Margaret Moses is logged in Logo

Sven Janno^,^a^,¹, Matti M. Holi^,^b^,^,, Katinka Tuisku^b and Kristian Wahlbeck^,^c

^a Department of Psychiatry, University of Tartu, 31 Raja St., Tartu, 50417, Estonia
^b Department of Psychiatry, Helsinki University, P.O. Box 320, HUS 00029, Finland
^c National Research and Development Centre for Welfare and Health, Finland

Received 3 March 2004; accepted 13 May 2004. Available online 23 July 2004.

1. Introduction

Neuroleptic-induced akathisia (NIA) is a common adverse effect of traditional antipsychotics (Barnes and McPhillips, 1999). The diagnosis requires (1) characteristic restless movements and (2) typical subjective complaints of restlessness referable to the legs, inner tension, and discomfort (APA, 2000). Assessment of NIA is commonly based on clinician's subjective judgement through unstructured clinical observations or through rating scales, of which the most established is Barnes Akathisia Rating Scale (BARS) (Barnes, 1989). A diagnostic threshold score of 2 on the BARS global item has been used to identify NIA cases (Barnes, 1989).

Movement disorders can be measured objectively by recording motor activity (Hoff et al., 1999). Motor activity in NIA has been measured in a few studies (e.g., Braude et al., 1984 and Gardos et al., 1992). Measurement of "controlled rest activity" by actometry has discriminated NIA patients from healthy controls, and symptomatic NIA from NIA in remission (Tuisku et al., 1999). Standardized actometry may be suitable in finding mild or hypokinetic NIA cases (Tuisku et al. 2000), and in the assessment of NIA symptom severity (Tuisku et al., 1999 and Poyurovsky et al., 2000).

As there are no reports on the validity of rating scales compared to objective movement recording in the assessment of NIA, we aimed to evaluate a standardized actometric method (controlled rest activity measurement) and BARS in a large clinical population of unselected schizophrenia patients by comparing them in case identifying and severity measurement of NIA.

2. Experimental procedures

2.1. Subjects

We recruited 99 chronic schizophrenic institutionalized adult patients from a state nursing home in Estonia. Inclusion criteria were DSM-IV diagnosis of schizophrenia or schizoaffective disorder, stable antipsychotic medication (for at least 1 month), and age of 18–65 years. Diagnosis was made using a semi-structured interview according to DSM-IV criteria for schizophrenia by a psychiatrist (SJ) and medical records. Patients with severe somatic illness and neurological illness were excluded. Written, informed consent was obtained from the subjects and the study was approved by the Ethics Review Committee on Human Research of the University of Tartu. Data were collected from October 29, 2001 to March 27, 2002. An experienced clinician (SJ) assessed all the subjects to identify NIA cases in accordance with DSM-IV. The DSM-IV diagnostic criteria for other neuroleptic-induced movement disorders (NIMD) were also checked.

2.2. Measurements

Clinical NIA symptoms were assessed by BARS and the motor symptoms of NIA were objectively measured by actometry. The actometric recording was performed during sitting in a standardized clinical interview for 30 min, a method described previously as measuring "controlled rest activity" in NIA. (Tuisku et al., 1999). Controlled rest activity is a parameter of motor activity in a situation where sitting still is adequate and expected, but not instructed or required. The actometers (PAM3; IM-Systems, Baltimore, USA) were attached to the ankles of the subjects to measure lower limb motor activity. Actometers are small, computerized movement detectors of matchbox size, which do not influence normal moving of the patient. The mode of data collection was digital integration, and the sampling rate was 40 Hz and the chosen epoch was 0.1 s. PAM3 records acceleration signals exceeding 0.1 g. When properly calibrated and used in an adequately standardized setting, no technical, environmental or physiological artefacts occur.

2.3. Statistical analyses

The correlation between the lower limb actometric activity index (the mean activity of right and left ankle) and BARS global score, was analysed by Spearman correlation test. Differences between the NIA and the non-NIA patient median values in these variables were analyzed by Mann–Whitney's two-tailed U-test. The performances of the two instruments in NIA case identification were evaluated by receiver operating characteristics (ROC) analyses. Validity coefficients (specificity and sensitivity) for different thresholds were calculated and the cut-off points were defined by optimal trade-off between sensitivity and specificity. The software used in analyses was SPSS 11.0 (SPSS, Chicago, IL, USA).

3. Results

Of the 99 participants, 45 (45.5%) were male and 54 (54.5%) female. The mean age was 49.7 (S.D. 9.5) years. The mean continuous treatment in hospital or in nursing home was 13.6 (S.D. 9.0) years. Seventy-nine (79.8%) patients used conventional antipsychotics (70 on low-dose, and 9 on high-dose neuroleptics) and 20 (20.2%) used clozapine. The mean daily chlorpromazine equivalent dose was 328 (S.D. 221) mg. The prevalence of NIA according to DSM-IV criteria was 31.3%. Among patients with NIA 10 had parkinsonism, 12 tardive dyskinesia; 3 of them had all three conditions. The total prevalence of NIMD in the whole sample according to DSM-IV was 61.6%. Actometric data was missing for one male patient due to non-cooperation.

The BARS global score did not correlate significantly with lower limb activity index neither in the total population (r=0.189, p=0.063), nor in the NIA subgroup (r=0.159, p=0.393). After controlling for the effect of presence or non-presence of parkinsonism and/or tardive dyskinesia DSM-IV diagnoses (post-hoc analysis of co-variance), a significant correlation between BARS global and the lower limb activity index could be found (r=0.265, p=0.009).

The median BARS global score for NIA patients was significantly higher from that of non-NIA patients (2 and 0, respectively, U=−8.20, p=0.000). The median lower limb activity index for NIA patients was also significantly higher than that of non-NIA patients (19870 and 10119, respectively, U=−2.82, p=0.005).

