ORIGINAL ARTICLE
Effect of sleep deprivation on the electrophysiological signature of habituation to noxious laser stimuli €rtner1, R.-D. Treede1 S. Schuh-Hofer1, U. Baumga 1 Chair of Neurophysiology, Centre of Biomedicine and Medical Technology Mannheim, Heidelberg University, Mannheim, Germany
Correspondence Sigrid Schuh-Hofer E-mail:
[email protected] Funding sources This work is part of the ‘European’ Collaboration, which has received support from the Innovative Medicines Initiative Joint Undertaking, resources of which are composed of financial contribution from the European Union’s Seventh Framework Program (FP7/ 2007–2013) and EFPIA Companies’ in kind contribution. Conflicts of interest RDT has lectured or served on advisory boards for or received grants from the following companies: Astellas, Boehringer-In€nenthal, Kade, Eli gelheim, Galderma, Gru Lilly, Merz, Nycomed, Pfizer. SSH has lectured for Allergan. The authors declare they have no competing interests.
Accepted for publication 11 February 2015 doi:10.1002/ejp.698
Abstract Background: Sleep deprivation induces hyperalgesia. However, this pronociceptive effect is not reflected at the electrophysiological level, since sleep restricted subjects show amplitude reduction of Laser-evoked Potentials (LEP). We aimed to explore the contribution of habituation to this paradoxical LEP amplitude decline. Methods: We compared LEP’s of 12 healthy students (23.2 1.1 years) after habitual sleep (HS) and a night of total sleep deprivation (TSD). Twelve repetitive laser stimulus blocks (each comprising twenty stimuli) were applied under three attention conditions (‘focusing’ – ‘neutral’ – ‘distraction’ condition). Stimulus blocks were split in part 1 (stimulus 1–10) and part 2 (stimulus 11–20). The contribution of habituation to the TSD-induced LEP amplitude decline was studied by calculating the percentage amplitude reduction of part 2 as compared to part 1. Individual sleepiness levels were correlated with (1) averaged LEP’s and (2) the degree of habituation. Results: TSD induced hyperalgesia to laser stimuli (p < 0.001). In contrast, depending on the attention condition, the P2 amplitude of the N2P2complex was significantly reduced (‘focusing’: p = 0.004; ‘neutral’: p = 0.017; distraction: p = 0.71). Habituation of the P2 amplitude to radiant heat was increased after TSD (‘focusing’: p = 0.04; ‘neutral’: p < 0.001; distraction: p = 0.88). TSD had no significant effect on N1 amplitudes (p > 0.05). Individual sleepiness correlated negatively with averaged P2 amplitudes (p = 0.02), but not with the degree of habituation (p = 0.14). Conclusion: TSD induces hyperalgesia and results in attentiondependent enhanced habituation of the P2 component. Increased habituation may – to a substantial degree – explain the TSD-induced LEP-amplitude decline. For this article, a commentary is available at the Wiley Online Library.
1. Introduction Sleep disturbances are highly prevalent in pain patients, irrespective of their pain aetiology (Finan et al., 2013). Beyond night-time pains interfering with restorative sleep, there is increasing evidence of insomnia specifically harming pain patients by exacerbating their clinical pain condition (Irwin et al., 2012). Consistent with this, experimental studies © 2015 European Pain Federation - EFICâ
indicate the risk of developing spontaneous pain being significantly increased both by sleep fragmentation (Smith et al., 2007) and long-term sleep deprivation (Haack et al., 2007, 2009, 2012). In addition, already one night of total sleep deprivation (TSD) is able to induce generalized – mechanical and thermal – hyperalgesia (Kundermann et al., 2004; Schuh-Hofer et al., 2013). To date, little is known about the effect of disturbed sleep on cortical processing of noxious
Eur J Pain 19 (2015) 1197--1209
1197
Effect of sleep deprivation on habituation of LEP’s
What’s already known about this topic? • Restricted sleep results in dissociation of perceptual and electrophysiological outcome measures of nociception. While pain sensitivity increases, amplitudes of laser-evoked potentials (LEP) decrease. What does this study add? • One night of sleep deprivation results in attention-dependent enhancement of habituation to noxious laser stimuli. • Enhanced habituation contributes to the sleepdeprivation-induced LEP amplitude decline.
