropoetin	5/12 (42%) 10 (67%)	28/63 (44%) 63 (89%)	.05
Sevelamer	13 (76%)	67 (94%)	.04
?-Blocker	7 (41%)	47 (66%)	.09

Abbreviations: BP, blood pressure; bpm, beats per minute.

* P value comparing Lasix versus no Lasix groups using Fisher exact test or independent, two-sample t test.

measures, the average number of ED visits was about 1.1 visits less in furosemide-treated patients compared with patients not receiving furo- semide (95% confidence interval, -4.4 to 2.1). However, there was no significant relationship between furosemide total daily dose and mean, median, or peak Potassium levels. Patients on diuretics were more likely also to be taking ESAs and sevelamer (P<= .05). The use of oral alkali agents, ?-blockers, and sevelamer was not correlated with fewer ED visits.

Discussion

Currently, patients receive prescribed medications from their prima- ry care physicians, ED physicians, and nephrologists who attend acute dialysis rounds, and there is no uniform practice regarding diuretic management. Information about use of ?-blockers, which confers mor- tality benefit especially in patients with documented heart failure [3],

Fig. 1. Study flow diagram.

1554 S. Ahmed et al. / American Journal of Emergency Medicine 34 (2016) 1552-1555

30

25

20

15

10

5

0

No furosemide

Furosemide

Total ER visits

Fig. 2. Number of ED visits among the no furosemide and furosemide groups.

and phosphate binders was collected because it may decrease symp- tomatology and as an indicator of how much patient coordination the patients received. The ED providers impart no information regarding di- alysis follow-up, and patients drive the frequency of ED visits and refer- ral to dialysis is determined by prescribed criteria. This study was initiated because loop diuretics could potentially mitigate 2 major com- plications of ESRD, hyperkalemia and volume overload, that frequently necessitate emergency dialysis. We observed that loop diuretics could potentially be beneficial in this respect and suspect that if we were able to determine residual renal function that the difference in ED visits would be significant. Residual renal function is an important determi- nant of diuretic response, and those who have no or marginal residual function with urine volumes below 100 mL may not benefit from diuretics [4]. We speculate a larger sample size controlling for residual renal function would reveal a significant independent effect of loop diuretics.

Our review also revealed that there may be an independent associa- tion between ESA use and decreased ED utilization. This may be ex- plained by increased doctor-patient time, less symptomatology while on ESAs, or other mechanisms including reduced blood pressure, re- duced left ventricular hypertrophy, and optimization of heart failure [5,6]. These factors should also be investigated in future studies.

The cause of ESRD was not unequivocally determined in most cases in our cohort but is made presumptively based on presence or absence of comorbidities in many cases, as is the case in most patients in the United States who develop ESRD. The causes vary, and in some cases, slower progression of renal disease can result in an additional benefit from diuretic use to control potassium and volume. In the population

Table 2

Etiology of ESRD in the study population
Etiology of kidney disease	No. of patients
Hypertension	46
Diabetes	38
Idiopathic/unknown	28

Glomerulonephritis, other than lupus nephritis 4

Lupus nephritis 2

IgA nephropathy 2

Nephrolithiasis 1

Abbreviation: Immunoglobulin A, IgA.

we studied, the cause of ESRD in over a third of the patients was listed as unknown or idiopathic (Table 2), and given their patterns of immigration from Central America, many may have Meso-American ne- phropathy. Meso-American nephropathy is becoming a more recog- nized cause of ESRD among Latin Americans, in particular among those who have worked in agricultural industries [7]. The cause of ESRD is not determined (eg, with biopsy or serology) even among pa- tients who have full access to health care and is most commonly pre- sumed based on comorbidities.

In many states and municipalities across the United States including Texas, there is a vibrant debate regarding the provision of life-saving treatment for uninsured, undocumented patients with ESRD. Since 1972, the Center for Medicare and Medicaid Services has ensured that citizens and legal residents who require dialysis have access to regularly scheduled dialysis. However, uniform public policy and formal funding mechanisms are absent regarding the management undocumented pa- tients with ESRD. It is in this complex environment that physicians on the frontlines at indigent care hospitals have to deliver care to undocu- mented patients with ESRD. In their role as advocates for patients, phy- sicians are left to develop approaches and best practices for the management of these vulnerable patients. Establishing best practices can help communities with a large population of patients ineligible for maintenance dialysis better care for patients avoiding serious complica- tions and manage costs until a permanent solution can be developed.

There are some limitations to this study because of its retrospective nature, but a randomized, controlled trial involving these vulnerable set of patients would have substantial logistic and ethical challenges. An- other limitation was information about residual renal function (eg, 24- hour Urine collections) was not available. Differences in residual function would have likely correlated with the effectiveness of loop diuretics. Furthermore, additional medications that may contribute to the health of patients, such as ?-blockers, ESAs, and oral alkali, were explored in this study, but the sample size was too small to detect differences.

Conclusions

The management of uninsured, undocumented patients with ESRD is a difficult challenge to physicians and health care systems. Novel ap- proaches to delivering quality care while minimizing costs and improve outcomes are greatly needed. We have not established clear evidence

S. Ahmed et al. / American Journal of Emergency Medicine 34 (2016) 1552-1555 1555

for a benefit of loop diuretics in undocumented patients with ESRD, but suggest that providers consider the approach in patients with residual renal function with the aim of mitigating adverse complications associated with ESRD. If possible, we also recommend appropriate use of ESAs, which can help limit symptomatology and need for transfu- sions. Ultimately, undocumented patients with ESRD require more study so consensus management guidelines can be developed.

References

1. Baker B, Rytina N. Estimates of the unauthorized immigrant population residing in the United States. Department of Homeland Security; 2012.
2. Raghavan R. When access to chronic dialysis is limited: one center’s approach to Emergent hemodialysis. Semin Dial 2012;25:267-71.
3. Cice G, Ferrara L, D’Andrea A, D’Isa S, Benedetto A, Cittadini A, et al. Carvedilol in- creases two-year survival in dialysis patients with dilated cardiomyopathy. J Am Coll Cardiol 2003;41(9):1438-44.
4. Medcalf JF, Harris KP, Walls J. Role of diuretics in the preservation of residual renal function in patients on continuous ambulatory Peritoneal dialysis. Kidney Int 2001; 59:1128-33.
5. Thomas J, Teitelbaum I. Preservation of residual renal function in dialysis patients. Adv Perit Dial 2011;27:112-7.
6. Kumra R, Bargman J. A review of diuretic use in dialysis patients. Adv Perit Dial 2014; 30:115-9.
7. Correa-Rotter R, Wesseling C, Johnson RJ. CKD of unknown origin in central America: the case for a Mesoamerican nephropathy. Am J Kidney Dis 2014;63: 506-20.