In ROC analysis AUC for BARS global was 0.971 (CI=0.945–0.998), and for lower limb activity index 0.683 (CI=0.578–0.787).

The validity coefficients and the optimal cut-off points of the methods are presented in Table 1.

Table 1. Validity coefficients and optimal cut-off points (bolded text) of Barnes Akathisia Rating Scale (BARS) Global score, and Lower Limb Activity Index measured by actometry during 30-min controlled rest in a clinical population of 99 chronic schizophrenia patients

4. Discussion

This methodological study was the first to compare BARS and actometric lower-limb controlled rest activity in NIA in a large naturalistic population with different NIMD.

4.1. Case identification

Both measures differentiated NIA from non-NIA group, but BARS proved to be superior to actometry in finding NIA cases. Although earlier reports suggest discriminant validity for quantitative actometry between pure NIA and non-NIA subjects, in this study it was not capable of accurately identifying NIA patients from the naturalistic sample of the study. This may be due to confounding effect of other NIMD manifesting with hyperactivity (parkinsonian tremor and tardive dyskinesia). The qualitative analysis of actometric movement patterns (Tuisku et al., 1999 and Rapoport et al., 1994) could differentiate between the different NIMD, but it was beyond the scope of this article. One reason for the better performance of the BARS in identifying DSM-IV defined NIA cases may be the similarity in content between the inquired items and the methodological similarity of the examination procedures.

4.2. Symptom severity measurement

Actometric lower limb activity and BARS Global score correlated weakly but significantly. The correlation became statistically significant only after controlling for other NIMD. The instruments may measure somewhat different constructs; Actometry records only movements, while BARS includes a subjective experience in its score.

4.3. Validity of BARS

BARS has been suggested to need objective validation by movement measuring (Barnes, 1989). The convergence between actometry and BARS in this study (the correlation and the ability to discriminate between NIA and non-NIA groups) provide some objective support for the validity of BARS.

4.4. Utility of actometry

With appropriate cut-off, actometry is very sensitive in finding akathisia cases, but asking about subjective symptoms (e.g., BARS awareness item) may be needed in order to specifically rule out other NIMD in a naturalistic population. After the NIA diagnosis has been made, however, actometry could be a useful instrument for measuring symptom change induced by, e.g., therapeutic intervention. BARS is a considerably rough measure of symptom severity, as it has only 6 possible global scores (0–1–2–3–4–5). Minor changes may be difficult to detect by BARS compared to actometry, in which quantitative activity count has proved to be a sensitive follow-up tool (Tuisku et al., 2002).

5. Conclusions

In a naturalistic population with several co-morbid NIMD, BARS is more practical and suitable as a DSM-IV based screening method for NIA than quantitative actometry. Actometry, however, may offer an objective additional option for assessment of symptom severity if other hyperactive NIMD are controlled for.

References

American Psychiatric Association, 2000. Diagnostic and Statistical Manual of Mental Disorders. (4th ed.), American Psychiatric Press, Washington, DC.

Barnes, T.R.E., 1989. A rating scale for drug induced akathisia. Br. J. Psychiatry 154, pp. 672–676.

Barnes, T.R.E. and McPhillips, M.A., 1999. Critical analysis and comparison of the side effects and safety profiles of the new antipsychotics. Br. J. Psychiatry 38S, pp. 34–43.

Braude, W.M., Charles, I.P. and Barnes, T.R.E., 1984. Course, jerky, foot tremor: tremographic investigation of an objective sign of acute akathisia. Psychopharmacology 82, pp. 95–101.

Gardos, G., Teicher, M.H., Lipinski Jr., J.F., Matthews, J.D., Morrison, L., Conley, C. and Cole, J.O., 1992. Quantitative assessment of psychomotor activity in patients with neuroleptic-induced akathisia. Prog. Neuro-psychopharmacol. Biol. Psychiatry 16, pp. 27–37.

Hoff, J.I., van Hilten, B.J. and Roos, R.A., 1999. A review of the assessment of dyskinesias. Mov. Disord. 14 5, pp. 737–743.

Poyurovsky, M., Nave, R., Epstein, R., Tzischinsky, O., Schneidman, M., Barnes, T.R., Weizman, A. and Lavie, P., 2000. Actigraphic monitoring (actigraphy) of circadian locomotor activity in schizophrenic patients with acute neuroleptic-induced akathisia. Eur. Neuropsychopharmacol. 10 3, pp. 171–176. Abstract | Full Text + Links | PDF (92 K)

Rapoport, A., Stein, D., Grinshpoon, A. and Elizur, A., 1994. Akathisia and pseudoakathisia: clinical observations and accelerometric recordings. J. Clin. Psychiatry 55 11, pp. 473–477.

Tuisku, K., Lauerma, H., Holi, M.M., Markkula, J. and Rimon, R., 1999. Measuring neuroleptic-induced akathisia by three-channel actometry. Schizophr. Res. 40, pp. 105–110. Abstract | Full Text + Links | PDF (121 K)

Tuisku, K., Lauerma, H., Holi, M.M., Honkonen, T. and Rimon, R., 2000. Akathisia masked by hypokinesia. Pharmacopsychiatry 33, pp. 147–149.

Tuisku, K., Wahlbeck, K., Holi, M., Ahlgren, A. and Lauerma, H., 2002. The treatment response to pramipexole in restless legs syndrome measured by actometry (abstract). Nord. J. Psychiatry 56 2, p. 117.



		This Document


			SummaryPlus
			Full Text + Links
			·Full Size Images
			PDF (75 K)



		Actions


			E-mail Article

Abstract

Article Outline