stimuli. Laser-evoked Potentials (LEP) are well acknowledged to selectively study the cortical response to noxious input, since brief infrared laser pulses are known to activate Ad and C-fibres without co-activation of non-nociceptive b-fibres (Bromm and Treede, 1984, 1987; Garcia-Larrea et al., 2003; Treede et al., 2003; Cruccu et al., 2010). The main LEP parameters comprise a negative deflection over the contralateral temporal lobe (N1) and a biphasic vertex response (N2P2 complex). To what extent the subjective pain percept is reflected by the magnitude of phase-locked LEP’s is a matter of ongoing debate (Legrain et al., 2011). However, under highly standardized study conditions, the N2P2 amplitude of alert healthy subjects changes concurrently with the subjective painfulness of a noxious stimulus (Carmon et al., 1978; Garcia-Larrea et al., 1997). So far, two LEP studies on the interrelation of lacking sleep and nociception have been published. The first study, performed in our laboratory, explored the effect of sleep restriction on LEP’s of healthy subjects. Sleep restricted subjects, although verbally expressing pain hypersensitivity, exhibited an attention-dependent LEP amplitude decrease. This unexpected finding indicated the subjective and objective correlates of nociception being differentially processed (Tiede et al., 2010). More recently, sleep deprivation was shown to impact on energy thresholds required to induce a detectable LEP. Compared to habitual sleep, laser energy thresholds were significantly increased after total sleep deprivation – despite concurrent hyperalgesia (Azevedo et al., 2011). The mechanism underlying the apparent dissociative effect of sleep loss on the perceptual and
1198 Eur J Pain 19 (2015) 1197--1209
S. Schuh-Hofer et al.
electrophysiological correlates of nociception is currently unknown. However, in a past human study on pain response plasticity, uncoupling of subjective and objective measures of nociception could be explained by a mismatch of percentage habituation at the electrophysiological and subjective level (Bromm and Scharein, 1982). Taking this finding into account, we aimed to explore the impact of sleep deprivation on habituation to noxious stimuli. We hypothesized that one night of total sleep deprivation (TSD) changes the degree of habituation to laser stimuli, thereby contributing to the so far unexplained LEP amplitude reduction. To test this hypothesis, healthy volunteers were examined in a cross-over design, contrasting a night of TSD with habitual sleep (HS).
2. Materials and methods This electrophysiological study was part of a large study on the effect of sleep deprivation on somatosensory processing. The study was conducted in accordance with the Declaration of Helsinki and approved by the Local Ethical Committee of the Medical Faculty of Mannheim. All participants signed the informed consent prior to study enrolment. Details regarding inclusion criteria and study design have previously been described (Schuh-Hofer et al., 2013). Of 14 healthy students participating in the study, the EEG of 12 healthy students (age 23.2 1.1 years; seven men, five women) could be analysed. The EEG of the remaining two volunteers had to be excluded from the analysis since it was contaminated with artefacts. Importantly, none of the women complained of symptoms characteristic for a premenstrual syndrome (neither at screening nor during the study). Because of the electrophysiological significance of habituation in migraineurs, our study population was very carefully selected by a headache specialist (SSH) in order to exclude migraine as a confounding factor. None of the subjects met the diagnosis of migraine [1.1, 1.2] or probable migraine [1.6] according to ICHD-II (Headache Classification Subcommittee of the International Headache Society, 2004). Only women with a regular menstrual cycle were included and experiments exclusively took place during the transition time of the menstrual to follicular phase, thus ensuring that all women using contraceptives were examined during their pill-pause (see Table 1). Demographic data and results from screening tests (Beck Depression Inventory (BDI; Beck et al., 1961)
© 2015 European Pain Federation - EFICâ
S. Schuh-Hofer et al.
Effect of sleep deprivation on habituation of LEP’s
Table 1 Demographic data. No.
Age
Gender
Sleeping hrs (work days)
Sleeping hrs (weekend)
BDI
STAI-Trait
PSQI
Mensis HS
Mensis TSD
Contraceptive
1 2 3 4 5 6 7 8 9 10 11 12
32 20 21 21 27 21 27 21 20 22 20 26
Female Female Female Female Female Male Male Male Male Male Male Male
7.5 7.5 7.5 7.5 7.5 6.5 8.0 8.0 7.0 7.0 7.0 7.5
8.5 9.5 9.5 9.0 8.5 8.0 9.0 9.0 8.5 8.0 9.0 8.0
1 3 1 1 0 0 0 1 0 0 0 0
27 36 35 27 26 21 24 31 21 35 29 29
2 2 2 3 0 3 2 1 2 4 3 2
6 7 3 3 4
6 3 3 4 4
No No Yes Yes Yes
The columns ‘Mensis HS’ and ‘Mensis TSD’ indicate at which day of the menstrual cycle the experiments were performed. BDI: Beck Depression Inventory (range 0–20, cut-off 0.05). Results are shown in Supporting Information Figure S2A and S2B. As illustrated in Figure 1, irrespective of the sleep condition, higher N2P2 amplitudes correlated positively with higher pain ratings (HS: regression coefficient r2 = 0.96; TSD: regression coefficient r2 = 0.87). However, similar to our previous sleeprestriction study, there was a marked leftward and upward shift of the regression line after TSD compared with HS, thus confirming the previously observed dissociation of perceptual and electrophysiological correlates of nociception.
3. Results Before starting LEP experiments, the mean level of sleepiness after a night of sleep deprivation was significantly higher (71.9 5.5, range 34.4–96.6) as compared to a night of habitual sleep (VAS 16.4 3.7, range 1–42.6; p < 0.001). At the end of the study, sleep-deprived subjects were still more sleepy as compared to habitual sleep (59.9 7.6; range 6.7–81; p < 0.01). However, in seven of the 12 study participants, the sleepiness level had decreased by up to 80% (range of reduction: 5%–80%). In the remaining subjects, sleepiness scores either remained unchanged or slightly raised (range 0–16%).
