• 
• 

Three Part Question

In [patients presenting to the Emergency Department with a seizure], does [serum prolactin measurement] help to [differentiate true seizure from pseudo-seizure]?

Clinical Scenario

The local ambulance service bring in a 22 year old male who was witnessed by shoppers to have a seizure in the town centre, but there are no longer any eye-witnesses available to give a collateral history. You wonder whether a prolactin level will help you to differentiate between an epileptic seizure and a pseudoseizure.

Search Strategy

Medline 1948 to Week 1 April 2011, via OVID interface, and Embase 1980 to week 15 2011
[seizure OR convulsion] AND [prolactin] AND [ Pseudoseizure OR Non-epileptic OR Psychogenic]. Limit to abstracts, English language, humans and human.

Search Outcome

A total of 32 different articles were found. 12 were relevant to the clinical question; 7 studies and 5 Review articles

Relevant Paper(s)

Author, date and country	Patient group	Study type (level of evidence)	Outcomes	Key results	Study Weaknesses
Willert et al 2004 Germany	Consecutive patients undergoing long term video-EEG monitoring for non-invasive evaluation for epilepsy surgery or classification of aetiologically unknown seizures. 44 patients recruited; 32 with epileptic seizures (ES), 12 with psychogenic non-epileptic seizures (PNES) and 16 healthy controls.	Case control	Baseline and post-ictal venous blood samples (at 10min, 20min, 30min, 1h, 6h, 12h and 24h) for prolactin (PRL).	No differences between baseline values for groups. Significant elevation of post-ictal prolactin for ES (p<0.001 at 10min declining to p<0.05 at 6h) and PNES(p<0.002 at 10min declining to 0.004 at 6h ). Patients with ES had significantly greater increase of prolactin levels within first 6 hrs than patients with PNES (p<0.001 at 10min).	Small sample. Baseline PRL needed for comparison. Excluded acute CNS disorders and patients on other medication.
Mishra et al. 1990 India	77 patients studied; 35 true seizures, 20 pseudoseizures, 22 healthy persons.	Case control	Post-ictal and inter-ictal serum prolactin levels.	Inter-ictal prolactin levels similar in all patients. Significant rise in prolactin levels in true epilepsy (p<0.001) but not simple partial seizures. No change in prolactin levels in pseudoseizure cases.	No details of recruitment methods. No reference range for prolactin given.
Collins et al. 1983 Ireland	35 inpatients;18 with generalised tonic-clonic (GTC) seizures, 10 with partial seizures, 8 with pseudoseizures (one of these had both GTC and psychogenic seizures and included twice).	Case control	Prolactin levels drawn at 5min, 15min, 30min, 60min, 90min, 2h, 3h post-seizure.	Plasma prolactin level was elevated in all patients in generalised seizure group and normal in partial seizure and pseudoseizure groups.	Inpatient population used. No baseline levels taken. IVC insertion may have contributed to rise in prolactin.
Shukla et al 2004 India	Patients attending neurology services at the All India Institute of Medical Sciences in a 1 yr period. Recruited into 2 groups: psychogenic non-epileptic seizure (19 patients) and complex partial seizures (CPS) (17 patients).	Case control	Prolactin levels collected within 15-20 minutes of occurance of ‘habitual event’ which was monitored with video telemetry and/or EEG. Recorded events reviewed independently by 2 separate clinicians to determine diagnosis.	No significant difference in mean levels of prolactin between the 2 groups (p = 0.24).	Baseline prolactin levels not performed despite planning. Modification of anticonvulsants may have contributed to altered prolactin levels. Inconsistent methods of monitoring patient groups. CPS group all had radiological abnormalities on MRI.
Alving 1998 Denmark	91 patients over a 2 year period were studied. 33 excluded, leaving 58 patients: 38 with epileptic seizures (ES), 20 pseudo-epileptic seizures (PES)	Case Control	Prolactin levels measured post-ictally (15 mins) and baseline (at 2 hours after first sample).	ES resulted in significant post-ictal elevation of prolactin when compared with baseline (p<0.001). However, after PES the difference between postictal and baseline levels also reached significance (p< 0.01), although less pronounced.	No details of recruitment methods. Inconsistent methods of monitoring patients. Several of PES group had been treated with anticonvulsants prior to study.
Mehta et al 1994 India	87 patients who had seizures in hospital; 60 with generalised tonic-clonic seizures, 18 with partial seizures, 9 with pseudoseizures, and 10 controls.	Case control	Post-ictal blood samples taken at 10min, 20min, 30min, 60min and 120 minutes to measure prolactin levels. Basal levels taken in 8 generalised seizure group, 3 partial and 10 controls.	Serum prolactin levels remained significantly elevated for up to 30 mins post-ictally in the generalised tonic-clonic group. Corresponding levels in pseudoseizures within normal limits. Insignificant rise after partial seizures.	Baseline levels not taken from all patients. Small numbers of pseudo-seizure relative to true epilepsy patients.
Rao et al 1989 Germany	12 patients with refractory seizures (6 epileptic and 6 pseudoseizures) and 28 matched healthy subjects undertaking simultaneous video/EEG and drug monitoring.	Case control	Blood levels of prolactin measured every 15 mins for 2 hours post-seizure.	Serum prolactin levels increase immediately after epileptic seizure post-seizure, and decline thereafter to baseline levels. No rise in patients with psychogenic seizures and no decline post-seizure.	Small sample. No p values given. Possible influence of anticonvulsant drugs in patients. Nocturnal seizures excluded.

