ORIGINAL ARTICLES

Choice Reaction Time and Adequacyof Dialysis: A New Application of anOld Method

Suhail Ahmad

University of Washington, Seattle, Washington, U.S.A.

Background: Choice Reaction Time (CRT) is the time it

takes for a subject to accurately respond to a flashing panel

of lights. The CRT has been used to assess the quality of

dialysis in hemodialysis patients and to assess the neuro-

logical impairments in patients with Parkinson’s disease.

Methods: Three groups of end-stage renal disease

(ESRD) patients on three different renal replacement

therapies were tested using CRT: intermittent peritoneal

dialysis (IPD, n¼ 11), thrice weekly hemodialysis

(HD, n¼ 22), and well-functioning kidney transplant

(Tx, n¼ 6). A group of volunteers with normal renal

function (NL, n¼ 12) was also tested.

Results: The CRT was significantly longer in IPD patients

(618� 89 ms) than observed in the other three groups

(p< 0.0001). CRT in HD patients was 461� 50 ms,

which was significantly longer than in Tx patients

(396� 25 ms, p¼ 0.05). However, the CRT in the Tx

patients was no different from the NL (382� 22,

p¼ 0.32). There was a strong negative correlation between

CRT and weekly creatinine clearance in the IPD group

(r¼� .96) and between the dialysis index and CRT in HD

(r¼� 0.79).

Conclusions: CRT may be a useful tool in assessing the

adequacy of dialysis.

Hemodial Int. 2003; 7(2):118--121.

Key words

Choice reaction time, uremia, adequacy of dialysis,

transplantation, dialysis index, weekly creatinine

clearance

Introduction

Multiple organ systems are adversely affected by advan-

cing renal failure and uremia. Severe uremia can cause

generalized encephalopathy and convulsions; however,

subtle changes in brain functions are much more preva-

lent and are more difficult to assess. Renal replacement

therapies improve brain functions, but to varyingdegrees. The improvement may be dependent on the

adequacy of dialysis or the level of renal function of the

transplanted kidney.

Since the 1970s, we have used a test called choice

reaction time (CRT) and have observed that the CRT

in a patient alters with changes in well-being and mental

acuity, but is reproducible in a constant state of health.

For example, in the same patient the CRT is prolongedwhen measured immediately post-dialysis compared to

the pre-dialysis value, the magnitude of the difference

corresponding to the quality of dialysis [1,2]. In past

studies, we have used the delta-CRT (post minus pre-

dialysis CRT) in assessing the impact of a particular

change in dialysis treatment on the quality of the treat-

ment. For example, delta-CRT value was significantly

larger following acetate compared to bicarbonate dialysis[1]. However, the pre-dialysis CRT, reflecting the state of

health in a group of subjects, has not been compared in

different populations to assess its utility in judging

adequacy of dialysis. This manuscript describes previously

unpublished results of a study conducted in the early

1980s, in which the CRTs were compared among hemo-

dialysis, intermittent peritoneal dialysis, and transplant-

ation patients, and volunteers with normal kidneyfunction.

Materials and Methods

Three groups of end-stage renal disease (ESRD) patients

and one group of normal volunteers with normal renal

function (NL, n¼ 12) were evaluated by CRT [1,2]. The

three ESRD groups included patients on intermittent

peritoneal dialysis (IPD, n¼ 11), on hemodialysis (HD,n¼ 22), and those who had functioning renal allografts

(Tx, n¼ 6). The CRT was performed in the morning in

Correspondence to:

Suhail Ahmad, MD, Scribner Kidney Center, 2150 North 107th

Street, Suite 160, Seattle, WA 98133, U.S.A.

email: sahmad@u.washington.edu

118

Tx and NL, and pre-dialysis in IPD and HD, prior to a

mid-week dialysis session.

The adequacy of HD was measured by Dialysis Indexfor vitamin B12 [3]. The dose of peritoneal dialysis was

measured by weekly creatinine clearance. The serum

creatinine for the Tx patients is given in the Table I. All

normal volunteers had normal serum creatinine values.

