Neurology India Medknow Publications on behalf of the Neurological Society of India ISSN: 0028-3886 EISSN: 1998-4022 Vol. 58, Num. 2, 2010, pp. 220-224

Neurology India, Vol. 58, No. 2, March-April, 2010, pp. 220-224

Original Article

Prevalence of abnormal psychometric tests and critical flicker frequency after clinical recovery of overt hepatic encephalopathy

Praveen Sharma, Barjesh Chander Sharma, Shiv Kumar Sarin

Department of Gastroenterology, G. B. Pant Hospital, New Delhi-110 002, India

Correspondence Address: Barjesh Chander Sharma, Department of Gastroenterology, Room 203, Academic Block, G. B. Pant Hospital, New Delhi-110 002, India, drpraveen_sharma@yahoo.com

Date of Acceptance: 29-Oct-2009

Code Number: ni10059

PMID: 20508339

DOI: 10.4103/0028-3886.63797

Abstract

Background: Abnormal psychometric tests are common in cirrhotic patients without overt encephalopathy. It may seriously impair the patient's daily functioning and quality of life. Prevalence of abnormal psychometric tests after the clinical recovery of overt encephalopathy has not been studied.
Materials and Methods: Consecutive 80 patients (age, 44.5±9.8yr; M: F 66:14) with hepatic encephalopathy (HE) who recovered and met inclusion criteria were evaluated for abnormal psychometric tests ( > 2 SD) and critical flicker frequency (CFF < 38Hz). All patients underwent number connection tests-A and B, figure connection tests -A and B (if illiterate), digit symbol test, object assembly test, CFF and arterial ammonia after one month of clinical recovery of HE.
Results: Prevalence of abnormal psychometric tests (> 2SD) was seen in 58 (72.5%) patients. [33 (66%) of 50 patients on lactulose and 25 (83%) of 30 patients not on lactulose, P = 0.12]. Arterial ammonia was significantly higher in patients with abnormal psychometric tests compared to patients without it (86.3±22.2 vs 50.4±11.5 mmol/l, P = 0.01). CFF was < 38Hz in 50 patients (62.5%). Patients with ≥2 abnormal psychometric tests had CFF significantly lower than with normal psychometric tests (35.6±2.3 vs 40.7±2.4 Hz, P = 0.001). CFF sensitivity and specificity in diagnosing ≥2 abnormal psychometric tests was 82 and 83% respectively. CFF was significantly correlated with psychometric tests [NCT-A (-0.672, P = 0.001) and B (-0.743, P = 0.001), digit symbol test (-0.533, P = 0.001), object assembly test (-0.659, P =0.001) and arterial ammonia level (-0.385, P = 0.02)].
Conclusion: About 73% of patients who recovered from HE had abnormal psychometric tests ( > 2SD). CFF is a simple reliable tool to find out the presence of abnormal tests.

Keywords: Critical flicker frequency, hepatic encephalopathy, psychometric tests

Introduction

Hepatic encephalopathy (HE) is defined as a disturbance in the central nervous system function because of hepatic insufficiency and is manifested by a spectrum of abnormalities affecting cognition, attention, functional ability, personality and intellect. ^[1] Overt HE occurs in 30-45% of cirrhotic patients and development of HE is associated with a poor prognosis. ^[2],[3] Minimal hepatic encephalopathy (MHE) is characterized by slight cognitive abnormalities and may seriously impair patient′s daily functioning and quality of life. ^[1] MHE affects 30-75% of patients with cirrhosis, is difficult to detect but can be identified with psychometric tests. ^{[4],[5],[6],[7],[8]} The pathogenesis of MHE is thought to be similar to that of HE, and ammonia plays a key role. ^[9],[10] Studies had shown that MHE patients over a period of follow-up develop HE more than those who did not have MHE. ^{[3],[11],[12]} In an American Association for the Study of Liver Diseases (AASLD) survey, only very few hepatologists were able to test for MHE more than 50% of the times and 38% respondents never tested for it. Main barriers to MHE testing included adding time to clinic visits and lack of standardized norms. ^[13] Critical flicker frequency (CFF) has recently been shown to be correlated with psychometric tests in the assessment of MHE. ^[7],[8] Prevalence of abnormal psychometric tests after clinical recovery of HE has not been well studied. We evaluated the prevalence of abnormal psychometric tests after the clinical recovery of HE and utility of CFF in the assessment of these abnormal tests.

Materials and Methods

From March 2007 to April 2008, consecutive cirrhotic patients who recovered from HE were enrolled. Cirrhosis was diagnosed on a clinical basis involving laboratory tests, endoscopic evidence, sonographic findings and liver histology, if available. The exclusion criteria were history of taking lactulose or any antibiotics in the past six weeks prior to this episode of HE, alcohol intake during the past 12 weeks, patients on secondary prophylaxis for spontaneous bacterial peritonitis, hepatocellular carcinoma, previous transjugular intrahepatic portosystemic shunt (TIPS) or shunt surgery, significant co morbid illness such as heart, respiratory, or renal failure and any neurologic diseases such as Alzheimer′s disease, Parkinson′s disease and nonhepatic metabolic encephalopathies. Patients with color blindness and mature cataract, diabetic retinopathy and patients on psychoactive drugs, such as antidepressants or sedatives were also excluded. Detailed information about the study was given to all the patients and informed written consent was taken. Psychometric tests and CFF were done on the same day after (30.2±6.6 days) clinical recovery of HE [Figure - 1].

