Sleep-Related Violence and Sexual Behavior in Sleep: A Systematic Review of Medical-Legal Case Reports

METHODS

RESULTS

We found 699 references; all abstracts were screened and 62 articles with potentially relevant material were examined in detail, leading to the final identification of 27 articles (26 in Pubmed and 1 in PsychINFO), containing a total of 35 medical-legal cases.

Sixteen of the above 27 articles, reporting a total of 18 cases (9 of SRV^29–36 and 9 of SBS^{6,7,13,37–41}), met the inclusion criteria (Figure 1). The selected articles were published from 1985 to 2011 and included 9 single case reports,^{31–34,36–39,41} and 7 case series with at least 1 medical-legal case.^{6,7,13,29,30,35,40} Seven articles were published in psychiatry journals,^{6,7,29,30,34,35,38} 5 in legal medicine journals,^{33,36,37,39,41} 3 in sleep medicine journals,^13,31,32 and 1 in a sexual medicine journal.⁴⁰ In cases fulfilling the inclusion criteria but lacking data regarding the trial outcome, we directly contacted the authors. Each case was summarized in terms of the description of the event, the legal charge(s), the defense, the forensic evaluation, and the court verdict (Table 2 and Table 3).

The remaining 11 articles, reporting a total of 18 cases, were excluded because it was not clear whether the sleep disorder was actually purported as the defense during the trial, or because information about the forensic evaluation was lacking. A description of excluded cases is reported in Table S1 (supplemental material).

DISCUSSION

From the 1980s, with the advent of formal video-PSG monitoring techniques and the first systematic classification of sleep disorders, the forensic dimension of sleep medicine entered a very new era. Since then, there has emerged a relevant scientific medical literature on this topic. Nevertheless, the sleep defense still “strikes the heart of the criminal law jurisprudence”,⁵¹ which reflects the difficulty in evaluating the defendant's level of consciousness and the state of mind and volitional criminal intent. This requires a sleep expert opinion to provide objective and scientific elements supporting (or not) an underlying sleep disorder as the basis for the criminal behavior.

Our review has framed previously published medical-legal cases by means of a detailed analysis of legal issues, the defendant and victim characteristics, the circumstantial factors surrounding the crime, and the forensic evaluation. In order to have an official and international peer-reviewed foundation for establishing the presence of a sleep disorder, we included articles published since 1980, i.e., after the release of the first “Diagnostic classification of sleep and arousal disorders.”²⁸ Cases sufficiently informative about the forensic evaluation (i.e., only half of those retrieved) were published from 1985 to 2011, mainly in journals in the psychiatric field, but also in the fields of legal medicine, sleep medicine, and sexual medicine, thus calling attention to the growing multidisciplinary interest to publish these cases. Unfortunately, the lack of completeness of the remaining published cases prevented an analysis about all the cases that came to trial.

All SRV reports encompassed major crimes (murder or attempted murder), while in SBS cases the criminal charges ranged from sexual touching to rape. In 1 case, the charge of “indecent exposure”³⁷ demonstrates that sleep disorders may result in criminal behavior, even when physical contact between a defendant and the victim is lacking.

In most of the cases, the sleep disorder supporting the sleep defense was sleepwalking, which indicates that the sleep defense generally corresponds to a “sleepwalking defense.” The trial outcome was in favor of the defendant in all SBS and in two-thirds of SRV cases. However, we can not exclude a publication bias due to a possible greater interest engendered by an acquittal than by a conviction. In addition, some verdicts may be challenged on appeal, which could ultimately result in a different outcome for the defendant.

Almost all these legal cases concerned a single criminal episode, and had a single victim. In all cases, defendants were men of relatively young age; this was also found by Bonkalo in 20 cases of homicide collected from 1791 until 1969.²⁶ Beyond medical-legal cases, SRV with moderate to severe injuries is known to be frequently reported in males of relatively young age.^52–56

Moreover, based on our results, in most cases defendants had no prior criminal record, were employed, had a partner, and did not demonstrate any antisocial trait. In the context of otherwise unremarkable medical histories, the defendants' sleep history usually disclosed past or ongoing sleepwalking, frequently associated with other parasomnias and characterized by complex sleep behaviors. A family history, when investigated, revealed parasomnia in relatives in almost all cases.