3.2 Cognitive performance Sleep deprivation had no impact on the discrimination task (54.4 1.2% vs. 54.1 1.4%; p = 0.9; see Supporting Information Figure S3A). Compared to habitual sleep, study participants performed significantly less calculation steps after a night of TSD (104.3 11 vs. 80.2 8.4; p = 0.05; see Supporting Information Figure S3B). However, the performance index, though being lower after TSD, was not significantly different from the performance index after HS
3.1 Pain ratings Pain ratings (averaged across all attention conditions) were significantly increased after TSD (Intensity Scores: HS: 27.8 3.4; TSD: 38.0 3.9; p < 0.001, Unpleasantness ratings: HS: 33.8 3.5; TSD: 44.6 3.9; p < 0.001). The percentage increase of the intensity score (39.7 9.9%) was highly similar to the percentage increase of the unpleasantTable 2 Laser-evoked potentials – N1 amplitudes. Attention condition
Sleep condition
Focusing
HS TSD HS TSD HS TSD HS TSD
Neutral Distraction All conditions
N1 (lV) SEM 5.7 5.6 5.6 4.8 5.1 5.0 5.5 5.2
0.5 0.6 0.6 0.6 0.5 0.5 0.3 0.3
N1 (log) SEM 0.700 0.690 0.679 0.595 0.659 0.638 0.679 0.641
0.048 0.050 0.054 0.064 0.046 0.049 0.028 0.031
t=
p=
0.13
0.89
1.68
0.11
0.71
0.6
1.47
0.15
Raw data and log data of N1 amplitudes (mean SEM), separated for each attention condition. Results were compared in paired t-test.
© 2015 European Pain Federation - EFICâ
Eur J Pain 19 (2015) 1197--1209
1201
S. Schuh-Hofer et al.
1202 Eur J Pain 19 (2015) 1197--1209
0.07 1.8
0.06 1.9
0.022 0.030 0.049 0.041 1.079 0.677 0.525 0.508 0.6 0.7 0.4 0.5 0.71 0.37
13.1 12.2 3.9 3.0 0.05 for any of the attention conditions; data not shown). Separated for each task condition, the significant effect of TSD on the P2 amplitude decline (assessed at Cz) was restricted to the conditions ‘neutral’ (p < 0.01) and ‘focusing’ (p < 0.05), while there was no significant difference when subjects were distracted (p > 0.05; see Table 4 and Figure 2). As demonstrated by Figure 3 and Supporting Information Table S1, the attentiondependent effect of TSD on the P2 component is similarly reflected by analysis of the global field power (‘neutral: p = 0.02; ‘focusing: p = 0.04; ‘distraction’: p = 0.28; one-tailed paired t-test).
p=
3.4 Effect of TSD on the vertex potential N2P2
Raw data and log data of N2 and P2 amplitudes (mean SEM), averaged over all attention conditions, at the three vertex positions Cz, Pz and Fz. Significant results are marked in bold.
n.a. n.a. 5.8 0.5 6.4 0.6
lV SEM
Fz
log lV SEM
The amplitude of the N1 potential, averaged over all attention conditions, was slightly but not significantly reduced by TSD (HS: 5.5 0.3 lVM; TSD: 5.2 0.3 lV (p = 0.15). TSD had no effect on N1 latencies (p > 0.05 for any of the attention conditions; data not shown). Table 2 illustrates results of N1 amplitudes, separated for each attention condition.
n.a. n.a. 0.657 0.041 0.653 0.057
t=
3.3 Effect of TSD on N1
1.68
p=
(33.8 9.4 vs. 23.2 4.7; p = 0.384; see Supporting Information Figure S3C).
0.09
Effect of sleep deprivation on habituation of LEP’s
© 2015 European Pain Federation - EFICâ
S. Schuh-Hofer et al.
Effect of sleep deprivation on habituation of LEP’s
Table 4 N2 and P2 amplitudes – attention-dependent effects. Attention condition
Sleep condition
Focusing
HS TSD HS TSD HS TSD HS TSD
Neutral Distraction All conditions
p=
0.07 0.13 0.35 0.01
N2 Amp (lV) SEM 7.8 7.6 8.3 7.9 7.1 5.6 7.8 6.9
N2 (log) SEM
1.0 1.4 1.3 1.3 1.1 1.0 0.6 0.7
0.847 0.777 0.842 0.843 0.790 0.635 0.827 0.745
0.061 0.104 0.089 0.07 0.071 0.106 0.042 0.055
t=
p=
P2 (lV) SEM
0.80
0.44
0.01
0.99
1.68
0.12
1.51
0.14
17.9 14.25 18.7 15.24 15.4 14.1 17.62 14.7
1.2 1.2 1.6 1.7 2.1 1.2 1.0 0.8
P2 (log) SEM 1.242 1.134 1.248 1.141 1.142 1.126 1.22 1.145
0.029 0.039 0.045 0.064 0.062 0.045 0.028 0.028
t=
p= 0.004
3.63 0.017 2.8 0.71 0.38