Psychometric testing

All patients underwent a series of psychometric tests after clinical recovery from HE and included 2 number connection tests (NCTs), parts A and B; 2 figure connection tests (FCTs), parts A and B if patient was illiterate; and 2 performance subtests of the Wechsler Adult Intelligence Scale, the object assembly test and digit symbol test. In principle the FCT is similar to the NCT, except that numbers were replaced by figures. Psychometric test for NCT/FCT was considered abnormal when both NCT-A and NCT-B or FCT-A and FCT-B were abnormal. The test score is the time required to complete the test, including the time needed to correct any errors. Tests were considered abnormal when test score was more than mean th + 2 S.D of the age and education matched controls. ^[6],[14]

Measurement of CFF

CFF was done by HEPAtonorm analyzer in a quiet, semi-dark room. Patients were first instructed and trained about the procedure. Each patient was instructed about the fundamentals in performing the test and an exercise involving five trial runs was done before patient response was actually recorded. Flicker frequencies were measured eight times and the mean value was calculated. Measurement of the CFF thresholds was done by intra-foveal stimulation with a luminous diode. Decreasing the frequency of the light pulses from 60 Hz downward, the CFF threshold was determined when the impression of fused light turned to a flickering one. Critical flicker frequency was considered abnormal when value was < 38 Hz. ^[8]

Assessment and recovery of hepatic encephalopathy

Hepatic encephalopathy was defined using West-Haven criteria. ^[15] Recovery from the episode of HE was assessed independently by two senior consultants (co-authors) with more than 10 years of experience in the field of hepatology.

Management of hepatic encephalopathy

All the patients were admitted to an intensive liver care unit with intensive supervision of the vital functions, electrolytes and acid-base status. Management was based on ability to identify and treat the precipitating factors. All the patients were treated with lactulose in a dosage so that the patients had two to three semiformed stools/day and broad spectrum antibiotics (cephalosporins mainly or as per sensitivity report) as a part of intensive care protocol for the management of HE.

Dietary habits and medication after recovery of hepatic encephalopathy

Patients were allowed to take normal diet with salt restriction as they were taking before. Most of them were vegetarian. Protein intake was not restricted and their diet was supplemented with vitamins. Fifty patients were taking lactulose (30 ml twice daily so that they had two to three semiformed stools/day) as per their treating physician decision and 30 patients were not taking lactulose. All patients continued with minimal required diuretics, antiviral treatment and beta blockers if they were taking them before.

Blood tests, imaging and biochemical examinations

After overnight fasting, patient venous blood was taken and analyzed for routine liver function tests and hematologic parameters by conventional methods. Ultrasound abdomen and Doppler were done to see for large spontaneous shunt. Arterial ammonia was measured by ammonia Test Kit II of the PocketChem BA Device (Arkay, Inc., Kyoto, Japan) on the day of psychometric tests and CFF assessment. The blood sample was collected onto ice and tested immediately after collection. The continuous measurement range is 7-286 mmol/L; the normal blood ammonia level for healthy adults for this device is less than 54 mmol/L.

Statistical analysis and data management

Data processing was performed using the SPSS software packages. Data were expressed as mean± S.D. For a comparison of categorical variables, Chi Square and Fisher′s exact tests were used, and for continuous variables, a Mann-Whitney test for unpaired data. Correlations between variables were examined with a Pearson′s correlation and Cohen′s K statistics. The probability level of P < .05 was set for statistical significance.

Results

Two hundred and twenty patients with cirrhosis and HE were screened and 80 (36%) patients met the eligibility criteria for inclusion. The etiology of cirrhosis included: alcohol (n=29), chronic hepatitis B (n=25), chronic hepatitis C (n=12) and cryptogenic cirrhosis (n=14). The clinical and demographic characteristics of the patients are shown in [Table - 1]. Of the 80 patients, 43 were in grade 1 and 2, these patients were managed in the intensive care unit for acute variceal bleed (n=16), spontaneous bacterial peritonitis (SBP) with renal impairment (n=8), severe infections (n=13) and unknown cause with a suspicion of bleed (n=6). Twelve of the patients enrolled had spontaneous large shunt. Of these 12 patients, seven patients were taking lactulose and five patients were not on lactulose after the recovery from HE. All these patients had> 2 abnormal psychometry tests and CFF < 38 Hz. The median duration from the date of admission to the clinical recovery from HE was 8 days (range 4-19 days). All patients were neurologically normal, alert and no flapping tremor, ataxia, or dysarthria.