Choice Reaction Time (CRT) is the time in ms for

patients to correctly identify the color of a randomly

flashing panel. The average of 25 such responses is

taken to ensure reliability and is recorded as the CRTresult in milliseconds. The test actually measures the

delay in a person’s initiation of response after delivery

of an imperative signal [4]. The subject was seated in

front of the CRT instrument in a quiet room, the proced-

ure was explained by the study nurse or the author, and

the patient was allowed to practice before taking the final

test. The subject was asked to keep a button pressed

down with a finger of the dominant hand and watch forthe imperative signal, which was a flash of one of four

possible colors. The subject had to lift his/her hand from

the depressed button and push the button appropriate

for the color that was flashed. The machine measured the

time between the release of the first button and the

correct pressing of the required button. Incorrect

responses and releasing the button before the light was

seen were not considered valid responses. The average of25 such responses was taken as the result of the test. It

usually took about five minutes to complete one test.

Results

The average CRT in normal volunteers was 382� 22

milliseconds. The CRT was significantly longer in bothdialysis groups, being the longest in IPD (Figure 1). The

CRT in HD was significantly longer than in Tx and NL

groups, whereas there was no statistical difference

between Tx and NL.

The average weekly creatinine clearance for the IPD

group was 56.7 liters/week (range 42–74). The dialysis

index for the HD group was 0.99� 0.08. Analysis of

variance showed a strong inverse correlation betweenCRT and dose of dialysis in both IPD and HD groups

(Figures 2 and 3).

Discussion

Reaction times and choice reaction times have a long

history of use in medicine. By the late 19th century,

using a chronoscope, it was possible to measure thedelay in a person’s initiation of response after delivery

of an imperative signal [4]. The choice reaction time was

later distinguished from simple reaction time by intro-

ducing a requirement for the subject to decode varying

imperative signals and respond appropriately. In the

sophisticated instrument used for this study, the impera-

tive signal was one of four possible colors flashed, and

the subject’s accurate responses were noted.Many neuropsychological impairments have been

measured using CRT and simple reaction times [4]. The

abnormal CRT has been described in lead poisoning,

Parkinson’s disease, and other neurological disorders

[4,5]. Reaction time was first used in dialysis research at

Seattle’s Coach House dialysis facility in the late 1970s to

assess the overall quality of dialysis during many studies,

such as comparing acetate vs. bicarbonate dialysis [1] andcomparing the use of oxygen vs. air during dialysis [2].

It was observed that patients had a longer CRT immedi-

ately after hemodialysis compared to that obtained

immediately prior to that dialysis treatment. This delta-

CRT was smaller if patients had fewer intradialytic

problems; conversely, the delta-CRT was longer if qual-

ity of dialysis was poor. For example, in double-blind

studies it was found that the delta-CRT was significantlylonger when patients were dialyzed against acetate

compared to bicarbonate dialysis [1]. Similarly, delta-

CRT was significantly longer when air was used during

dialysis compared to the use of oxygen in a double-blind

study [2].

Abnormal prolongation in reaction time has been

reported in neurological diseases [4]; however, the use

of this test to assess the status of dialysis patients (notdialysis treatments), to our knowledge, has not been

reported. We tested four groups of individuals with vary-

ing degrees of renal and dialysis function in the late 1970s

and early 1980s, and the results show that the shortest

CRT was observed in normal and successfully trans-

planted individuals. The CRT was significantly pro-

longed in those on three times a week HD and still

TABLE I Age, gender, dialysis information, and renal function in the four study groups. There were no differences in age and gender distribution. All

subjects were Caucasians.

Gender Age, years

Weekly Creatinine

Clearance, L/week Dialysis Index

Serum

Creatinine,

mg/dL

Intermittent Peritoneal

Dialysis (n¼ 11) 6M: 5F 46.0� 11.4 56.7� 9.6

Hemodialysis (n¼ 22) 12M: 10F 48.5� 9.8 0.99� 0.08

Transplanted (n¼ 6) 3M: 3F 41.0� 6.1 2.4� 0.6

Normal (n¼ 12) 6M: 6F 46.8� 8.4 1.1� 0.2

Hemodialysis International, Vol. 7, No. 2, 2003 Ahmad

119

longer in those on IPD. Many IPD patients were clearly

underdialyzed, and a few HD patients were not reaching

the minimum targeted dialysis index of 1.0. Two patientswere transferred from IPD to HD, and their CRTs

decreased from 795 to 700 ms and from 695 to 570 ms.