Interestingly, differences between SRV and SBS cases were found in the victims' characteristics. Indeed, in most SRV cases the victim was an adult relative of the defendant, and of female gender in two-thirds of cases, while SBS victims were always females, and in most cases minors without a familial relationship with the defendant. Since it is known from the literature that in SBS cases, bed partners often experience physical injuries (ecchymoses, lacerations),⁸ our results suggest that the sexual offence is criminally reported only when the defendant is not known/not a relative or when the victim is a minor.

When reported, the time interval between the defendant's sleep onset and the crime was around 1-2 hours, consistent with the presence of a NREM parasomnia in most cases. In accordance with the published literature,²⁷ there was a close physical proximity between the defendant and the victim (sleeping in the same bed, sleeping the same room, sharing the same home) in almost all SRV cases and in all SBS cases.

Alcohol intake by the defendant on the night of the event was reported in 2 of 9 of SRV cases and in 6 of 9 of SBS cases. Despite alcohol intake being previously listed as a “precipitating factor” of confusional arousals, and a “risk factor” for sleepwalking,⁵⁷ the scientific evidence of alcohol-induced sleepwalking or confusional arousals as a defense to criminal behavior has been reviewed.^58,59 The recently published International Classification of Sleep Disorders (ICSD-3)¹⁰ highlights the absence of a compelling relationship between alcohol use and a disorder of arousal, stating that, in the presence of alcohol intoxication, disorders of arousals should not be diagnosed. The above position complies with the current Diagnostic and Statistical Manual of Mental Disorders edition that has removed alcohol from the list of possible triggers for sleepwalking, adding also a section on the differential diagnosis of alcohol blackout.⁶⁰ These major shifts in opinion have important implications for those forensic cases in which unspecified or excessive amount of alcohol intake was reported.

A sustained or coincidental exposure to stressful condition was often reported, while sleep deprivation and caffeine overuse were pointed out less frequently. All the above mentioned conditions have been reported in literature as potential triggering factors for SRV in sleepwalking and sleep terrors,^22,53 and a recent longitudinal survey of 100 sleepwalkers found that those with a history of SRV more frequently reported triggering factors.⁵⁶

The forensic evaluations differed from case to case for both SRV and SBS. A forensic evaluation restricted to neurological and/or psychiatric assessments, or at most an EEG, was carried out in the older cases, while all cases published after 1998 reported at least a PSG or video-PSG study for 1 to 6 nights, along with neuropsychological tests, brain computerized tomography, or magnetic resonance imaging in some cases. PSG at times was performed under sleep restriction or with sound/ tactile provocation of arousals during the night. In two cases followed by the same author,^36,41 a PSG alcohol challenge study was performed, with limited details provided, and without an explicit rationale. In 1 case,³⁴ the Court ordered to replicate the same forensic workup performed in a previous case.³¹

Our review concludes that in most cases, the advice of a sleep expert is requested in cases of a suspected or purported sleepwalking-related felony.

A consensus about guidelines to follow in the forensic assessment of such cases is still lacking. Nevertheless, there is broad agreement that the sleep expert workup should include at least the following steps:

1. History of sleep disorders should be carefully investigated in relatives.^6,21,35

2. A complete description of the defendant's lifetime history of any motor behavior during sleep, preferably from both the defendant and possible witnesses (present and former bed partners/relatives/friends) should be obtained, and details about age at onset, the usual timing of the event during the sleep, the degree of amnesia, and both duration and frequency of episodes should be investigated.^1,6,21,35,61

3. Information about sleep/wake habits, drugs (prescribed or illicit), herbal products, and habitual caffeine and alcohol consumption/abuse should be collected.^6,35,53

4. Along with information about the event, circumstantial factors of both the person's life and the hours prior to the episode is essential: stressful events, sleep deprivation or excessive fatigue, and intake of alcohol and other substances should be investigated.^6,35,61

5. Complete physical, neurologic, and psychiatric evaluations along with administration of standardized questionnaires for sleep disorders should be carried out.^{4,15,20,35,62}

6. A video-PSG study to identify or rule out other sleep disorders associated with abnormal motor behaviors (RBD, nocturnal frontal lobe epilepsy) or possibly triggering sleepwalking (OSA, periodic limb movements) should be performed with standard polysomnographic monitoring⁶³ and with an extensive scalp EEG, electromyographic monitoring also of the arms, and time-synchronized audiovisual recording.^4,15,20,21 To increase the possibility to capture an event, the documentation of nocturnal episodes with home video using a camera with infrared night vision function could be useful.^4,6,62,64 Combining video and PSG monitoring at home may allow for longer recording periods and minimize bias from monitoring in a sleep laboratory setting. However, home video-PSG recording prevents the possibility of technician/physician intervention during and after a parasomnia episode. For these reasons, home video-PSG should be performed, when feasible, in conjunction with a sleep laboratory study.