Comment(s)

There is a poor quality of evidence available in these studies. None of these studies were based within the Emergency Department; their results cannot be extrapolated to the ED as a baseline measurement of PRL was used as a comparison in most of the studies we have looked at. There appears to be several confounding factors which influence PRL values: PRL can be altered in response to treatment with anticonvulsant treatment , circadian rhythms and physiological stresses, i.e. vascular access insertion. This makes a one-off prolactin measurement in the Emergency Department of questionable value in differentiating epileptic seizures and pseudoseizures. Serum prolactin may provide a role, in conjunction with other modalities, in the investigation of recurrent seizure activity.

Clinical Bottom Line

There is some evidence to suggest that prolactin is raised after an epileptic seizure (and not after a pseudo-seizure), however there is insufficient evidence to suggest a one-off prolactin measurement in patients attending the Emergency Department following a seizure event is useful.

Level of Evidence

Level 3 - Small numbers of small studies or great heterogeneity or very different population.

References

1. Willert C, Spitzer C, Kusserow S, Runge U. Serum neuron-specific enolase, prolactin, and creatine kinase after epileptic and psychogenic non-epileptic seizures Acta Neurologica Scandinavica 2004 Vol 109; 318-323
2. Mishra V, Gahlaut DS, Kumar S, Mathur GP, Agnihotri SS, Gupta V Value of serum prolactin in differentiating epilepsy from pseudoseizure The Journal of Association of Physicians of India 1990 Vol 38 No 11 pp 846-847
3. Collins WCJ, Lanigan O, Callaghan N Plasma prolactin concentrations following epileptic and pseudoseizures Journal of Neurology, Neurosurgery and Psychiatry 1983; 46: 505-508
4. Shukla G, Bhatia M, Vivekanandhan S, Gupta N, Tripathi M, Srivastava A, Pandey R, Jain S Serum prolactin levels for differentiation of non-epileptic verus true seizures: limited utility Epilepsy and Behaviour 2004; 5: 517-521
5. Alving J Serum prolactin levels are elevated also after pseudo-epileptic seizures. Seizure 1998; 7: 85-89
6. Mehta S, Dham S, Lazar A, Narayanswamy A, Sai Prasad GS Prolactin and cortisol levels in seizure disorders The Journal of the Association of Physicians of India 1994 Vol 42 PT 9 PP 709-712
7. Rao ML, Stefan H, Bauer J Epileptic but not psychogenic seizures are accompanied by simultaneous elevation of serum pituitary hormones and cortisol levels Neuroendocrinology 1989; 49: 33-39