Similarly, in one HD patient, the CRT improved from

460 ms to 395 ms after being transplanted.

In this cross-sectional study, one difference among

the four groups is the degree of uremia; the IPD group

was most uremic and the transplant group was least

uremic. Thus, CRT may be a sensitive test to distinguishthe effect of the adequate dose of dialysis on the neuro-

logical system. The strong correlation between the

measured dose of dialysis and the CRT further supports

this hypothesis.

Recent emphasis on patient outcomes (hospitaliza-

tion and survival) as a measure of adequacy of dialysis,

while pertinent in defining the adequate dose of dialysis,

is not helpful in deciding the impact of dialysis dose in

individual patients. Clinical assessment utilized in the

past, such as considering hematocrit and blood trans-fusion requirements, are not useful because of the use

of erythropoietin. The neurological assessment by nerve

conduction velocity is cumbersome, painful to the

patient, and not practical for clinical use. Overall health

and well-being have been assessed by various question-

naires, valuable as research tools, but of limited value in

clinical practice. The CRT is easy to administer, takes

only a few minutes, and appears to be sensitive enough tomeasure general health. Previously, the dose of dialysis,

as measured by the dialysis index, has been shown to

influence the neurological system (nerve conduction

velocity). Thus, it seems appropriate to report these

previously unpublished data.

The CRT is an easily administered test; however, the

test is nonspecific, and any impairment in well-being

would influence the test. These impairments include phy-sical as well as other impairments; thus, a decrease in

vision, motor function, or even a severe headache would

slow down the response time. Many other factors such as

age can influence the CRT, particularly in a chronically

ill patient population. In spite of the fact that CRT is not

a specific test for uremia, it may be useful in assessing

dialysis adequacy.

In the current study, the groups were comparable inage, gender, and other respects; however, this study was

not randomized or blinded, and had only a limited num-

ber of subjects. A larger prospective study is needed to

assess the usefulness of the CRT in judging the impact of

the dose of dialysis.

FIGURE 3 Correlation between dialysis index and choice reaction time

(CRT) in a group of 22 patients on hemodialysis.

700

600

500

400

300

200

100

0

CRT, ms

IPD HD TX NL

618 89±p < 0.0001

461 50±

p = 0.05

396 25±

p = 0.32

382 22±

FIGURE 1 Choice Reaction Time (CRT) in patients on IPD, HD, and

after transplantation (Tx) compared to volunteers with normal renal

function (NL). Values are mean� SD. The difference between IPD

and HD groups: p< 0.0001, between HD and TX: p¼ 0.05 and between

Tx and NL: p¼ 0.32. (IPD¼ Intermittent peritoneal dialysis,

HD¼ hemodialysis, Tx¼ transplantation)

FIGURE 2 Correlation between weekly creatinine clearance and choice

reaction time (CRT) in a group of 11 patients on intermittent peritoneal

dialysis.

Choice Reaction Time and Uremia Hemodialysis International, Vol. 7, No. 2, 2003

120

References

1 Pagel MD, Ahmad S, Vizzo J, Scribner BH. Acetate and

bicarbonate fluctuations and acetate intolerance during

dialysis. Kidney Int. 1982; 21(3):513–518.

2 Ahmad S, Pagel M, Shen F, Vizzo J, Scribner BH. Effects

of oxygen administration on the manifestation of acetate

intolerance in dialysis patients. Am J Nephrol. 1982;

2(5):256–260.

3 Babb AL, Strand MJ, Uvelli DA, Mulutinovic J, Scribner BH.

Quantitative description of dialysis treatment: A dialysis

index. Kidney Int Suppl. 1975; 2:S23–29.

4 Henderson L, Dittrich WH. Preparing to react in the

absence of uncertainty: I. New perspective on simple

reaction time. Br J Psychology. 1998; 89(Pt 4):531–554.

5 Williamson AM, Teo RK. Neurobehavioral effects of

occupational exposure to lead. Br J Ind Med. 1986;

43(6):374–380.

Hemodialysis International, Vol. 7, No. 2, 2003 Ahmad

121