It should be emphasized in the forensic context that irrespective of whether an event compatible with sleepwalking is recorded during PSG, this will not conclusively indicate that the defendant was or was not sleepwalking at time of the criminal event. Indeed, the only direct evidence of whether or not a criminal act occurred during a state of parasomnia comes from any eyewitness testimony and/or evidence obtained from the scene of the crime. Nevertheless, sleep experts have the duty to pursue as much as possible objective data to support or not support a diagnosis of sleepwalking. Accordingly, they should first evaluate the conditions in which a sleepwalking episode may occur, namely a genetic predisposition in the presence of an increased pressure for slow wave sleep and factors favoring arousals or fragmenting sleep.⁴

From this perspective, the Montplaisir group's sleep-deprivation protocols⁶⁵ have been considered by some authors as a “promising novel approach to the forensic use of PSG in sleepwalking cases” for ruling out or greatly minimizing the probability that the accused is in fact a sleepwalker.⁶⁶ In predisposed individuals, 25-38 hours of sleep deprivation increases the number and the complexity of sleepwalking events recorded in the laboratory during sleep recovery, whereas similar sleep deprivation does not lead to sleep behaviors in subjects with no history of sleepwalking.^65,67,68 In addition, several studies have disclosed neurophysiological abnormalities in sleepwalkers' slow wave sleep, even on nights without episodes, namely the absence of NREM sleep continuity, the overall decrease in slow wave activity during the first sleep cycles, and the increased cyclic alternating pattern rate.^69–74 Further scientific evidence is needed to establish that these PSG findings are stable neuro-physiological markers of sleepwalking. Therefore, at the present time, they may only provide indirect and circumstantial evidence in courtroom.

Cartwright and Guilleminault have recently reported that a PSG spectral analysis was used during a trial as a basis of an expert witness testimony.⁷⁵ The reliability of this analysis in a forensic setting has been seriously questioned on the grounds of the lack of sufficient sensitivity, specificity, and stability over the time.⁷⁶ In particular, according to some studies,^77–79 arousals from deep sleep and hypersynchronous delta waves lack sufficient sensitivity and specificity to be used as diagnostic markers.¹⁰ In their rebuttal, Cartwright and Guilleminault claimed that several studies make “a low slow wave activity a strong candidate to be a manifestation of the genetic vulnerability to abnormal delta arousals and therefore likely to be a stable characteristic of sleep.”⁸⁰

Finally, due to the complex nature of SRV and SBS, the opinion of a properly credentialed sleep expert^81,82 and a multi-disciplinary approach are highly recommended in forensic cases.^16,20

Forensic sleep medicine is at an embryonic stage. Our review points out that many medical-legal cases of SRV and SBS, reported in the form of small case series and case reports, do not provide essential information about a proper forensic evaluation. Cases sufficiently informative offer the conclusion that court trials for SRV and SBS involved relatively young and healthy adult males as the defendants, while victims were usually adult relatives in SRV cases and unrelated teenage or young girls in SBS cases. Although many sleep disorders could result in assaultive behaviors, in most cases sleepwalking was implicated, and the sleepwalking defense turned out to be generally successful. However, no common protocols for the forensic evaluation were utilized.

An international expert consensus among sleep experts, medical-legal experts, and psychiatrists for the forensic evaluation of SRV and SBS cases should be developed, which we consider to be an urgent priority. This report has provided a foundation for advocating such an international consensus.

Finally, enhancing the publication of accurate and comprehensive reports of medical-legal cases involving SRV and SBS would provide essential information for sleep medicine experts who participate in forensic evaluations, and a relatively homogeneous body of data for ongoing scientific research.

DISCLOSURE STATEMENT

This was not an industry supported study. The authors have indicated no financial conflicts of interest.

SUPPLEMENTAL MATERIAL