Results of psychometric tests and CFF

Of the 80 patients, 14 were illiterates, 47 had ≤12 years of formal education and 19 were graduates. All the patients could do digit symbol test and object assembly test. Sixty six patients could do NCT-A and B while the remaining 14 patients did FCT-A and B. Fifty-eight (72.5%) patients had abnormal psychometric tests: 33 (66%) of 50 patients on lactulose and 25 (83.3%) of 30 patients not on lactulose had ≥2 abnormal psychometric tests, P = 0.12 [Table - 2]. Arterial ammonia was significantly higher in patients with ≥2 abnormal psychometric tests compared to patients with < 2 abnormal psychometric tests (86.3±22.2 vs 50.4±11.5 mmol/l, P = 0.01).

All the patients could perform CFF, eight (10%) patients, however, needed two training sessions to properly understand the fundamentals involved in the performance of test. CFF was < 38Hz in 50 (62.5%) patients. Ten patients had only psychometric impairment without CFF impairment, four had only CFF impairment without psychometric impairment and 46 had both CFF and psychometric impairments. Comparing patients who had both psychometric and CFF impairment (n=46) versus those who had either psychometric (n=10) or CFF impairment (n=4) only, we found a significant difference in arterial ammonia level (87.4±25.8 vs. 68.8±21.6 umol/l, P = 0.03) while no difference with respect to age, education level, etiology of cirrhosis or treatment with lactulose [Table - 3]. Patients with ≥2 abnormal psychometric tests had CFF significantly lower than those who did not have (35.6±2.3 vs 40.7±2.4 Hz, P=0.001). CFF sensitivity and specificity in diagnosing patients with ≥2 abnormal psychometric tests was 82% and 83% respectively. CFF had positive predictive, negative predictive value and diagnostic accuracy of 92%, 67% and 82.5% respectively. Calculating agreement by the Cohen′s K statistics between the classifications based on psychometric tests (≥2 abnormal psychometric tests) and CFF (< 38Hz) was substantial (κ =0.66). CFF was significantly correlated with psychometric test (NCT-A and B, digit symbol test, object assembly test), Child score and arterial ammonia level [Table - 4].

Discussion

There was high prevalence (73%) of ≥2 abnormal psychometric tests after clinical recovery from HE. CFF had sensitivity and specificity of 82% and 83% respectively in the diagnosis of patients with ≥2 abnormal psychometric tests. Prevalence of MHE in cirrhotics with no previous episodes of HE is variable, ranging from 30-75% depending upon the criteria used for screening. ^{[4],[5],[6],[7],[8]} Cognitive deficits in these patients, impaired psychomotor speed, attention, and visual perception will have major difficulties in safely performing routine activities of life. ^{[16],[17],[18],[19]} In this study most of the patients even after improvement from HE (grade 0), had abnormal psychometric tests. Evaluation of patients with cirrhosis by mini mental status examination (MMSE) has been suggested before evaluating them for MHE and the cut off score suggested was 24 or more. ^[20],[21] In this study, psychometric tests and CFF was done on the same day of clinical recovery from HE. At the time of evaluation all the patients were neurologically normal, alert and no flapping tremor, ataxia, or dysarthria. Gut-derived nitrogenous substances play a major role in the pathogenesis of HE and pathogenesis of MHE is thought to be similar to that of overt HE. Specifically, ammonia is thought to be a critical factor in the pathogenesis. ^[9] Ammonia- induced alterations in cerebral blood flow and glucose metabolism have shown that there is a significant decrease of glucose utilization of various cortical regions that correlate with the patients cognitive functions. ^[10]

In this study patients with abnormal psychometric tests had a higher ammonia level when compared to patients without. The main barriers to test psychometric test in cirrhotics are, adding time to clinic visits and lack of standardized norms. Advantages of using computerized tests over pencil and paper tests is that there is less reliance on the considerable motor activity required to perform certain of the psychometric tests. Associated motor deficits and movement disorders may give false impression of significant cognitive deficits. CFF is highly reproducible parameter with little bias for training effects, education, age, daytime, or interexaminate variability. ^[7],[8],[22] The only motor response needed in pressing the button The sensitivity and specificity of CFF (< 38Hz) in identifying patients with ≥2 abnormal psychometric tests in this study was 82 and 83% respectively. We found it to be a simple bedside tool for identifying patients with abnormal psychometric tests even after recovery from overt HE. A CFF value of < 38Hz was found to be predictive of further bouts of HE in patients who did not have HE before. ^[8] Thus, CFF may replace psychometry tests in the future, however, more studies are needed as we have shown in our earlier study that after the recovery from HE, the risk of development of overt HE depends on the presence of abnormal psychometry tests and not CFF. ^[23] Lactulose is effective in the management of MHE. ^[24] However, in our study there was no difference in abnormal psychometric tests between patients who were on lactulose and those who were not on lactulose. The mean duration of lactulose treatment, however, was one month in our study and it is quite possible that longer duration of treatment might be effective. The other reason could be that this study was not powered to look at the effect of lactulose on abnormal psychometric tests. In conclusion, patients who recovered from HE had high prevalence of abnormal psychometric tests and CFF is a simple tool to identify the presence of psychometric test abnormalities.

References

Copyright 2010 - Neurology India

The following images related to this document are available:

Photo images